This content outlines the initial stages of a product manager's lifecycle, focusing on validating a product opportunity and its potential value before design and development. It emphasizes a structured, hypothesis-driven approach to ensure alignment and feasibility.
Mind Map
Click to expand
Click to explore the full interactive mind map • Zoom, pan, and navigate
Hi everyone, good morning
all the new folks as well as the folks
joining from the older cohorts. Uh we
will just wait for three or four more
hi everyone. Good morning. Um, all
right. So, what we're going to talk
about today is uh we're going to talk
about uh focusing on how the life cycle
of a product manager works or what are
the things that a product manager does.
Once there has been a strategic idea
that has come to the place right. So um
as we uh looked at the roles when we
looked at the orientation we also looked
at trying to understand that exactly
when a PM is uh there exactly when there
is an idea that comes to the product
manager their job is then to execute it
especially when you're a solution based
PM and of course from a problem space PM
uh it is important for you to understand
this so that you can lead and drive your
team accordingly. So what we're going to
look at today is the first two
components of um of of going ahead and
solving the problem which is feature
opportunity validation and feature
design. At first we would of course look
at feature opportunity validation or
what we will look at in this case is
what are the different things that you
must do. What are the different steps
that you must do as a product manager?
No matter uh whether the idea of doing
something has come on your own or has
come elsewhere but whatever is the uh
whatever are the steps that you must do
to become someone who is ready to go
ahead and validate an opportunity
whenever it comes to you. Right? So once
again uh the opportunity is going to
come to you via uh maybe someone senior
maybe your uh own ideas maybe your own
observations. We will talk about how uh
you know opportunities can come and how
you need to look at them later. But what
is more important to begin with or what
is more important to start with is to
understand exactly the points that you
would need to know or the steps that you
would need to know to validate whether
an opportunity is worth uh chasing or not.
not.
Now the first thing or the first uh
point that you would look at or the
first point that uh you would look at
okay just of course before that any
questions whenever you have I'll pause
for questions after different sections
just feel free to raise your hand and
I'll uh be able to answer them um and
unmute you and answer them u so yeah so
let's start so the first thing of course
or the first point of course that uh we
would want to look at is opportunity
validation now consider Consider this.
You have got the uh clarity from a
senior leader or you've got the clarity
from someone who is uh uh focusing on um
uh you know building stuff uh and and
someone who's focusing on deciding what
to build. So you've got that clarity.
Now your job or your role is that okay
now that we have this clarity now that
we have this understanding that okay
this is what uh we are supposed to build
or this is what something that we could
build what is it that we should do to go
ahead and actually build it. Now one
important thing uh that you must
understand. One important thing that you
must uh look at here is to focus on
uh one one important thing that you look
at here is to focus on that even if you
are in a place where the opportunity or
the valid or the or the or the task that
you need to do has actually come from
someone senior has actually come from
someone with more experience from you.
It can be something that you could uh
you know focus on and try to see and say
that okay um
you could go ahead and say that okay now
that this senior person has told me to
do this or have they found this
opportunity it may be easy for me to
just go ahead and implement it. But in
product management since you are the
owner of your career since uh the the
the responsibility of what you have done
at the end rests with you. It is very
important for you to be someone who has
actually validated whatever they're
doing or to be someone who has a lot of
backing in terms of the problems or in
terms of the in terms of the solutions
that they are building. So essentially
even if something comes from someone
senior it does not mean that it's your
role next to just go ahead and blindly
implement it. Definitely not. What mean
what it means is that okay you have an
idea that has come and now the first
step before you even go to uh you know
building or thinking about how that
product would look and feel is to
validate the opportunity or is to
validate what exactly the opportunity is
going to be. Right now let's try and
understand how do you validate an
opportunity. So there are three
components of any opportunity uh
whenever an opportunity comes whenever a
feature comes. So when I say opportunity
what I mean in this case is a feature
right. So let's assume that someone from
let's assume that you are a product
manager at Amazon. Uh then someone has
come in and said that hey quickcommerce
is a great opportunity or someone has
come in and said that hey you know
personalized reviews are a great
opportunity. Now they may have their own
reasons to come up with this. They might
have looked at some reviews. They might
have looked at some observation. They
might have looked at some competition.
But now as a PM, even as a solution PM,
it's your job to go ahead and look at
those opportunities, to go ahead and
look at those
Now there are three parts to this
opportunity validation flow. What are
the three parts? Strategic fit, business
value, and user value. So for a product
manager to confidently pursue an
opportunity, so since we're taking the
Amazon example, let's assume that an a
feature idea or an opportunity came to
you which was around personalized reviews,
reviews,
right? So someone came up to you,
someone senior came up to you and asked
you that hey uh could you I mean this is
something that I have figured or you
yourself figured it out from different
discussions or wherever and the
understanding was that personalized
reviews is a great opportunity to double
tap into. Now for you to confidently go
through this, for you to confidently go
through this, you need to actually go
ahead and do three types of validations
before you can actually confidently say
that yes, this is something that I will
chase. What are these three types of
validation? The first is strategic fit.
The second is business value and the
third is user value.
convincing your manager. Now sometimes
these ideas may come from your manager
themselves but even then your goal or
your job is to convince your manager and
your job is to convince whatever your
manager has said or whatever your
manager has uh uh presented by creating
an initial hypothesis. We'll just call
that IH for now. Right? Creating an
initial hypothesis is what the first
part of your uh opportunity validation
uh begins with where your role is to go
ahead and convince uh and and present an
initial hypothesis to your manager or
your leadership that you are that gives
you the benefit or that gives you the
alignment to going ahead and doing
business and user value research. Right?
So let's just let me just break it down
step by step. Step one is for your
manager to get convinced, right? Even
though the idea has come from your
manager, it is still important for you
to do some basic work and convince them.
You can convince them by creating an
initial hypothesis,
right? This initial hypothesis is what?
This initial hypothesis is more around
the business value, the user value and
in general the strategic value that the
feature is going to give uh for the
product. Then once the initial
hypothesis is approved at this stage
which means that the manager has is
aligned to go ahead and uh uh you know
invest uh your time and your money and
your uh bandwidth and everything into
this stage. It is now your job to go
ahead and do a deeper validation of the
business value and of the user value.
We'll look at it step by step and we'll
try and understand to understand this
better and to understand these products
or the way better we are going to use a
tool as well. Uh so this is called
sculp. Uh this is uh proprietarily built
by us uh I mean by by me and team uh
itself. uh which is basically a SAS that
helps you uh manage your work or that
helps you follow the steps that we are
going to learn as we learn uh as we go
along. Um once you sign up on this uh
and if you uh let me know you would get
uh pro access to uh the tool so that uh
for the next one year you can use it uh
and practice all the steps that we have
mentioned but essentially we are going
to apply the work that a PM is supposed
to do on this tool step by step
especially for a solution based PM. Now
let's assume that we've got an idea. Uh
we will call this idea as um
Then uh we are going to also go ahead
and give it a description giving ratings
and reviews based on a user's previous purchase.
just exchange to personal
views for Amazon and now we can select
the phase. So we're going to put it in
validation because as of now this has
just come as an idea. This has just come
as an idea. Uh maybe it has come from
you know looking at reviews. Maybe
someone has come up with this uh u uh in
genuinely itself. It could have come
from multiple places. Uh but it has come
as an idea and you can put a focus start
date. So now this comes here and it's
going to take me to the project's
homepage where we are going to follow
the steps that you need to do to ensure
that you are aligned or that whatever
you are trying to do is uh aligned with
whatever the person in front of you or
your manager in most cases is going to
talk about. Right? So how do we do that?
The first step whenever you are
validating whether a problem statement
uh is uh is even valid is like I said
conducting a manager briefing. Now once
again repeating the idea may have come
from your manager but all that means is
that it means you are going to get uh a
faster alignment on that but even with
that even if the idea is coming from the
manager briefing you must give your
manager a understanding of what you're
going to do and only then pro proceed further
further
right. Uh
now what is that step? The first step is
conduct manager briefing. Of course,
Sculp allows you to, you know, mark
these steps and all of those things. And
it also allows you to use templates that
effectively explain how you're going to
do this. So, we're going to use the
template uh to conduct manager briefing.
And now this is what we're going to use
right in this uh we're just going to use
our feature. And uh that's what we're
going to put here as well.
Right now when you're going to your
manager to get uh uh uh to get alignment
for something like this, the first thing
that you would talk about is the
strategic fit. Now this is the easy part
because this talks about the fact that
uh you know uh do you have the right
mission and vision? Do you have the
right company strategy? Do you have the
right product strategy? And is it
aligned with the goals that your team is
facing? Right? Uh once again mostly your
manager would be aware of these things.
They would be aware of you know what the
mission and vision of the company is,
what the strategy is, what the product
strategy is and what your team goal
goals are. But it still is great to kind
of put it into place. uh there is a
example as well where uh let's assume
that you're working on Uber and you're
working on reducing cancellations. So in
this case you would of course talk about
the mission and vision of Uber in terms
of you know improving people's life with
the best world's transportation. You
will then talk about the company
strategy where you know the strategy of
the company is to provide drivers
flexible earning opportunities and
offering rides at more and more
affordable rates. Product strategy is
empowering drivers with flexible earning
opportunities to work for them. And then
team goals is the driver experience team
that you are a part of and essentially
create an alignment on these four.
Right? So if if I am supposed to as a
product manager take up something around
a manager briefing around personalizing
reviews for Amazon, my job is to first
write down these four for my product and
my team rather and then take alignment
from the manager on this. Right? Very
straightforward thing because you know
you would usually know very well about
your team. You would usually know very
well about your strategy, about the
company and the mission and vision. But
in every time when you fill this, you
would have to fill it with alignment
on why personalizing reviews for Amazon
aligns with the mission and vision of
Amazon. Why it aligns with your team,
why it aligns with the product strategy
and why it aligns with the company
strategy. Right? Usually this would be
uh very straightforward. You would just
write you know standard stuff here in
terms of how alignment is. This is very
Right. So it essentially means that you
would have a lot of uh uh you know
clarity in terms of uh showcasing this
properly in terms of being uh uh you
know as verbose here as possible and get
alignment on it as quickly as possible.
Right now once you're done with this uh
this is the easier part. Now is when you
start creating
your first set of initial hypothesis,
right? So once again if I if I go back
to sculp and uh you know kind of show
you the steps. What are the steps? The
first step is conducting manager
briefing. In manager briefing we talk
about the strategic fit.
Then we talk about the user value
hypothesis. So once again very uh clear
demarcation what we are talking about
here is our viewpoint of what the user
value is and then the business value
hypothesis which is our viewpoint of
what the business's value is right so at
this step what we are doing is we are
creating hypothesis around okay if we
are personalizing the review for uh
Amazon this is the value that we expect
it can get Right? Most of the numbers
here are assumed.
assumed.
and there is very less interaction with
actual users at this stage because at
this stage you're doing user value
hypothesis. Now of course a few people
confuse this with saying that okay hey
shouldn't if I'm going to my manager
shouldn't I go ahead with a lot of data
with a lot of uh user research all of
that done that is something that you do
in the coming steps even before you
spend time on going ahead and doing this
you actually go ahead and get an
alignment from your manager saying that
hey this is my hypothesis this is what
I'm thinking of this is why I think this
can work And now I want uh your
alignment to go ahead and do my user
value research which is do my interviews
and and create my business funnels and
so on and so forth. Right? So
essentially in the manager briefing step
what you end up doing is you end up
first of all writing the user value. So
you just end up writing uh exactly what
uh your hypothesis has been exactly how
uh you know you think that this is going
to work out or this is going to add
value and then once that is done you
then go ahead and talk about the user
value hypothesis but once again very
very uh strongly reminding you that this
is initial hypothesis
right this is not a hypothesis that is
for later this is initial hypothesis
which means that this hypothesis is the
starting point of your journey of making
the feature. It is not expected of you
to be very very uh deep and analytical
at this stage. It is not expected of you
to have a lot of user interviews and all
of those at this stage. It is expected
but it is expected in the coming stages
of this entire situation.
Right? So once again, don't confuse this
with um um don't confuse this with uh uh
with with a lack of depth. The thing
here is that you want to get an
alignment as soon as possible so that
you can use that alignment and move
forward in your journey.
Right? Now once that is sorted or once
that is done, you essentially go ahead
and start typing or typing in your user
hypothesis. So let's take an example and
see what a user hypothesis can look
like. The first part is the user profile
hypothesis. Now what is a user profile
hypothesis? User profile hypothesis is
nothing but uh just a straightforward
understanding of your users's uh profile
where you can split it into two. One is
the demographics and formographics.
Formographics is more of the kind of
companies they work in. uh if if it's a
business then what is the kind of
business what is the kind of spend
turnover etc demographics is if you're
doing it for the consumer and then talk
about their product behavior and usage
profile right so now you don't need to
have a research done to get this answer
if you feel challenged uh in the
situation that you are having to do a
lot of research to even get to this
answer or get to this situation it means
that there is some gap somewhere it is
something that you would have to work on
to ensure that you are able to fix it
because you shouldn't be having any gaps
at this stage uh in terms of creating
this hypothesis. If you have that uh gap
where you are unable to think that okay
which is the kind of user that is going
to profit from this then you would have
to observe more
because that's one of the biggest uh you
know uh confusions or one of the biggest
mistakes that product managers do where
they feel that a lot of the answers of
whatever they're doing a lot of the
answers of whatever they are building
actually need to come from end users
where that is not the truth. The truth
is that you have to have your own
hypothesis and then your own hypothesis
is going to then convert this uh into this.
this.
Right? So that is how the entire
structure of this is done and that is
how the entire structure of it is built
up. Right? So you begin with the user
profile hypothesis talk about that talk
about so in the case of uh you know
Amazon we are going to talk about you
know what is our initial hypothesis for
someone who is going to use personalized
reviews. So um if I were to think uh
about this and go uh deeper of course
Amazon is in uh uh in this thing but one
thing that I can very very clearly uh
you know kind of uh think of or kind of
put is um folks who are interested in uh
getting the maximum value out of
whatever they're building or whatever
they are buying sorry.
So this is something u uh uh
this is something that I can definitely
think of as as one of the uh demographic
attributes. Of course I can also look at
maybe focus on a certain section of the
audience, certain tier of the audience
and so on and so forth. Uh and then go
ahead and uh you know talk about this.
Similarly their product behavior. Now I
have to have an initial hypothesis
around this. So one of the hypothesis
that I can have is uh folks who check in
at least 5 to 10 reviews before they
place an order usually wait for
discounts so that the same thing is
available later and also
uh are assured by a return policy on Amazon.
Amazon.
Right? So now this is another behavior
that I have written down where you know
my product behavior is that you know
people check at least 5 to 10 reviews
before buying they place an order uh
after that and also you know uh they
also keep checking if there are
discounts available on the price that
they have been quoted so that they can
get the maximum value out of it. Now you
can have a different hypothesis you can
have a completely different segment that
you want to solve for. For example, you
may want to solve for, you know, just
focused on maybe electronics users and
and and stuff like that. Uh you may want
to solve for um uh some some other type
of user and and solve for that. So
that's something that you would do and
go ahead and present it here. Right?
Again, very varied, but you should have
your own viewpoint here. And if you're
not able to create that viewpoint, you
should observe users more or look at
other people's research more to get that
viewpoint. Right now the next part is
the user's problem hypothesis. So here
you break it down into two parts. What
is the user's problem and what is and
why is the problem so important. So what
is a user's problem? Of course you would
talk about problem area and signals
which is where uh
what is the problem area of the user and
what are the signals they get and what
is the uh why is the problem important
you know the severity of it uh and um
and then the number of users impacted
right so now let's talk about this for
Uber for example the problem area is
that at the end of the session the
drivers want their last ride to bring
them closer to their desired destination
however for those who live outside busy
areas it's it isn't easy to get paired
with uh people who are going to those
areas, right? Um what are some of the
signals, the reports and complaints
about u uh where the customers have uh
received significantly less uh I mean uh
okay this is something else written by
mistake. uh the report the signals would
be more about focusing on you know the
complaints that have come or the drivers
that have complained that you know they
go or also another thing that would
happen is people would go ahead and like
mentioned here that at the end of their
driving session they would forced to
close the app and just go without a ride
and so on and so forth and what is the
number of users impacted once again no
data yet it's just your hypothesis so uh
you are and you have to of course
communicate it clearly to your leader or
your manager that I'm not looking at uh
numbers as of now. I'm just presenting
my viewpoint of it and once that my
viewpoint gets better, once my viewpoint
gets aligned, I'm going to then go ahead
and do a deeper research on it. Right?
For the Amazon example, I mean the
problem areas would be
users buying a product that had high
ratings but uh found them to be not
usable for them because they valued certain
certain
significant smaller proportions of that
product. So this could be that you know
I as a user want to value um
uh battery life uh for my phone but uh
the reviews were great but they were
great because they were more about um uh
you know on the phone the camera quality
and camera quality is not valid for me
right um uh so I have a very specific
value and certain significantly smaller
portion proportions of the product is
what I target on what would be the
signals high refund in cert certain
categories where people buy for multiple
core value propositions. So this would
be a signal that I would uh talk about
where you know my my thought process is
that okay there are certain uh core
uh products where there are more than
one core value proposition. So phones is
of course a great example uh for this.
Clothes is also a great example where
some people would you know want to uh uh
would want to validate and would want to
have comfort and some people would want
to have style difference and so on and
so forth. So while something is rated
high on one thing, it could be rated low
on the other. Uh some people have a
bigger problem with power consumption of
an appliance while the others might be
focusing only on getting uh the job done
because they might not necessarily use
it. So what is the high refund that you
get or what is the high cancellation
that you get from here right now?
Severity very severe because uh if one
of the core values is not aligned the
product makes becomes uh really less
usable for the user.
So again this is my hypothesis. This is
not uh something that I am uh you know
using data to do. I'm just of course
data is going to come in but I'm just
putting down my hypothesis. This is
where I may have to do some guesstimate.
We'll talk about guesstimates as well in
one of the interview sessions. But this
is where I may have to do some
guesstimates in terms of going ahead and
saying that hey
um maybe 20 or 10 to 15% of the returns
may be impacted by this. Again this is
more of a guesstimate here. So I can be
a little I can give a range or or things
like that. And then what is the
alternatives here right? The alternative
here would be for people to check the
product online on Amazon and then go
ahead and physically find it in a store,
try it for some time, take multiple
opinions and then decide whether they
want to buy it based on the best price
available. Right? That's that's a
behavior that uh you know you would have
seen in multiple people and so on and so
forth. The next part of the user is the
goal. Once again, you go ahead and talk
about what the goal of the user is. So,
the goal here for for the Uber example
would for the driver would be to, you
know, uh get paired with rides uh that
match their final destination,
especially when they're closing the day.
Quantitatively, it's going to impact
cancellation rate. It's going to also
increase driver earnings and also driver
supply would increase at odd hours
specifically. And non- goals are
basically goals that you don't want to
support. The non- goals are usually uh I
mean features or or directions that you
don't want to take. For example, you
don't want to create route preferences
beyond the end of the day. You don't
want a drivers to be able to say that
no, I just want to go on this road all
the time and that is why I'm going to
create this. Right? Similarly, for the
Amazon example, the user goal would be
to get the best out of every purchase.
Tell Amazon what matters for me while
buying something. and then uh get the
best recommendations based on that. So
that is for the user. Quantitatively,
this would mean lesser refunds and
cancellations on these products and the
users would come back and keep using
Amazon more. Once again, I'm not I'm not
taking any data. I'm not looking at some
numbers here, but I'm giving a
quantitative assessment of what it would
look like, what metrics would impact.
And non- goals would be I don't want to
reduce the price of the products for uh
for users when they get
personalizations. I only want to select
the best for them which may mean a more
expensive choice as well. So this is
basically a non- goal that I'm taking
where I'm categorically saying that hey
price is something that I'm not going to
look at here. Uh price is something that
I don't want to optimize for at this
stage. which I want to optimize for the
best match with the user and then based
on that I want to present it to the user.
user.
Right? So essentially this is how you
create your initial hypothesis. Right?
Once again, uh, of course, we we did a
very quick exercise here, but it should
take you a maximum of two to three hours
to create this initial hypothesis. And
this is enough for you to go ahead and
talk to your uh manager or talk to
whoever uh is available to go ahead and
talk about uh this as a problem
statement to uh to basically capture,
right? uh we'll talk about the next step
which is what is the business value
hypothesis but we'll take a few
yeah I'm sure there are plenty of
questions uh okay is saying can we get
this template yes of course so you will
get this templates access um right after
this session uh both in the assignment
that you would get for this session and
also on sculp that I showed so you would
get access to that. Uh okay, Gome is
saying where would discovery fit? Uh so
I mean u so let me help you kind of
break this down because this is a
slightly confusing notion that comes
especially when you start looking at
this. Now problem discovery while most
places or everywhere that you see is
taught as the first thing that a PM is
supposed to do there is a small catch there.
there.
Practically what happens is that you
don't get a chance to do problem
discovery unless you become a problem
space PM. Right? what we discussed last
time when we discussed that okay what is
a problem space PM and what is a
solution space PM now only when you're
in a problem space role will you do
problem discovery what happens in a
solutionbased role which we are talking
about now and in the next few sessions
we'll talk about that itself is you get
a problem that has been discovered
already and you have to now take it
further to the next direction take it
further to the next step right Now this
is very different from what is usually
uh you know uh the usual convention that
okay the first thing would be problem
discovery because uh in most situations
what would happen is that you would
already have a problem with you and then
you have to validate right so to answer
your question this happens after uh the
validation part happens after discovery
but essentially as a solution space PM
you are not responsible of doing this
discovery on your own. It's more about
someone else who is usually in a problem
space role who would do this and then
come back to you to say that okay this
is how the uh problem is and then you go
ahead and build it out right um so what
other tools can be used so we'll dive
deep into these uh for now the starting
point of this is let's say no other
tools the starting point of this and
that's something that you'll have to get
comfortable with is only you and your
product chops or your product knowledge,
right? That that's one thing that you
know most um new PMs or most PMs
sometimes kind of falter at the first
thing that you have to hone and you have
to get better at is your hypothesization
of things. Right? A lot of times what
happens is you always look for answers.
You look for tools. You look for
external reasons. You look for external
pointers and you're not enough and
you're not able to focus enough on what
you as a product manager feel or you as
a product manager think. So the first
thing we'll get to tools as well but the
first thing that you have to be
comfortable with is your hypothesis. I
mean the excel or sculpt that I showed
is something that can be uh you know
used probably but otherwise
uh you know the idea that you have to be
comfortable with how this works in your
hypothesization should be the first thing.
thing.
Okay Gum said feeling kind of fast. Okay
I'll try and slow down but feel free to
ask any question that you missed. Uh
okay so Abdul is saying example is a
little confusing so I'll just try and
paint another example of uh the entire
situation here so let's assume that you
are working at uh Amazon right and uh
what happened at Amazon is u there would
be a senior PM or there would be a lead
PM or there would be a group PM or there
would be a director of PM
they will come with a problem that they
have discovered on their own based on
their experience. Right? Right. So in
product what happens is when you are in
slightly earlier roles especially when
you are in solution space roles like we
discussed in the first uh uh this thing
when you're in solution space role what
happens is work comes externally right
so someone who is at a senior level is
going to go ahead and give you work is
going to go ahead and tell you that hey
this is something that you have to um uh
do and You then take that idea and start
working on it. Right? To give you
another example, let's take an example
of Uber, right? So what would happen is
someone who is senior, someone who is um
you know into a group PM or whatever
kind of role will actually come to you
and tell you that hey you know I feel
that uh we should integrate a AI based
ride search. Right? This has come either
from my mind either from reviews could
could be anywhere but this is a
hypothesis that will come to you. Now it
sounds very counterintuitive because it
doesn't sound very structured that okay
someone will randomly give me a
situation or a pointer to solve but
that's exactly how it works practically
right I can tell you that no in reality
you should go ahead and do discovery on
your own but truly if you're into a
solution space role you will not be able
to do discovery on your own right so
someone will give you this problem
statement now as a PM or as a solution
space PM your job will be to take that
step back and say that okay now that I
am you know doing this uh now that I
have been given this problem statement
or given this thing to build I will go
ahead and create my hypothesis of why
this should work or this should not work
right so the essence here is that you
don't have to go ahead and you know jump
into whatever has been given to you by a
senior you have to first create your own
hypothesis which means that you have to
first create your own in-depth
hypothesis of things and then once
you've done that you take it to your
manager and say that hey this is the
problem statement that I have this is
the problem statement that was given to
me or that you gave to me or someone
senior gave to me and this is how I want
to approach this so that's you know
basically what I'm trying to communicate
and then of course to do that you
conduct the manager briefing based on
the template we just discussed
okay we'll see we'll take some questions
questions from people who are speaking
as well. Yes, Garvid, go ahead.
>> Uh yeah, hi M. So, uh basically I'm just
a bit confused on the Uber uh example
that you mentioned because earlier we
were discussing about the cancellations,
right? Uh so am I getting confused on
the idea because this one seems to be on
the last ride of the rider.
>> Okay, so there are two different
examples that we took. One was a
existing field example that we took
where we took the example of saying uh
you know how can you reduce
cancellations on the driver side
>> and then we took another example of how
can uh you know we personalize reviews
for Amazon. So these are two different
ideas that came in and the idea was to
try and fill manager briefing documents
for both of them.
>> No no actually uh I'll try to phrase it
better. So
>> basically on Uber we were discussing
about the cancellations right? So how
are cancellations exactly related to the
last ride of the rider? I'm confused on
just the Uber idea.
>> Okay. Okay. Got it. Got it. So the
hypothesis there was that okay because
in the last ride uh
>> it's an observation right you when you
speak to people or when you try to book
rides at late night a lot of the times
the taxi drivers come back with a
hypothesis saying that I will not go
because I'll not get a return fair from there.
there.
>> Yeah. Or what they would communicate is
the fact that I
>> place. Yeah,
>> correct. Or what they would communicate
is that no no I won't go because I have
to go home and my go my home is on the
opposite side.
>> So it's coming from just that
observation that hypothesis right.
>> We would validate it with data, we would
validate it with numbers, we would
validate with everything. But the first
point of it actually comes from someone
observing it in different places. So
someone looking at this and this could
be a normal I mean a solutions based PM
as well but usually it would be someone
in all of these situations who would
take that step and say that okay I spoke
to let's say I I spend some time with
you know 5000 drivers and all of that
and a lot of them end up mentioning the
same thing which is that at night I
actually uh you know don't get more
rides so that is why I usually don't
take rides at night because I have to
travel a lot back home so that's where
the hypothe hypothesis is coming from.
Right? Similarly, the Amazon also could
come from personal observation, could
come from reading some reviews, could
come from figuring out okay why do
people get disappointed on high-rated
products, right? And then some things
that could come out is okay, they could
get disappointed because they had
thought of the rating as an overall
thing, but then what matters for them is
something different. I think Zomato
recently did something very similar
where what they did was they said that
okay ratings don't matter anymore as a
blended average. What essentially
matters is what is important to you. So
for example if I order uh you know a lot
of um let's say uh Indian food Indian
cuisine and so on and so forth or for me
maybe the packing or the hygiene matters
a lot more or for me maybe the quantity
matters a lot more or the all of those
things. So for me the ratings are
different from someone else. So this is
just an observation that comes in.
Usually it comes in from higher up
right. You can of course as a PM also
take the owners owners of doing
discovery and coming up with this but
typically in 90 95% of the scenarios
this observation or this problem
statement will come from the higher up.
And as you grow uh you know upwards in
the ladder as you become better and
better in the ladder you become better
and better in terms of uh um you know
doing the problem discovery and you get
more and more opportunity from that perspective.
perspective.
>> Right. Right.
>> That makes sense.
>> Got it. So
>> yes. Yes. So just a quick uh question on
that part. So
>> like the feature opportunity validation
then would you say is mostly the problem
PM's job? I mean the uh the PMs in the
problem space.
>> No, no, no. So what happens is the PMs
in the problem space would give you a
problem statement. So if I am let's say
a PM at Amazon, what would happen is a
senior PM or a group PM or someone will
have given me this problem that hey
Gervid do you I think you know what I've
been noticing is that um you know a lot
of people see 4.8 reviews, 4.9 reviews
but they're still very disappointed
because to them something else matters.
Now it's your job to take that entire
thing and convert it into something
completely different on your own. Right?
So there you have to validate it because
it has come as an idea right out of
their mind or right out of their
observation. So it's not necessary that
that idea is 100% correct. So you have
to take a step back and say that okay I
will now conduct my own validation
because I have this hypothesis that this
may be the reason right and then I will
validate it against three things. I
validate it if it's strategically
aligned. Mostly if it is coming from
seniors or it is coming from people in
more problem space, it will already be
strategically aligned. But still I don't
want to you know dive into something
that not everyone is aligned on. So
that's the first thing that I would do.
>> The second thing that I would do is then
I would validate if my users are aligned
with it. And the third thing is if my
business is aligned. Now even to do this
exercise the first thing I would have to
do is I would have to create my own
hypothesis right I mean so what I'm
trying to say is one mistake that a lot
of PMs and especially early PMs do is
that they get a problem statement and
they directly go to the user with that
problem statement or they go to the
business with that problem statement and
try to pitch that problem statement to
them and see if it is working for them
and so on and so forth. Whereas the
better way or the right way to do it is
to get the problem statement, create
your own hypothesis on that problem
statement which is basically what do you
think as a BM is it right? Is it worth
chasing? What is your hypothesis? And
then taking that hypothesis to you know
whoever uh is the user or whoever is the
business which we'll see in the next steps.
steps.
>> Got it. Got it. Clear. Thanks for
>> Okay, we'll move for now. Yeah, we can
go ahead.
Yeah, sorry. Um, so I think a part part
of that got answered in the previous
question, but uh what I wanted to check
is do you also have to do this for that
for like really obvious sort of problems
that um is like standard to be done as a solution,
solution,
>> right? So as a rule of thumb in a very
practical meth manner anything that uh
you know is big enough to involve like 2
to 3 weeks of full effort. That effort
could be you know you working with a
designer, you working with a developer.
Anything that involves like that amount
of time you should probably do a
validation before you invest more energy
into it or you should at least do a
manager briefing before you invest more
energy into it. But yeah, if this is
like very straightforward, you know,
like for example, you're implementing a
profile section on the product or you're
implementing um you know, a Google login
on the product. Now, those are things
that you don't require a lot of
validation for. Those are just things
that you have to do. So, that's
something that you could just very
straightforwardly do.
>> Yeah, makes sense. >> Yeah.
>> Yeah.
Okay. Yes. Uh I can only see M, but
please go ahead.
Um Mallay my question was um basically
let's say I hypothesized something and
uh that doesn't sit well with the
manager uh with whom I'm giving the
briefing on. >> Mhm.
>> Mhm.
>> So uh if that's the case does the like
the loop goes on like I I go on and
hypotheize again or do does the manager
uh help me hypothetize it again?
>> Correct. So it's an iterative process
and now it depends on the dynamics that
you have with your manager. In most
cases that I have seen one or two loops
is what it what it needs because then
the manager will you know be open to
kind of take a step back and say that uh
um okay let me help you out with this
let me work with you on this and so on
and so forth. So that's something that
the manager would usually do um and they
would be pretty okay with that. But this
entire validation process is essentially
a loop. It's it's a it's an iterative
process where you would go with the
hypothesis. The manager would say no no
no this is not true. I've seen something
else happen in reality. Uh and and this
is probably a thought process that you
have missed or something like that. And
then you would iterate a little bit on
top of it. But if you have a good rapper
with your manager, they usually it will
take like one or two cycles and then it
would be and they both of you will be
then aligned on what you need to do next.
next. >> Okay.
>> Okay.
Yeah. Uh yes, go ahead.
>> Hi M. So I have a question like in
typically B2P scenarios like you are
developing a application or a system
which solves a problem uh like uh you
internal users or for an enterprise.
let's say for a bank uh and then in that
case like how do you come up with use
cases uh feature validation uh in that
sense if
>> right so so the understood so just to
reiterate what you're saying is that in
fintech and banking and these kind of
regulated sectors how do you come up
with validation is that correct
>> yes uh so yeah so because you talk to
different users and uh so they might
they might be having some like priority
with and you have your own hypothesis
with your own observations. >> Mhm.
>> Mhm.
>> But uh the product road map or the
product execution will have a different
uh set of features which they need to
execute. So how do you uh position your
like how do you convince and go with
this process in in that case like
>> okay so those are two different things.
First I'll talk about you know how do
you uh do the hypo hypothesis validation
process. So that's pretty much similar.
We'll talk about you know how do you do
user validation next uh in terms of
surveys in terms of speaking to users
interviews and all that and then we'll
go deeper into it in next sessions as
well in terms of how do you conduct
research but essentially the answer to
you know validating this be it
enterprise be it B2B be it B2C is going
with your hypothesis not just one going
with three or four hypothesis and
expecting your users to give you a
validation or an invalidation of that
one by one. So that's part one. Part two
of this is more around the fact that uh
once you have this clarity of uh you
know what is going on in terms of uh um
what is the road map that the product
team already has. uh because again like
I said most of these problem statements
that will come to you especially as a
solutionbased BM will come to you from
people who are deciding the road map
right unless you're in that leadership
position where you are creating the road
map of the product this will come to you
from people who are creating the road
map. So most of the times the problem
statements that will come to you will
actually be aligned unless and until you
are trying to find your own problem
statements which is where of course you
would have to you know create more
narratives more measurements more
success metrics and all of those kind of
things to convince people. We'll talk
about that as well but most of the times
what will happen is it will come to you
already filtered. So it will come to you
already in a way where these things will
be you know already presented to you in
a way where you already know exactly how
uh uh this road map is and you know
>> Yeah thanks M. So uh my question you
partly answered uh so I was just looking
around the answers of like whether we
are going to cover the problem space
related uh discovery and other phases
also uh at some point.
>> Yes of course of course. So how it is
structured is we'll first cover what a
solution space PM does because that's
you know more executionoriented more uh
you know rounded from that perspective
and then we'll jump to what a problem
space PM does. So things like uh
>> things like
>> oh I can hear myself again but anyway so
things like strategy things like uh road
mapping vision things like uh finding
PMF things like scaling product delivery
creating teams uh distributing work all
of those things is what we'll also study
but first we'll fix figure out solution
space so that the basics are set like
what we're doing now plus data
experimentation AB testing those kind of
concepts uh and then we'll move to um uh
problem space.
Okay, I'll take a few from chat now. Um
so there are a couple of questions on
data and in terms of hypothesis
creation. So this is u actually a very
uh valid question and this is both from
Rashmi and Anamika which is so we are
not looking at data right we are
essentially in a position where you know
my ask is to not even look at anything
and come up with hypothesis and it might
make you feel a little uneasy or a
little overwhelmed that okay how do I
make a hypothesis when we don't have
numbers or when we don't have users or
all of those kind of things at the And
that is what makes you a great product
manager. I mean there are multiple
multiple things that you can you know
look at or there are multiple things
that you can feel about uh uh these
things but at the end the core product
sense that you have right is essentially
a result of the number of hours that
you've spent with your users right so a
good barometer of judging how well or
how nicely you are positioned as a
product manager in a particular company
or in a particular domain or wherever
you are is how How many hypothesis can
you come up with in general based on the
time that you have spent with users? So
if you're let's say into SAS or into
logistics what because that's what your
uh essential core job is right a lot of
times people confuse
meeting users looking at data and all of
those things as the core job of a PM. Um
but just taking a step back the core job
of a PM is to start with what they feel
right and that is why um even without
the help of data even without the help
of users telling you exactly what to do
as a product manager you should have
your own say or your own version of why
you feel certain things are there. Let's
assume that you are you know at working
at a company at an early stage startup
and you think and the retention of the
product is bad or uh there is something
that's not working out properly or there
is something that's not what is your
version of it. Why do you think that is
happening? Only and only when you have
that answer are you essentially equipped
to go further into that direction. Um
sometimes you may be in a situation
where you don't have the answers
immediately and that's of course when
you're starting but then that's the cue
for you to take it up and go up a notch
and go further a notch where your goal
should be to say that okay I don't know
u uh you know this right now but I will
spend more and more time with my users
and when I spend more and more time with
my users I will get this knowledge as I
go along and I will be able to
hypothesize so That is where uh the
point is and on Rashmi's question of you
know what if we are very far away from
the current numbers when do you pivot
and how would you impact so we'll just
talk about that we'll talk about you
know how your hypothesis is not the end
all of it uh you would have to do a lot
more refinement on both user and
business value where you look at actual
data and so that you're not very far
off. So what we studied so far was just
the manager briefing though that was
step one. Now in the step two and three
what we will do is we will validate
these numbers and we will validate our
understanding with actual users and
actual data. So by the end of it we will
be in a position where we've looked at
actual numbers and we are not you know
just working on assumptions. But
starting with your hypothesis or
starting with your assumptions is
important because if you don't start
there you are actually essentially
saying that okay I don't have a
viewpoint on my my own product. I will
just rely on my users to give me that
viewpoint and that's not uh you know how
Okay. I think Abhinav's question is
something that we uh answered. So it's
pretty good. Uh Swami also has a very
similar question which is does the same
approach of hypothesis? Yes, absolutely.
So the same approach follows for a B2B
product as well. The only thing is in
B2B what happens is and my experience of
working in B2B also is that the problems
that come um validating it them is very
very easy because you always get
problems that are uh you know that are
very important for users right so users
already know that this is something that
is very important for them users already
know that this is something that is very
serious for them because when you're
doing B2B it's mostly around their
business it's mostly around their
productivity, their profits, their
income. So they would always be very
serious about it. So the validation part
becomes very easy because they give you
very clear-cut validation. But the flow
of you coming up with the hypothesis of
saying that okay, I have a CRM. I feel
my hypothesis is that people waste a lot
of time on uh you know mis calls in the
product and I should solve it. If that
is the hypothesis that you can you go
with you would you find it easier to
find validation because people take you
seriously but the rest of the process is
absolutely the same.
So men said actually a very interesting
point which is in that case you need to
be an expert in the domain for your own
hypothesis validation
to a great extent you are right and that
is why uh you know great PMS or as you
grow into the ladder into a domain domain expertise becomes very important
domain expertise becomes very important right so there are three kind of pillars
right so there are three kind of pillars to PMing we'll talk about these pillars
to PMing we'll talk about these pillars later as well but just to give you a
later as well but just to give you a brief the three pillars are your
brief the three pillars are your customer empathy which is how much are
customer empathy which is how much are you able to think like a customer, your
you able to think like a customer, your creativity where how much you are able
creativity where how much you are able to figure out new different solutions
to figure out new different solutions and then finally your domain expertise.
and then finally your domain expertise. So you're absolutely right coming up
So you're absolutely right coming up with a better hypothesis is directly
with a better hypothesis is directly proportional to being a better expert in
proportional to being a better expert in the domain. But you can hack this or you
the domain. But you can hack this or you can become a better domain expert by
can become a better domain expert by spending more time with your users.
spending more time with your users. Right? So it's not necessarily that a
Right? So it's not necessarily that a domain expert is only because of the
domain expert is only because of the experience or the number of years you've
experience or the number of years you've spent in but domain expertise can come
spent in but domain expertise can come if you spend more and more number of
if you spend more and more number of hours with your users as well.
hours with your users as well. Okay. So uh the validation phase is
Okay. So uh the validation phase is where in the product life cycle. Yes,
where in the product life cycle. Yes, it's second it's basically the first
it's second it's basically the first thing for a solution space PM but in an
thing for a solution space PM but in an overall life cycle someone will discover
overall life cycle someone will discover the problem that will be a problem space
the problem that will be a problem space PM. Then the problem space PM will pass
PM. Then the problem space PM will pass it on to you or whoever is the solution
it on to you or whoever is the solution space PM and then you will start with
space PM and then you will start with your first step which is opportunity
your first step which is opportunity validation. And the first step of
validation. And the first step of opportunity validation is basically
opportunity validation is basically speaking to your manager and asking your
speaking to your manager and asking your manager and telling them that hey this
manager and telling them that hey this is what I'm trying to do. Should I go
is what I'm trying to do. Should I go ahead and do this or not?
ahead and do this or not? Okay. Nupur has a couple of questions.
Okay. Nupur has a couple of questions. By hypothesis do you mean our own
By hypothesis do you mean our own predictions we do for a particular
predictions we do for a particular feature like how it feels like? Uh yes.
feature like how it feels like? Uh yes. So by hypothesis what I mean is not
So by hypothesis what I mean is not predictions but your own uh I mean your
predictions but your own uh I mean your own explanation of a certain situation.
own explanation of a certain situation. So if in the example we took we are
So if in the example we took we are seeing cancellations on Uber rides. So
seeing cancellations on Uber rides. So as a PM at Uber what do you think these
as a PM at Uber what do you think these cancellations are for? Why do you think
cancellations are for? Why do you think these cancellations are happening? What
these cancellations are happening? What is the core reason for the
is the core reason for the cancellations? If you don't have your
cancellations? If you don't have your own version of this then there is an
own version of this then there is an acute need for you to spend more time
acute need for you to spend more time with the drivers spend more time with
with the drivers spend more time with the users to understand why they cancel
the users to understand why they cancel stuff right so that is I mean I wouldn't
stuff right so that is I mean I wouldn't call it prediction I'll call it
call it prediction I'll call it explanation this is basically your
explanation this is basically your explanation of why a particular feature
explanation of why a particular feature is being used or not used in the way
is being used or not used in the way it's supposed to be done
it's supposed to be done uh and the third second thing that
uh and the third second thing that nobody is asking is can this be valid
nobody is asking is can this be valid for a new product absolutely it can be
for a new product absolutely it can be valid for a new product as well in for a
valid for a new product as well in for a new product it starts with a hypothesis
new product it starts with a hypothesis only right if you if you look at any I
only right if you if you look at any I mean even if you go at a founder level
mean even if you go at a founder level and and so on all the founders when they
and and so on all the founders when they start a company or they start a product
start a company or they start a product they actually start with a hypothesis
they actually start with a hypothesis that they pitch to investors where they
that they pitch to investors where they say hey this is this is what you know
say hey this is this is what you know our thought is this is how we've
our thought is this is how we've validated this thought speaking to so
validated this thought speaking to so many users and looking at this business
many users and looking at this business and so on and so forth and do we want
and so on and so forth and do we want can you fund us or something like that
can you fund us or something like that So it's absolutely valid for new
So it's absolutely valid for new products as well. But for new products
products as well. But for new products again uh what Mahindra had said earlier
again uh what Mahindra had said earlier as well your expertise in the domain
as well your expertise in the domain becomes important and your uh you know
becomes important and your uh you know your time spent with the users become
your time spent with the users become important because the more time you
important because the more time you spend with the users the better you are
spend with the users the better you are going to be at creating these
going to be at creating these hypothesis.
hypothesis. Okay. Okay, Rashmi is asking as you said
Okay. Okay, Rashmi is asking as you said u hypothesis validation is iterative say
u hypothesis validation is iterative say suppose it's not getting settled with
suppose it's not getting settled with the manager in one two and deadlines are
the manager in one two and deadlines are running out would you propose to go for
running out would you propose to go for the right rough hypothesis and get the
the right rough hypothesis and get the manager to agree um I mean see I
manager to agree um I mean see I personally believe that at the end you
personally believe that at the end you are not going to win too many battles by
are not going to win too many battles by not having your manager
not having your manager Oh sorry about that uh uh so uh you
Oh sorry about that uh uh so uh you wouldn't win too many battles if your
wouldn't win too many battles if your manager is not extremely aligned with
manager is not extremely aligned with where you are right so even if it is not
where you are right so even if it is not working out properly or even if it is in
working out properly or even if it is in a situation where uh you know it is
a situation where uh you know it is taking multiple iterations it's okay
taking multiple iterations it's okay it's fine get your manager on board get
it's fine get your manager on board get your manager at the same side as you are
your manager at the same side as you are um because at the end I don't think
um because at the end I don't think you're going to like I said you're going
you're going to like I said you're going to do too well if your manager is not
to do too well if your manager is not aligned with uh at some point it will
aligned with uh at some point it will break and and and it's better to you
break and and and it's better to you know get it out of the way right at the
know get it out of the way right at the first step itself because at the end you
first step itself because at the end you know your manager will be one of the
know your manager will be one of the most important people in evaluating your
most important people in evaluating your work and in evaluating the result of
work and in evaluating the result of whatever you've built. So there's no
whatever you've built. So there's no need to kind of circumvent that and that
need to kind of circumvent that and that would be my recommendation. Even if it's
would be my recommendation. Even if it's taking longer, it's it's okay. You can
taking longer, it's it's okay. You can you should, you know, take that time and
you should, you know, take that time and give your manager the time to
give your manager the time to essentially become comfortable with the
essentially become comfortable with the uh with your hypothesis and give it.
Okay. Uh Bas saying great point having your own viewpoint makes total sense.
your own viewpoint makes total sense. Yes, absolutely. Like I said, right, I
Yes, absolutely. Like I said, right, I mean if it's your product, if you are
mean if it's your product, if you are working and if you are building that
working and if you are building that product, it should it's absolutely
product, it should it's absolutely essential that you have a viewpoint on
essential that you have a viewpoint on that. If you don't have a viewpoint on
that. If you don't have a viewpoint on that, you essentially are completely
that, you essentially are completely hamstrung. You're dependent on other
hamstrung. You're dependent on other people to tell you what should be built
people to tell you what should be built in the product. And that's something
in the product. And that's something that you have to fundamentally change
that you have to fundamentally change because everywhere else um because the
because everywhere else um because the product life cycle is which starts with
product life cycle is which starts with discovery and so on and so forth. It's
discovery and so on and so forth. It's always taught that you see data and you
always taught that you see data and you come up with answers and you see users
come up with answers and you see users and come up with answers. But a big
and come up with answers. But a big point that gets missed is one of the
point that gets missed is one of the biggest things that you would do as a PM
biggest things that you would do as a PM to do well as a PM is to generate better
to do well as a PM is to generate better hypothesis is to generate better
hypothesis is to generate better explanations. Once you start doing that,
explanations. Once you start doing that, the rest of the things keep uh keep
the rest of the things keep uh keep following. Okay. Mind is saying are
following. Okay. Mind is saying are there any frameworks for hypothesis
there any frameworks for hypothesis validation? We'll talk about that. So
validation? We'll talk about that. So we'll talk about how can you validate
we'll talk about how can you validate users and how can you validate business
users and how can you validate business next? Uh so just hang on to that. I
next? Uh so just hang on to that. I think we we will talk about that as
think we we will talk about that as well. Uh Gopti has has sent a lean
well. Uh Gopti has has sent a lean canvas validation framework which we can
canvas validation framework which we can use but I'll talk about the framework
use but I'll talk about the framework that I recommend in just a bit.
that I recommend in just a bit. Um
Um okay uh yes Bjit go ahead uh I will take
okay uh yes Bjit go ahead uh I will take some from uh chat also now.
some from uh chat also now. >> Yeah you almost cover Hi Mallay you
>> Yeah you almost cover Hi Mallay you almost cover everything. So how much
almost cover everything. So how much important the lean UX hypothesis
important the lean UX hypothesis because
because >> I mean see the lean UX hypothesis is is
>> I mean see the lean UX hypothesis is is is important but it actually is more
is important but it actually is more important in the next step which is the
important in the next step which is the design step that we do opportunity
design step that we do opportunity validation stage UX is not that uh I
validation stage UX is not that uh I mean UX is not that much at the
mean UX is not that much at the forefront because at this time you're
forefront because at this time you're deciding whether this is worth chasing
deciding whether this is worth chasing or not right so that is why it's you you
or not right so that is why it's you you just take take a step back and you know
just take take a step back and you know do only the validation in terms of
do only the validation in terms of whether it's worth solving or not the UX
whether it's worth solving or not the UX part of it or the lean UX validation
part of it or the lean UX validation that you do we do will do that we'll
that you do we do will do that we'll study that as well but we'll study it
study that as well but we'll study it when we study design which is the next
when we study design which is the next part of this life cycle
part of this life cycle >> one more question I think this is cover
>> one more question I think this is cover or not so what is the difference between
or not so what is the difference between ptm and phone pay while lots of user is
ptm and phone pay while lots of user is in phone pay and not in ptm
in phone pay and not in ptm is it the one impact for UX hypothesis.
is it the one impact for UX hypothesis. >> Yeah, of course. Of course. So, so see
>> Yeah, of course. Of course. So, so see there are multiple things, right? I
there are multiple things, right? I mean, if you look at and and I mean
mean, if you look at and and I mean Phone Pay and PTM are actually a good
Phone Pay and PTM are actually a good example, but if you have you have to
example, but if you have you have to evaluate their entire flow, right?
evaluate their entire flow, right? Right. From day zero to day one, Phone
Right. From day zero to day one, Phone Pay had a lot of focus on, you know,
Pay had a lot of focus on, you know, their wallet for a big amount of time.
their wallet for a big amount of time. Then after a while, it became focused on
Then after a while, it became focused on UPI and they had movie tickets and
UPI and they had movie tickets and multiple things going on. and phone pay
multiple things going on. and phone pay was more towards focusing only on UPI,
was more towards focusing only on UPI, scan and pay, scan and pay, scan and
scan and pay, scan and pay, scan and pay. So there are multiple kind of flows
pay. So there are multiple kind of flows and those things. So it it is difficult
and those things. So it it is difficult to bucket them into like one single
to bucket them into like one single reason in what the difference is. But we
reason in what the difference is. But we will do a case study on fintech. We'll
will do a case study on fintech. We'll do a case study of own pay versus Google
do a case study of own pay versus Google pay versus PTM in one of the hackathons
pay versus PTM in one of the hackathons as well where you will get a chance to
as well where you will get a chance to kind of evaluate it in a much more
kind of evaluate it in a much more deeper sense. But yeah, it is um I mean
deeper sense. But yeah, it is um I mean there's a multitude of things that that
there's a multitude of things that that are different for this.
are different for this. >> Thank you.
>> Thank you. >> Yeah.
>> Yeah. Um okay. So just asking if you feel if
Um okay. So just asking if you feel if you feel the problem Okay. Sorry. If you
you feel the problem Okay. Sorry. If you feel the problem that the problem space
feel the problem that the problem space PM have identified won't create
PM have identified won't create meaningful impact. No. So if your
meaningful impact. No. So if your validation uh in the next steps that we
validation uh in the next steps that we will discuss comes out with a result
will discuss comes out with a result that okay this is actually not a
that okay this is actually not a problem. the users are not really you
problem. the users are not really you know too bothered about this problem.
know too bothered about this problem. you go back to the problem space PM and
you go back to the problem space PM and present your hypothesis validation and
present your hypothesis validation and say that hey I went with what you had
say that hey I went with what you had talked about I created these hypothesis
talked about I created these hypothesis I spoke to users did business value
I spoke to users did business value we'll just talk about that in a bit on
we'll just talk about that in a bit on how you do it and it's come out with a
how you do it and it's come out with a position where it's not worth chasing
position where it's not worth chasing and then you iterate with them in terms
and then you iterate with them in terms of whether it's worth doing or not yes
of whether it's worth doing or not yes sushma this should be documented the
sushma this should be documented the template I showed is one of the ways to
template I showed is one of the ways to document it uh so that is how basically
document it uh so that is how basically it should be documented
Okay, Amit is saying continuation to Rashmi's question. How crucial deadlines
Rashmi's question. How crucial deadlines are when you are simply creating
are when you are simply creating hypothesis? I don't think there will be
hypothesis? I don't think there will be any se Yeah, you're right. Absolutely. I
any se Yeah, you're right. Absolutely. I mean, when you're at the hypothesis
mean, when you're at the hypothesis stage, deadlines are still far away
stage, deadlines are still far away because usually PMS don't work that
because usually PMS don't work that strongly against deadlines. Deadlines
strongly against deadlines. Deadlines are mostly work mostly, you know, a
are mostly work mostly, you know, a headache of design and developers
headache of design and developers because a PM is a thinking job, right?
because a PM is a thinking job, right? So you can get the right idea and you
So you can get the right idea and you can get the right thinking steps in like
can get the right thinking steps in like 5 minutes and you can take like 50 days
5 minutes and you can take like 50 days also. So usually a PM is not very
also. So usually a PM is not very hardressed against deadlines. Uh so
hardressed against deadlines. Uh so that's why it wouldn't be a challenge
that's why it wouldn't be a challenge and aligning your manager is is
and aligning your manager is is important. Uh but what is better is of
important. Uh but what is better is of course I mean what is more uh prone to
course I mean what is more uh prone to deadlines is is probably a um
deadlines is is probably a um like a developer or a designer.
All right. Moving on. The next part of what we are going to discuss is the
what we are going to discuss is the business value hypothesis. And this is
business value hypothesis. And this is of course easier because once again at
of course easier because once again at this stage we are only hypothesizing. So
this stage we are only hypothesizing. So we don't have uh we don't want to go to
we don't have uh we don't want to go to a data team or analytics and look at
a data team or analytics and look at numbers. We will do that. We will do
numbers. We will do that. We will do that when we refine the business value
that when we refine the business value in the next step that we're going to
in the next step that we're going to discuss. But right now the goal that we
discuss. But right now the goal that we have is to just look at business value
have is to just look at business value and fill in these things. Right? So you
and fill in these things. Right? So you go ahead and fill in the decision makers
go ahead and fill in the decision makers of you know who are the decision makers
of you know who are the decision makers that are available. You go ahead and
that are available. You go ahead and talk about who out of the business teams
talk about who out of the business teams will be informed that we are building
will be informed that we are building something like this and then talk about
something like this and then talk about that and then the business goals again
that and then the business goals again qualitatively the business goal is to
qualitatively the business goal is to improve the driver experience because
improve the driver experience because that will lead to better business
that will lead to better business outcomes. Quantitatively it will
outcomes. Quantitatively it will increase the driver supply and reduce
increase the driver supply and reduce cancellations. And non-goal is directly
cancellations. And non-goal is directly increasing revenue. Like I mentioned,
increasing revenue. Like I mentioned, the same non-goal concept here applies
the same non-goal concept here applies as well where I'm I'm not allowing to
as well where I'm I'm not allowing to get more rides because of this. So I I'm
get more rides because of this. So I I'm not going to get more rides because of
not going to get more rides because of this or I'm not going to increase
this or I'm not going to increase revenue because of this. Once again,
revenue because of this. Once again, just the same to have alignment on this.
just the same to have alignment on this. For the Uber example again, decision
For the Uber example again, decision makers for reviews would be the catalog
makers for reviews would be the catalog team, product leadership, uh certain
team, product leadership, uh certain catalog leaders
catalog leaders and then uh informed would be customer
and then uh informed would be customer support, marketing, sales, uh I mean
support, marketing, sales, uh I mean everyone uh would be there. You can
everyone uh would be there. You can again uh figure out who is going to be
again uh figure out who is going to be uh there. The business goals of course
uh there. The business goals of course qualitatively
qualitatively in improves user experience, allows them
in improves user experience, allows them to get value for money
to get value for money and quantitatively reduces refunds,
and quantitatively reduces refunds, reduces uh cancellations after orders
reduces uh cancellations after orders have been placed
have been placed and non- goals is increasing conversion
and non- goals is increasing conversion or revenue directly from the
or revenue directly from the orders that are being placed. Right? So
orders that are being placed. Right? So very clearly gives me a very clear
very clearly gives me a very clear picture of uh all the things that are
picture of uh all the things that are going to be there. Very quick on if
going to be there. Very quick on if you've figured out the user profile and
you've figured out the user profile and user hypothesis business hypothesis
user hypothesis business hypothesis should be straightforward. Again you
should be straightforward. Again you don't have to go really deep into it. Uh
don't have to go really deep into it. Uh you have to of course mention to your
you have to of course mention to your manager as well that this is a briefing
manager as well that this is a briefing a short briefing where I'm just trying
a short briefing where I'm just trying to talk about uh you know what is
to talk about uh you know what is exactly my feature. Once again like
exactly my feature. Once again like Swapnil said if you feel that hey
Swapnil said if you feel that hey decision makers and informed is
decision makers and informed is something that remains fairly constant.
something that remains fairly constant. We have a set of people that we inform
We have a set of people that we inform everything about and I don't want to
everything about and I don't want to fill this you can of course remove this
fill this you can of course remove this and just stick to the goals and say that
and just stick to the goals and say that qualitatively I want to improve user
qualitatively I want to improve user experience. I want them to get value for
experience. I want them to get value for money. Quantitatively I want to reduce
money. Quantitatively I want to reduce um refunds or cancellations and um that
um refunds or cancellations and um that is my game here. And I'm not focusing on
is my game here. And I'm not focusing on revenue. So I'm not increasing the
revenue. So I'm not increasing the conversion or revenue that is coming
conversion or revenue that is coming directly from the orders. So that's not
directly from the orders. So that's not part of the expectation that you would
part of the expectation that you would have right once this business value
have right once this business value hypothesis is done uh on sculp of course
hypothesis is done uh on sculp of course you can just attach it here so that you
you can just attach it here so that you can have a access to this later. So I'll
can have a access to this later. So I'll just show this uh
just show this uh uh okay sorry Amazon
uh okay sorry Amazon hypothesis
hypothesis and save it. So this is going to come
and save it. So this is going to come here where you can just go ahead and uh
here where you can just go ahead and uh uh present it. Um I mean just go ahead
uh present it. Um I mean just go ahead and if you click on it, it is just going
and if you click on it, it is just going to open the file that you just created
to open the file that you just created uh for you to use it and then you can
uh for you to use it and then you can complete this step. Right? So the first
complete this step. Right? So the first step create user value create uh a
step create user value create uh a hypothesis on that strategic fit is
hypothesis on that strategic fit is fairly static like a lot of you
fairly static like a lot of you mentioned. So you that can remain static
mentioned. So you that can remain static and then user value can become from
and then user value can become from there and then you complete this step.
there and then you complete this step. Okay. Now once this is complete the next
Okay. Now once this is complete the next step and the uh and the most important
step and the uh and the most important step whenever you're building any
step whenever you're building any feature is research where your goal now
feature is research where your goal now is to refine all the value that you have
is to refine all the value that you have seen from the users coming in. Your job
seen from the users coming in. Your job is to say that okay I have so many
is to say that okay I have so many things that have uh uh that I have
things that have uh uh that I have hypothesized in my original manager
hypothesized in my original manager briefing how can I validate it right so
briefing how can I validate it right so let's have a look at that template as
let's have a look at that template as well now refining user value I'll just
well now refining user value I'll just make a copy of this
right so refining I think user value has got these four steps. You begin by
got these four steps. You begin by planning the user validation which is
planning the user validation which is how are you going to get this validated.
how are you going to get this validated. We'll talk about that. Second is you're
We'll talk about that. Second is you're going to conduct the user validation.
going to conduct the user validation. You're going to actually go ahead and
You're going to actually go ahead and look at it. Third is you're going to
look at it. Third is you're going to consolidate this and fourth is you're
consolidate this and fourth is you're going to create something that we will
going to create something that we will call a user value map which is nothing
call a user value map which is nothing but the final outcome of whatever is
but the final outcome of whatever is going on. Right? We'll take a B2B
going on. Right? We'll take a B2B example for this where our example would
example for this where our example would be for increasing the usage of a feature
be for increasing the usage of a feature called threads on Slack for uh
folks who uh want to understand what threads is. So
want to understand what threads is. So threads are very simple. So Slack you
threads are very simple. So Slack you can have messages. Now for each message
can have messages. Now for each message you can create a thread. So this is how
you can create a thread. So this is how the thread looks where it can come as
the thread looks where it can come as replies and then once you click on the
replies and then once you click on the message that message is available here.
message that message is available here. So what this allows you to do is it
So what this allows you to do is it allows you to see this uh allows you to
allows you to see this uh allows you to uh very uh make your chat very concise.
uh very uh make your chat very concise. So don't uh otherwise if you just have
So don't uh otherwise if you just have messages that they might get lost. So
messages that they might get lost. So you could use that from there. Now what
you could use that from there. Now what we are taking is as a hypothesis here or
we are taking is as a hypothesis here or what we have taken as a hypothesis here
what we have taken as a hypothesis here is we want to increase the value of the
is we want to increase the value of the threads or we want to increase the
threads or we want to increase the threads of slack. So the first part of
threads of slack. So the first part of it is to plan the user validation and
it is to plan the user validation and that is what we're going to view there.
that is what we're going to view there. So now you would have multiple
So now you would have multiple hypothesis here. I think someone had
hypothesis here. I think someone had mentioned that what happens if we have
mentioned that what happens if we have multiple I think another had mentioned.
multiple I think another had mentioned. So yes you would have multiple
So yes you would have multiple hypothesis. This would be part of the
hypothesis. This would be part of the user problem hypothesis that you have
user problem hypothesis that you have put here. So here I have put five user
put here. So here I have put five user hypothesis. So the first is a hypothesis
hypothesis. So the first is a hypothesis saying now all of these hypothesis are
saying now all of these hypothesis are why people are not using threads on
why people are not using threads on Slack. Once again to just uh for people
Slack. Once again to just uh for people who have not used Slack just to show you
who have not used Slack just to show you on Slack you can send a message like
on Slack you can send a message like this and then once you click uh you can
this and then once you click uh you can then click a reply icon here and then it
then click a reply icon here and then it opens a thread and that thread is inside
opens a thread and that thread is inside that message. So now there are a few
that message. So now there are a few hypothesis. The first is users may be
hypothesis. The first is users may be treating Slack as unnecessary clutter
treating Slack as unnecessary clutter due to poor onboarding or lack of
due to poor onboarding or lack of benefits. Hypothesis is that improving
benefits. Hypothesis is that improving the discoverability of this will
the discoverability of this will increase adoption. So this means that
increase adoption. So this means that people just there's just been a bad
people just there's just been a bad onboarding and that is why people don't
onboarding and that is why people don't realize the value of it and that is why
realize the value of it and that is why they just look at it as something that
they just look at it as something that is unnecessary.
is unnecessary. The second hypothesis is that a
The second hypothesis is that a hierarchy in Slack usage where what
hierarchy in Slack usage where what happens is very active dominant users
happens is very active dominant users dictate how to use Slack and if they are
dictate how to use Slack and if they are just replying to messages without using
just replying to messages without using threads everyone is just going to follow
threads everyone is just going to follow it. Right? So what is the perceived
it. Right? So what is the perceived etiquette of using slack? That is our
etiquette of using slack? That is our hypothesis that hey our hypothesis is
hypothesis that hey our hypothesis is that we are going to go ahead and u
that we are going to go ahead and u say that because there is a hierarchy in
say that because there is a hierarchy in companies people may just blindly follow
companies people may just blindly follow the structure of using slack that their
the structure of using slack that their seniors are and that is why there isn't
seniors are and that is why there isn't going to be enough adoption. The third
going to be enough adoption. The third is the reply in thread button is not
is the reply in thread button is not intuitive or not emphasized. That is why
intuitive or not emphasized. That is why users just completely ignore it and a
users just completely ignore it and a redesign may be something that will
redesign may be something that will help. The fourth is people don't see
help. The fourth is people don't see value. So even if people have seen that
value. So even if people have seen that threads is something that they say they
threads is something that they say they don't say that they don't see that there
don't say that they don't see that there is enough time that is being saved,
is enough time that is being saved, there isn't enough productivity that is
there isn't enough productivity that is being improved and so on and so forth
being improved and so on and so forth and they just go back to their default
and they just go back to their default messages. The fifth is that uh you know
messages. The fifth is that uh you know it it varies with channel to channel
it it varies with channel to channel where on some channels where threads are
where on some channels where threads are being used it is fine but uh on other
being used it is fine but uh on other channels it's not. So people know they
channels it's not. So people know they get value out of it but it's just
get value out of it but it's just inconsistent where uh since using uh
inconsistent where uh since using uh normal messages is easier they just end
normal messages is easier they just end up uh you know going back to them. Now
up uh you know going back to them. Now all of these hypothesis are coming
all of these hypothesis are coming without any validation or without any
without any validation or without any research which means that there is an
research which means that there is an acute need to go ahead and validate. How
acute need to go ahead and validate. How many of these are actually making sense
many of these are actually making sense and how many of these are just uh you
and how many of these are just uh you know my or our version of what is going
know my or our version of what is going on and not true.
on and not true. There are five validation methods or
There are five validation methods or actually four validation methods that
actually four validation methods that you could choose from each of these
you could choose from each of these hypothesis. Right? So if you have like
hypothesis. Right? So if you have like several hypothesis, you can choose
several hypothesis, you can choose multiple ones and you can do them
multiple ones and you can do them together. So but essentially what
together. So but essentially what happens is all of these hypothesis that
happens is all of these hypothesis that you have here are uh you can either
you have here are uh you can either choose to validate them by data. We're
choose to validate them by data. We're going to talk about data uh as well. um
going to talk about data uh as well. um in in further classes where we're going
in in further classes where we're going to talk about how can you analyze data
to talk about how can you analyze data but you can choose to you know do select
but you can choose to you know do select that okay I'm going to select data and
that okay I'm going to select data and I'm going to validate it from this so
I'm going to validate it from this so based on the hypothesis you decide which
based on the hypothesis you decide which of these four is the best method to
of these four is the best method to validate or invalidate it right so for
validate or invalidate it right so for example um users treating them as uh
example um users treating them as uh unnecessary clutter and not uh using
unnecessary clutter and not uh using them is something that you may say that
them is something that you may say that hey this is something that I want to see
hey this is something that I want to see people use it and only then I'll be able
people use it and only then I'll be able to do it or actually I want to see
to do it or actually I want to see people in person and then only I'll be
people in person and then only I'll be able to do it and based on that you
able to do it and based on that you decide that field interviews and virtual
decide that field interviews and virtual interviews is a part of this or not
interviews is a part of this or not similarly the hierarchy point is
similarly the hierarchy point is something that you feel that hey this is
something that you feel that hey this is something that data can tell this is
something that data can tell this is something that I I can send an anonymous
something that I I can send an anonymous survey to people and once they fill it I
survey to people and once they fill it I will know that whether I'm thinking this
will know that whether I'm thinking this is correct or not and that may validate
is correct or not and that may validate this or reply to thread is also
this or reply to thread is also something that I can do a survey and
something that I can do a survey and data analysis and look at it. Um the
data analysis and look at it. Um the value part is uh you know where s time
value part is uh you know where s time is being saved or not. This is not
is being saved or not. This is not something that I will know by data
something that I will know by data because this is a very personal user
because this is a very personal user experience of whether they're getting
experience of whether they're getting value out of this or not. So a survey
value out of this or not. So a survey may not be good enough for me to do
may not be good enough for me to do this. I can actually go ahead and you
this. I can actually go ahead and you know do a interview to find out and then
know do a interview to find out and then channel by channel is also something
channel by channel is also something that I can look at data I can look at
that I can look at data I can look at whether something is being used more in
whether something is being used more in particular channels like threads is
particular channels like threads is being used more in 20% of the channels
being used more in 20% of the channels of a company and not used in 80 and so
of a company and not used in 80 and so on and so forth. So after you've
on and so forth. So after you've finalized your hypothesis and got an
finalized your hypothesis and got an alignment from your manager, you come
alignment from your manager, you come here where you start validating the user
here where you start validating the user hypothesis and the first thing you do is
hypothesis and the first thing you do is you write down the hypothesis here and
you write down the hypothesis here and decide for each of the hypothesis what
decide for each of the hypothesis what methods are you going to use. Right? The
methods are you going to use. Right? The methods are very self-explanatory.
methods are very self-explanatory. uh if you look at it data is can you
uh if you look at it data is can you find data points that validate or
find data points that validate or invalidate this. Typically uh you know
invalidate this. Typically uh you know usage related pointers here are very
usage related pointers here are very useful. For example, you may go ahead
useful. For example, you may go ahead and say that hey if uh my hypothesis is
and say that hey if uh my hypothesis is that threads on slack is being used on
that threads on slack is being used on you know five channels and not being
you know five channels and not being used on 10. I can very just look at data
used on 10. I can very just look at data and say that is it true or not. Right?
and say that is it true or not. Right? In the previous example, uh if uh if
In the previous example, uh if uh if reviews was a uh was a pointer that I
reviews was a uh was a pointer that I that I had, I could very go very quickly
that I had, I could very go very quickly go ahead and analyze uh the uh the the
go ahead and analyze uh the uh the the re the refund or cancellation reasons
re the refund or cancellation reasons that people have given and based on that
that people have given and based on that decide that this is exactly what I going
decide that this is exactly what I going to say, right? Um survey is something
to say, right? Um survey is something that of course takes a lot more time
that of course takes a lot more time that will take a lot more uh um u that
that will take a lot more uh um u that that would that can answer some quick
that would that can answer some quick objective questions. So a survey can
objective questions. So a survey can answer a very quick objective question
answer a very quick objective question like saying that hey uh
like saying that hey uh is is there a hierarchy that is causing
is is there a hierarchy that is causing this problem? Now if the answer could be
this problem? Now if the answer could be yes the answer could be no you go ahead
yes the answer could be no you go ahead and finalize whether the answer is yes
and finalize whether the answer is yes or no and run a quick survey for it.
or no and run a quick survey for it. Field and virtual is more when you're
Field and virtual is more when you're trying to build something new when
trying to build something new when you're trying to analyze whether this is
you're trying to analyze whether this is going to give value to my users or not
going to give value to my users or not and that is how you can go ahead and
and that is how you can go ahead and look at it. So based on this logic very
look at it. So based on this logic very simply you just break this down um into
simply you just break this down um into these multiple
these multiple uh directions and then you figure out
uh directions and then you figure out who you are going to interview right so
who you are going to interview right so you figure out exactly who you are going
you figure out exactly who you are going to interview by figuring out your target
to interview by figuring out your target audience right now what is your target
audience right now what is your target audience your target audience is again
audience your target audience is again very closely related to the user
very closely related to the user profiles that you created earlier but
profiles that you created earlier but all you need to do on your target
all you need to do on your target audience in this case is give them these
audience in this case is give them these uh answers where you go ahead and uh you
uh answers where you go ahead and uh you know present it as attributes, present
know present it as attributes, present it as stakeholdership
it as stakeholdership uh the need and the profile. So you
uh the need and the profile. So you basically create a profile who you are
basically create a profile who you are going to interview. So in this case for
going to interview. So in this case for example uh one of the hypothesis that we
example uh one of the hypothesis that we had was hierarchy. One of the hypothesis
had was hierarchy. One of the hypothesis we had was that there are you know
we had was that there are you know dominant users who will dictate norms.
dominant users who will dictate norms. So we have a attribute of someone who's
So we have a attribute of someone who's a reluctant collaborator. So they are
a reluctant collaborator. So they are usually an enduser. They rarely initiate
usually an enduser. They rarely initiate threats. They only contribute to
threats. They only contribute to discussions. They need an easy and
discussions. They need an easy and intuitive way to keep track of their
intuitive way to keep track of their discussions without feeling overwhelmed.
discussions without feeling overwhelmed. Between 26 to 32 in their age.
Between 26 to 32 in their age. Experience is 1 to 3 years. Moderately
Experience is 1 to 3 years. Moderately techsavvy. Mobile first. Uses Slack 3 to
techsavvy. Mobile first. Uses Slack 3 to five times daily. But often misses key
five times daily. But often misses key context. Right. So very clearly we have
context. Right. So very clearly we have picked one profile. We've given the
picked one profile. We've given the stakeholdership which is what is their
stakeholdership which is what is their role in the usage of the feature. We've
role in the usage of the feature. We've given need which is what need does this
given need which is what need does this persona have and then what is their
persona have and then what is their demographic. Similarly, we've picked the
demographic. Similarly, we've picked the second dominant person as well. We've
second dominant person as well. We've looked at some power user who who
looked at some power user who who dictates team norms, encourages best
dictates team norms, encourages best practices but struggles with junior team
practices but struggles with junior team members adopting it where they want
members adopting it where they want clarity and cohesion in the channel
clarity and cohesion in the channel communication that they have but uh you
communication that they have but uh you know they are unable to manage multiple
know they are unable to manage multiple work streams. They're of course much
work streams. They're of course much higher in in in terms of their age and
higher in in in terms of their age and experience. Uh they use Slack on
experience. Uh they use Slack on multiple devices. They engage deeply
multiple devices. They engage deeply with you know advanced features like
with you know advanced features like threads, reminders, integrations all of
threads, reminders, integrations all of that. So they are power users of Slack
that. So they are power users of Slack but they need to now you know uh we need
but they need to now you know uh we need to interview them and understand there
to interview them and understand there usually if you have two to three
usually if you have two to three personas related to the hypothesis that
personas related to the hypothesis that you have. It's not that you can only you
you have. It's not that you can only you know talk about the uh second hypothesis
know talk about the uh second hypothesis with this person. You can actually
with this person. You can actually validate all of them with them if you're
validate all of them with them if you're speaking to them or if you're serving
speaking to them or if you're serving them or whatever you're doing uh with
them or whatever you're doing uh with them. But the idea is that you pick up
them. But the idea is that you pick up two to three target audience that have
two to three target audience that have correlation with some of the key factors
correlation with some of the key factors that are mentioned here. Right now once
that are mentioned here. Right now once this is done you go ahead and uh conduct
this is done you go ahead and uh conduct the user validation which means that you
the user validation which means that you go ahead and I mean the target audience
go ahead and I mean the target audience is the same and you go ahead and speak
is the same and you go ahead and speak to them. Now if your validation method
to them. Now if your validation method is data then essentially what you would
is data then essentially what you would need to do is you would need to find
need to do is you would need to find that corresponding data and present it.
that corresponding data and present it. We'll talk about mix panel as well in
We'll talk about mix panel as well in later sessions when we talk about uh how
later sessions when we talk about uh how to you know look at different data
to you know look at different data analysis tools. But essentially for now
analysis tools. But essentially for now if there is something that you have
if there is something that you have hypothesized of saying that okay users
hypothesized of saying that okay users drop off because of X then just looking
drop off because of X then just looking at the data of whether there is some
at the data of whether there is some proxy that you can find based on that
proxy that you can find based on that and you just fill it. A survey is again
and you just fill it. A survey is again a very pointed objective survey where
a very pointed objective survey where you go ahead and ask users questions on
you go ahead and ask users questions on some of the things that you've
some of the things that you've mentioned. So for example here if you
mentioned. So for example here if you were trying to ask a survey on the
were trying to ask a survey on the second one you could create a survey
second one you could create a survey that asks users u in terms of you know
that asks users u in terms of you know presenting them an idea of saying that
presenting them an idea of saying that hey um who sets the norms for slack uh
hey um who sets the norms for slack uh are you intimidated to kind of initiate
are you intimidated to kind of initiate a thread at times or uh what is who sets
a thread at times or uh what is who sets the slack etiquette uh is is that
the slack etiquette uh is is that something that you're mindful of and
something that you're mindful of and things like that that you can mention to
things like that that you can mention to validate or invalidate it field and
validate or invalidate it field and virtual have pretty much the same thing.
virtual have pretty much the same thing. Field of course has more observations
Field of course has more observations linked to it because if you are you know
linked to it because if you are you know directly working with folks in the field
directly working with folks in the field you would have a very clear
you would have a very clear understanding with uh with them and and
understanding with uh with them and and focusing on them. So that's something
focusing on them. So that's something that you could do uh directly and
that you could do uh directly and present it. And virtual is also the same
present it. And virtual is also the same almost. Now I mean there's not too much
almost. Now I mean there's not too much of a difference between virtual and
of a difference between virtual and field. But in virtual and field
field. But in virtual and field interviews what you would need to do is
interviews what you would need to do is you would need to prepare questions.
you would need to prepare questions. Right. U so now what are these
Right. U so now what are these questions? These questions are questions
questions? These questions are questions which help you get validation from
which help you get validation from whatever you have interviewed the user.
whatever you have interviewed the user. So whenever you're doing a field and um
So whenever you're doing a field and um virtual interview, your job is to
virtual interview, your job is to prepare these questions. Your job is to
prepare these questions. Your job is to prepare all of these pointers one by one
prepare all of these pointers one by one and then go ahead and do this
and then go ahead and do this questionnaire with the user. So in the
questionnaire with the user. So in the slack thread example, can you tell me?
slack thread example, can you tell me? So now when you're preparing these
So now when you're preparing these questions uh you need to structure them
questions uh you need to structure them in a way that you have questions to
in a way that you have questions to build rapo first
build rapo first then you have questions to build
then you have questions to build confidence around your hypothesis
confidence around your hypothesis then finally validate or invalidate your
then finally validate or invalidate your hypothesis at the end and then note down
hypothesis at the end and then note down these answers from there. What happens
these answers from there. What happens uh especially when you interview users
uh especially when you interview users see at the end whenever you're speaking
see at the end whenever you're speaking to someone one of the things that you
to someone one of the things that you have to be mindful of is that they are
have to be mindful of is that they are trying to tell you the best that they
trying to tell you the best that they think for you right so when you ask
think for you right so when you ask someone that hey is this a problem
someone that hey is this a problem statement that uh you know I should
statement that uh you know I should solve now their job is there they are
solve now their job is there they are trying to find out whatever in their
trying to find out whatever in their limited capacity can they think is the
limited capacity can they think is the best for
best for Right? Which is why it is in their
Right? Which is why it is in their interest or it is in their um uh how do
interest or it is in their um uh how do I say it is in their second nature to
I say it is in their second nature to kind of just go ahead and uh you know
kind of just go ahead and uh you know give you an idea or give you a pointer
give you an idea or give you a pointer directly. So they can just go ahead and
directly. So they can just go ahead and give you you know one point directly and
give you you know one point directly and say that hey this is something that you
say that hey this is something that you could go ahead and present or this is
could go ahead and present or this is something that you could go ahead and
something that you could go ahead and talk about. right now that is something
talk about. right now that is something that you can achieve uh by or or
that you can achieve uh by or or circumventing this is something that you
circumventing this is something that you can achieve by breaking down your
can achieve by breaking down your problems like this. So you first begin
problems like this. So you first begin with questions to build rapper then you
with questions to build rapper then you talk about questions to build confidence
talk about questions to build confidence around the hypothesis and then you talk
around the hypothesis and then you talk about some peak questions that validate
about some peak questions that validate or invalidate your hypothesis.
or invalidate your hypothesis. So questions to build rapo are you know
So questions to build rapo are you know can we tell me a typical day at work?
can we tell me a typical day at work? Second how do you usually communicate
Second how do you usually communicate with your team? Third what are your
with your team? Third what are your favorite and these favorite tools that
favorite and these favorite tools that you use daily. Now these are questions
you use daily. Now these are questions not really related to slack, not really
not really related to slack, not really related to uh threads in general but
related to uh threads in general but these are questions that build a rapper
these are questions that build a rapper so that there is a starting point and
so that there is a starting point and also answers to these questions can call
also answers to these questions can call back later when they answer other
back later when they answer other questions so that you can reference them
questions so that you can reference them again. So for example, if I mention here
again. So for example, if I mention here what are some of the favorite least
what are some of the favorite least favorite tools and someone mentions
favorite tools and someone mentions slack or someone doesn't mention slack
slack or someone doesn't mention slack or things like that you have a way to
or things like that you have a way to call back to this so that you have you
call back to this so that you have you know some understanding or you have some
know some understanding or you have some rapper built up the more time you spend
rapper built up the more time you spend on this the better answers you will get
on this the better answers you will get later on. So that's something that you
later on. So that's something that you can um you know directly talk about or
can um you know directly talk about or feel. The more time you spend on
feel. The more time you spend on building this rapper the better it
building this rapper the better it becomes for you. The second phase is to
becomes for you. The second phase is to build confidence around this hypothesis
build confidence around this hypothesis right where now you ask questions that
right where now you ask questions that are closely related to what your
are closely related to what your hypothesis is. So if you're trying to
hypothesis is. So if you're trying to validate uh your idea that uh you know
validate uh your idea that uh you know manager dictates the norms uh now when
manager dictates the norms uh now when you go ahead and say that hey around
you go ahead and say that hey around threads when you're replying to someone
threads when you're replying to someone how do you decide whether to start a
how do you decide whether to start a thread or reply in the chat right so now
thread or reply in the chat right so now based on this if they go ahead and
based on this if they go ahead and answer that no there is a norm everyone
answer that no there is a norm everyone does just DM so I just use that then you
does just DM so I just use that then you realize that okay there is validation
realize that okay there is validation for this right uh have you ever been
for this right uh have you ever been confused about how and where to respond
confused about how and where to respond in the Slack conversation. So again, not
in the Slack conversation. So again, not asking questions that are, you know,
asking questions that are, you know, very uh blatant yes no questions as
very uh blatant yes no questions as right now. Right now I'm just building
right now. Right now I'm just building confidence. Why? Because if I ask
confidence. Why? Because if I ask blatant uh yes no questions right away,
blatant uh yes no questions right away, uh I run the risk of the other person
uh I run the risk of the other person thinking that okay, the best for me
thinking that okay, the best for me right now is to just uh you know talk to
right now is to just uh you know talk to this person and answer and say yes to
this person and answer and say yes to whatever they're saying and that is why
whatever they're saying and that is why they may not give you an insightful
they may not give you an insightful answer.
answer. The third question around this would be
The third question around this would be can you call recall a time where you
can you call recall a time where you missed an important message and there
missed an important message and there was a problem because of that. So once
was a problem because of that. So once again you're building confidence around
again you're building confidence around threads. You're building confidence
threads. You're building confidence around what you're going to talk about.
around what you're going to talk about. Right. The final is the peak questions
Right. The final is the peak questions to to highlight where you go ahead and
to to highlight where you go ahead and uh go ahead and now directly answer the
uh go ahead and now directly answer the or directly ask the questions around
or directly ask the questions around what your hypothesis is. So you go ahead
what your hypothesis is. So you go ahead and answer the questions around saying
and answer the questions around saying that hey uh do you even find slack
that hey uh do you even find slack threads useful? Uh if slack threads were
threads useful? Uh if slack threads were removed how would your day change? Um
removed how would your day change? Um have you stopped some uh following
have you stopped some uh following something because it was too cluttered?
something because it was too cluttered? So you can basically now go ahead and uh
So you can basically now go ahead and uh mention this as I mean ask final level
mention this as I mean ask final level questions where you go ahead and present
questions where you go ahead and present this uh directly from there. Right? So
this uh directly from there. Right? So essentially what you do is uh you go
essentially what you do is uh you go ahead and uh create peak questions where
ahead and uh create peak questions where you ask them finally whether they are
you ask them finally whether they are aligned with the hypothesis or not and
aligned with the hypothesis or not and then you keep noting this with multiple
then you keep noting this with multiple users. Usually recommendation is to
users. Usually recommendation is to interview five to 10 users at least five
interview five to 10 users at least five to 10 proper interviews where you can
to 10 proper interviews where you can ask all of these questions and present
ask all of these questions and present them. So that's a good uh you know way
them. So that's a good uh you know way to go ahead and uh mention all of this
to go ahead and uh mention all of this one by one. So that's something that you
one by one. So that's something that you should do. Um and of course structure
should do. Um and of course structure your questions like this so that you are
your questions like this so that you are able to do that
able to do that right. Um once you have done this then
right. Um once you have done this then you go ahead and consolidate whatever
you go ahead and consolidate whatever you found. So target audience is of
you found. So target audience is of course the same but you consolidate your
course the same but you consolidate your observations on what has happened where
observations on what has happened where during the interviews users like Ria
during the interviews users like Ria mentioned they continue discussions in
mentioned they continue discussions in the main channel because it feels easier
the main channel because it feels easier and they don't want to lose visibility.
and they don't want to lose visibility. this supports one of the hypothesis that
this supports one of the hypothesis that we have. You could change it to this
we have. You could change it to this contradicts or this is a new finding. If
contradicts or this is a new finding. If it's a new finding, you can choose to
it's a new finding, you can choose to tackle it now or you can choose to
tackle it now or you can choose to tackle it in another feature. But if it
tackle it in another feature. But if it contradicts and a lot of your hypothesis
contradicts and a lot of your hypothesis are already contradicted then it's it's
are already contradicted then it's it's something that uh you know you would
something that uh you know you would have to go back uh to the drawing board
have to go back uh to the drawing board and figure out right similarly Amit has
and figure out right similarly Amit has shared that in his team threads are
shared that in his team threads are strictly enforced while in cross
strictly enforced while in cross function or client facing channel
function or client facing channel everyone ignores the norms. So this
everyone ignores the norms. So this supports your fifth hypothesis where you
supports your fifth hypothesis where you say that okay the hypothesis was that in
say that okay the hypothesis was that in some threads people follow and in some
some threads people follow and in some threads people don't follow and uh there
threads people don't follow and uh there and you had an observation again on top
and you had an observation again on top of it as well. So once you've done this
of it as well. So once you've done this uh once you have uh uh go ahead and then
uh once you have uh uh go ahead and then begin from there you actually mention
begin from there you actually mention all the hypothesis that were there that
all the hypothesis that were there that were supported or contradicted and any
were supported or contradicted and any other areas that you saw from there. So
other areas that you saw from there. So you finally synthesize the problems
you finally synthesize the problems whatever underlying insight was there
whatever underlying insight was there from this. So basically your hypothesis
from this. So basically your hypothesis that you had you basically now convert
that you had you basically now convert it into uh I mean if you remember what
it into uh I mean if you remember what we did here was we did problem severity
we did here was we did problem severity and uh alternatives. So you now go ahead
and uh alternatives. So you now go ahead and write it but now you go ahead and
and write it but now you go ahead and write this with a lot more authority
write this with a lot more authority because you've now researched or looked
because you've now researched or looked at data or surveyed users and finally
at data or surveyed users and finally got the outcome from them.
got the outcome from them. At the end once all of it is done you
At the end once all of it is done you create this user value map. This map is
create this user value map. This map is the same of what you began with when you
the same of what you began with when you were beginning with your original
were beginning with your original hypothesis but now you're doing it in a
hypothesis but now you're doing it in a much more structured way because now you
much more structured way because now you have the confidence of speaking to users
have the confidence of speaking to users and understanding what they need. So the
and understanding what they need. So the profile the problem and the goals in
profile the problem and the goals in this case now when you're presenting
this case now when you're presenting this you cannot actually go ahead and
this you cannot actually go ahead and just do a surface level thing. In this
just do a surface level thing. In this case, you will have to talk about you
case, you will have to talk about you know uh how many people did you meet?
know uh how many people did you meet? What was the number of people that said
What was the number of people that said that this is going to be valuable? What
that this is going to be valuable? What is the number of people that said it is
is the number of people that said it is not going to be valuable and so on and
not going to be valuable and so on and so forth. So that is why the user value
so forth. So that is why the user value map is basically a new version of the
map is basically a new version of the user profile hypothesis that you had
user profile hypothesis that you had here but now it's much more uh in stone
here but now it's much more uh in stone because now everything that you've said
because now everything that you've said has been u uh you know validated and and
has been u uh you know validated and and and verified from users. If you are in a
and verified from users. If you are in a situation where you see a lot of
situation where you see a lot of contradictions where you see that okay
contradictions where you see that okay we began with something that we thought
we began with something that we thought but clearly this is not something that
but clearly this is not something that people value or people are aligned with
people value or people are aligned with then you know you go ahead and actually
then you know you go ahead and actually take that call and say that okay this is
take that call and say that okay this is something that uh um that we won't be
something that uh um that we won't be going ahead and uh following or this is
going ahead and uh following or this is something that we won't be going ahead
something that we won't be going ahead and looking at. And what we're going to
and looking at. And what we're going to do is we're going to go back and rework
do is we're going to go back and rework our hypothesis based on the in
our hypothesis based on the in observations that we've had from our
observations that we've had from our users.
users. Right? Finally, you present them into
Right? Finally, you present them into your user value map where you go ahead
your user value map where you go ahead and present them as the profile, the
and present them as the profile, the problem, and the goals. And in this
problem, and the goals. And in this case, now a lot of your user validation
case, now a lot of your user validation work, right? If you've looked at some
work, right? If you've looked at some data, if you've surveyed users, if
data, if you've surveyed users, if you've profiled them, if you've spoken
you've profiled them, if you've spoken to them, now you actually have data. Now
to them, now you actually have data. Now you actually have the point that okay
you actually have the point that okay I've spoken to you know 10 users we've
I've spoken to you know 10 users we've surveyed 1,000 users we've looked at
surveyed 1,000 users we've looked at data from 100,000 users and you can
data from 100,000 users and you can mention all of that here
right and also once this is done you can of course go ahead and attach it once
of course go ahead and attach it once again to this so I'm just going to name
again to this so I'm just going to name it user value
it user value slack
slack threads and save it and I can complete
threads and save it and I can complete read this as well. So this is going to
read this as well. So this is going to be available for you to look at whenever
be available for you to look at whenever whatever document you had created at
whatever document you had created at that time and you move on to refining
that time and you move on to refining the business value. Before we go to
the business value. Before we go to defining the business value, the final
defining the business value, the final part of what we're going to cover today,
part of what we're going to cover today, uh we'll just pause for questions.
uh we'll just pause for questions. Questions can be asked on chat or you
Questions can be asked on chat or you can just raise your hand and start
can just raise your hand and start asking them.
hello. I cannot hear me. Ah, okay. Um so we'll
I cannot hear me. Ah, okay. Um so we'll start from uh
Yeah, I was just hearing myself for some reason. Ah. Um, okay. So, with we'll
reason. Ah. Um, okay. So, with we'll start with Mahindra's question on chat.
start with Mahindra's question on chat. If competitors have launched any new
If competitors have launched any new feature, how can we use their
feature, how can we use their research to hypothesis? So uh of course
research to hypothesis? So uh of course that is one valid way of doing um you
that is one valid way of doing um you know uh hypothesis validation which is
know uh hypothesis validation which is basically looking at your competitors
basically looking at your competitors and deciding if they have done something
and deciding if they have done something and you know they have put their mind
and you know they have put their mind and thought into it. So it might be
and thought into it. So it might be something that would be interesting.
something that would be interesting. However, I always recommend at least the
However, I always recommend at least the user validation to be done by speaking
user validation to be done by speaking to your users once. Maybe if your
to your users once. Maybe if your competitor has done something similar,
competitor has done something similar, you can reduce the number of people you
you can reduce the number of people you speak to, but at least uh you know a few
speak to, but at least uh you know a few users spoken to, a few users surveyed is
users spoken to, a few users surveyed is a good way to validate that.
a good way to validate that. Okay, Amar is saying we use Pendo in our
Okay, Amar is saying we use Pendo in our organization for research. Yeah, Pendo
organization for research. Yeah, Pendo is a good tool. Uh we will use that tool
is a good tool. Uh we will use that tool as well uh in the future uh when we
as well uh in the future uh when we practice and get some hands-on. But
practice and get some hands-on. But essentially I mean at the end it's more
essentially I mean at the end it's more about speaking to users and interacting
about speaking to users and interacting with them. Uh that's the crux and of
with them. Uh that's the crux and of course there are several tools that help
course there are several tools that help you manage that research. We'll use
you manage that research. We'll use Pendo and we'll also look at Craftful
Pendo and we'll also look at Craftful which is another great tool to do user
which is another great tool to do user sentiment for product managers. So we'll
sentiment for product managers. So we'll look at both of them as we go along but
look at both of them as we go along but yeah that's definitely something that
yeah that's definitely something that can be used. Uh now Abdul has a very
can be used. Uh now Abdul has a very practical question in terms of saying
practical question in terms of saying that okay ideally this will take some
that okay ideally this will take some time which means that uh you know you
time which means that uh you know you have to convince stakeholders who are
have to convince stakeholders who are expecting workable solutions then
expecting workable solutions then spending on this validation. Now this is
spending on this validation. Now this is a very pertinent question and this is a
a very pertinent question and this is a very real question where uh you know
very real question where uh you know what happens is stakeholders especially
what happens is stakeholders especially stakeholders around business and
stakeholders around business and stakeholders around um um
stakeholders around um um u outcome oriented stakeholders don't
u outcome oriented stakeholders don't maybe not may not have enough time to
maybe not may not have enough time to for you to validate things and go ahead
for you to validate things and go ahead with I mean if you if your company has a
with I mean if you if your company has a good product culture validation is
good product culture validation is something that will have to be you know
something that will have to be you know part inte integrated part of that flow.
part inte integrated part of that flow. What you can do although is you can pick
What you can do although is you can pick and choose when you want to do a deep
and choose when you want to do a deep validation, right? like uh I think
validation, right? like uh I think someone had said earlier as well uh I I
someone had said earlier as well uh I I forgot the name but essentially what
forgot the name but essentially what what would work is if you can go ahead
what would work is if you can go ahead and skip this validation for smaller
and skip this validation for smaller features for features where you know the
features for features where you know the expectation is very straightforward it's
expectation is very straightforward it's great but for larger features what you
great but for larger features what you could do is you could take the time to
could do is you could take the time to kind of work on it and do a proper
kind of work on it and do a proper validation and again align stakeholders
validation and again align stakeholders keep them informed about what you're
keep them informed about what you're doing and that's probably a good way to
doing and that's probably a good way to do do it. So maybe a balancing out act
do do it. So maybe a balancing out act of you know picking and choosing when to
of you know picking and choosing when to do this and when to not might be a good
do this and when to not might be a good way to uh you know kind of do this.
way to uh you know kind of do this. Okay, we'll take some from uh you know
Okay, we'll take some from uh you know voices. Well yes Pashan go ahead.
Yeah. Hi. I wanted to ask if like this is a very thorough research we need to
is a very thorough research we need to do and correct. So as a we need to do
do and correct. So as a we need to do all these exercise or there's a
all these exercise or there's a dedicated team like UX research or other
dedicated team like UX research or other people will help you with
people will help you with >> so usually bigger companies have a UX
>> so usually bigger companies have a UX research team that works with you on
research team that works with you on this. Smaller companies don't have that.
this. Smaller companies don't have that. So smaller companies don't have a UX
So smaller companies don't have a UX team in general and uh you know they uh
team in general and uh you know they uh in there the PM is supposed to do it but
in there the PM is supposed to do it but slightly bigger companies especially
slightly bigger companies especially some with decent funding they always
some with decent funding they always have someone to uh you know work with
have someone to uh you know work with work this with um so yeah so that's how
work this with um so yeah so that's how it it it works usually if you're in a
it it it works usually if you're in a MNC or in at least even a tier one or
MNC or in at least even a tier one or even like a series C plus funded startup
even like a series C plus funded startup uh in India you would have someone who
uh in India you would have someone who would own research as Well um I mean
would own research as Well um I mean from my experience both in our company
from my experience both in our company we have uh like a couple of UX
we have uh like a couple of UX researchers but we spent a lot of time
researchers but we spent a lot of time with uh because of having common
with uh because of having common investors with the research teams of
investors with the research teams of Zumato and Urban Company and um I mean
Zumato and Urban Company and um I mean yeah there it's it's a very focused wing
yeah there it's it's a very focused wing where uh you know PMS come with
where uh you know PMS come with hypothesis and what they think is the
hypothesis and what they think is the thing that they should build and then
thing that they should build and then this research team goes ahead and
this research team goes ahead and implements something similar to what
implements something similar to what they we said of or they have their own
they we said of or they have their own templates but something similar to what
templates but something similar to what we discussed where they go to market and
we discussed where they go to market and actually speak to users understand their
actually speak to users understand their sentiment and decide if what the PM team
sentiment and decide if what the PM team has come up with uh is accurate or not.
has come up with uh is accurate or not. So so yeah that's that's how it usually
So so yeah that's that's how it usually works.
Uh but like in smaller companies you have less team but you need to ship fast
have less team but you need to ship fast and
and >> yeah so so that's that's why you would
>> yeah so so that's that's why you would have to you know get validation with
have to you know get validation with fewer people right you would you would
fewer people right you would you would not have the luxury to you know let's
not have the luxury to you know let's say at a bigger company you would have
say at a bigger company you would have the luxury to speak to 15 20 30 40
the luxury to speak to 15 20 30 40 people take your time to do it but at
people take your time to do it but at startups maybe just five people seven
startups maybe just five people seven people 10 people and you go ahead and
people 10 people and you go ahead and build it but the steps remain the same
build it but the steps remain the same it's just the scale reduces Um uh
it's just the scale reduces Um uh unfortunately especially with the amount
unfortunately especially with the amount of resources that it takes to build all
of resources that it takes to build all these features in terms of dev salaries
these features in terms of dev salaries and their time and all of that and the
and their time and all of that and the opportunity cost as well. Uh not doing
opportunity cost as well. Uh not doing opportunity validation is uh is
opportunity validation is uh is essentially a bigger risk,
essentially a bigger risk, >> right?
>> Thank you. >> Sure.
>> Sure. Uh yes, man. Go ahead.
Uh yes, man. Go ahead. Hi Mallay. So my question is about the
Hi Mallay. So my question is about the same that I asked on chat is about you
same that I asked on chat is about you know the research on I mean how can we
know the research on I mean how can we utilize the research of uh our
utilize the research of uh our competitor like and I I wanted to throw
competitor like and I I wanted to throw some some uh competitors uh feature that
some some uh competitors uh feature that has been available in one company and
has been available in one company and other company is planning to copy that.
other company is planning to copy that. So how the steps will be for you know
So how the steps will be for you know those specific company for example there
those specific company for example there are two big uh gifting companies Snappy
are two big uh gifting companies Snappy and IGP
and IGP >> and Snappy has introduced this feature
>> and Snappy has introduced this feature called gifting agent AI gifting agent
called gifting agent AI gifting agent okay and this IGP wanted to replicate
okay and this IGP wanted to replicate that that entire feature do they really
that that entire feature do they really required to do all these uh you know
required to do all these uh you know hypothesis validation and product
hypothesis validation and product discoveries uh sorry this feature
discoveries uh sorry this feature discoveries and all those thing or uh
discoveries and all those thing or uh and and at at what level I mean if they
and and at at what level I mean if they required at what level and I have other
required at what level and I have other two companies as well like uh Vista
two companies as well like uh Vista print and printo both both are in the
print and printo both both are in the print tech okay
print tech okay >> it so there are multiple options
>> it so there are multiple options examples there so I'll tell you I'll
examples there so I'll tell you I'll tell you a thumb rule which is very
tell you a thumb rule which is very practical and how it works as well if
practical and how it works as well if the audience base is extremely similar
the audience base is extremely similar I'm not talking about the product
I'm not talking about the product offering I'm talking about the audience
offering I'm talking about the audience base if that base is extremely similar
base if that base is extremely similar There is no harm in copying your
There is no harm in copying your competitor. I mean the biggest example
competitor. I mean the biggest example is the soundbox. Uh PTM launched the
is the soundbox. Uh PTM launched the soundbox for all the people and uh then
soundbox for all the people and uh then Phone Pay, Google Pay, everyone has
Phone Pay, Google Pay, everyone has their own soundbox. Even HDFC has their
their own soundbox. Even HDFC has their soundbox and there isn't like too much
soundbox and there isn't like too much thought there because the end customer
thought there because the end customer is the same. So the only time you
is the same. So the only time you require to do this kind of a thorough
require to do this kind of a thorough research is when your end customer and
research is when your end customer and the person in front or the competitor's
the person in front or the competitor's end customer is even you know slightly
end customer is even you know slightly different right for example I'll give
different right for example I'll give you an example right so for us uh for
you an example right so for us uh for example ours is a billing software right
example ours is a billing software right now our audience is um small MSMES
now our audience is um small MSMES small turnover not not extremely large
small turnover not not extremely large turnover um maybe between two up to two
turnover um maybe between two up to two crores of turnovers in a year kind of
crores of turnovers in a year kind of businesses. Now on the just one second
businesses. Now on the just one second let me complete. So on the face of it
let me complete. So on the face of it like Vapar which is the billing software
like Vapar which is the billing software our company and Tally are competitors
our company and Tally are competitors but Tally's audience is very different
but Tally's audience is very different which is more about higher earning
which is more about higher earning higher turnover 57 cr 10 cr turnover
higher turnover 57 cr 10 cr turnover people. So even though the segment is
people. So even though the segment is the same something that we build will
the same something that we build will just based on copying tally will not
just based on copying tally will not work because the end user is slightly
work because the end user is slightly different. So even if the end user
different. So even if the end user becomes slightly different it's it's
becomes slightly different it's it's it's not worth copying in situations
it's not worth copying in situations such as Uber Ola IGP and and and the
such as Uber Ola IGP and and and the gifting thing that you said PTM phone
gifting thing that you said PTM phone pay there the audience is exactly same I
pay there the audience is exactly same I mean the same person is using both apps
mean the same person is using both apps so that is why it's it's fairly fine to
so that is why it's it's fairly fine to kind of you know replicate in those
kind of you know replicate in those situations.
situations. >> Yeah. So I wanted to highlight this
>> Yeah. So I wanted to highlight this slack uh example that you had given
slack uh example that you had given thread right that can available in Slack
thread right that can available in Slack but Google chat or teams Microsoft teams
but Google chat or teams Microsoft teams wanted to introduce do they really
wanted to introduce do they really required to do this?
required to do this? >> Yes. Yes they do because because Slack's
>> Yes. Yes they do because because Slack's target audience is startups
target audience is startups Microsoft teams and Google uh chat or
Microsoft teams and Google uh chat or whoever their target audience is bigger
whoever their target audience is bigger enterprises. So the difference between
enterprises. So the difference between the two uh the target audience is there
the two uh the target audience is there which is why you know they have not
which is why you know they have not blindly copied it so far and just
blindly copied it so far and just created it which is why it exists
created it which is why it exists completely differently. So that's sort
completely differently. So that's sort of how that's sort of when the
of how that's sort of when the differentiation exists.
differentiation exists. >> Okay. Okay. Got it. Yeah.
>> Okay. Okay. Got it. Yeah. >> Yes. Cool. Thank you. Yes. S go ahead.
>> hello. >> Yeah, sorry about that. Yes, go ahead.
>> Yeah, sorry about that. Yes, go ahead. While uh you know selecting the user
While uh you know selecting the user groups for the validation of the of our
groups for the validation of the of our hypothesis,
hypothesis, >> do we need to select users or can we
>> do we need to select users or can we rely on web for the sol uh for the
rely on web for the sol uh for the hypothesis validation cuz sometimes
hypothesis validation cuz sometimes people even even if they use the uh uh
people even even if they use the uh uh even if they use the website or an app
even if they use the website or an app something something like that they won't
something something like that they won't be able to give out the suggestions for
be able to give out the suggestions for that or they might not be knowing the
that or they might not be knowing the problem. Exactly.
problem. Exactly. >> You're right. So u I did create like a
>> You're right. So u I did create like a user interview GPT which I will share
user interview GPT which I will share with you guys in the assignment of this
with you guys in the assignment of this session as well which you can use as a
session as well which you can use as a typical user. It's trained on you know
typical user. It's trained on you know what 10 15,000 of our uh conversations
what 10 15,000 of our uh conversations that we've had with different users uh
that we've had with different users uh in both my experiences. So it's trained
in both my experiences. So it's trained on that. So that's something that you
on that. So that's something that you can definitely use. I mean an AI tool is
can definitely use. I mean an AI tool is essentially what you can definitely use.
essentially what you can definitely use. But somehow I feel that even with this
But somehow I feel that even with this uh situation there is still scope of at
uh situation there is still scope of at least 30 40% of your users of you
least 30 40% of your users of you speaking to your users to generate
speaking to your users to generate insights because that really is a
insights because that really is a different feel that really is a
different feel that really is a different level of insights especially I
different level of insights especially I mean uh when you sit with users when you
mean uh when you sit with users when you work with them when you understand uh
work with them when you understand uh observe them when you understand their
observe them when you understand their day-to-day that gives you a lot more
day-to-day that gives you a lot more insights in general that's what I feel
insights in general that's what I feel so yes sometimes users will be
so yes sometimes users will be incoherent. They will say things that
incoherent. They will say things that they don't mean. They will say uh things
they don't mean. They will say uh things that are not essentially right. But even
that are not essentially right. But even with that, I think your job or your role
with that, I think your job or your role should still be to um you know try and
should still be to um you know try and find at least a few users who you can
find at least a few users who you can speak to and then of course AI is also
speak to and then of course AI is also very important now and you can train
very important now and you can train models. There are enough you know AI
models. There are enough you know AI tools that are available that can act
tools that are available that can act like real users and give you good
like real users and give you good insights. But yeah, still some portion
insights. But yeah, still some portion of it to be done by users. uh to be done
of it to be done by users. uh to be done by actual users should be there.
by actual users should be there. >> Okay. Then how do you go forward with
>> Okay. Then how do you go forward with selecting which user group to validate
selecting which user group to validate our hypothesis like is there any
our hypothesis like is there any particular set of people?
particular set of people? >> So like in the example I I showed right
>> So like in the example I I showed right you would create a target audience. So
you would create a target audience. So based on you know your hypothesis of who
based on you know your hypothesis of who your target audience is you would try
your target audience is you would try and find people who are closest to that.
and find people who are closest to that. So not all the people would fit in
So not all the people would fit in exactly but you would find people who
exactly but you would find people who are close enough to that and then try
are close enough to that and then try and interview them. I think that's
and interview them. I think that's Sujit's question as well on uh chat too
Sujit's question as well on uh chat too in terms of of course you try and you
in terms of of course you try and you know map it as close to your hypothesis
know map it as close to your hypothesis as as possible. But if sometimes it's
as as possible. But if sometimes it's not possible you pick them at random and
not possible you pick them at random and keep interviewing them there. But yes,
keep interviewing them there. But yes, usually what you would do is you would
usually what you would do is you would uh you know figure out the ones that map
uh you know figure out the ones that map closest to your thought process from
closest to your thought process from there.
there. Okay, got it. Thank you.
Okay, got it. Thank you. >> Right. Okay. Uh, Sud is saying conduct
>> Right. Okay. Uh, Sud is saying conduct interviews. Should we consider any
interviews. Should we consider any particular guidelines to pick people?
particular guidelines to pick people? So, like I said, the target audience is
So, like I said, the target audience is what you would do. And then connecting
what you would do. And then connecting with users is very interesting. I'll
with users is very interesting. I'll tell you two or three practical things
tell you two or three practical things that have worked with us. What worked
that have worked with us. What worked really well for us early on when I had
really well for us early on when I had start just started uh uh you know
start just started uh uh you know leading the team like four five years
leading the team like four five years back. What had worked really well was
back. What had worked really well was the promise of a reward. So we did
the promise of a reward. So we did something where uh you know users could
something where uh you know users could express interest to speak to us and if
express interest to speak to us and if they spoke to us they stand a chance to
they spoke to us they stand a chance to win a gift card right and this was the
win a gift card right and this was the time when uh I don't know if you people
time when uh I don't know if you people remember or not but Google pay had
remember or not but Google pay had introduced scratch cards and they were
introduced scratch cards and they were very viral at that time people everyone
very viral at that time people everyone wanted to have a scratch card because it
wanted to have a scratch card because it used to give you a lot of good rewards.
used to give you a lot of good rewards. So that's what we did where we said that
So that's what we did where we said that okay if you give us an interview uh and
okay if you give us an interview uh and we used to show it to our target
we used to show it to our target audience the one that was there the
audience the one that was there the people who we really wanted the kind of
people who we really wanted the kind of people let's say who created 500
people let's say who created 500 invoices who created 600 invoices who
invoices who created 600 invoices who were let's say retailers or who were
were let's say retailers or who were wholesalers groceries they we used to
wholesalers groceries they we used to show them bigger rewards. So we used to
show them bigger rewards. So we used to show them that you know you can win up
show them that you know you can win up to 500 rupees on a scratch card and so
to 500 rupees on a scratch card and so on and so forth. We used to have a
on and so forth. We used to have a budget for it which is we used to say
budget for it which is we used to say that okay in total we'll spend this much
that okay in total we'll spend this much the rest we will just you know uh
the rest we will just you know uh essentially uh show better luck next
essentially uh show better luck next time. Uh that really worked very well
time. Uh that really worked very well and especially when you're early it
and especially when you're early it works very well once you are a bigger
works very well once you are a bigger company once you're an MNC users
company once you're an MNC users automatically feel a level of prestige
automatically feel a level of prestige interacting with you and then they would
interacting with you and then they would be you know more okay to you know speak
be you know more okay to you know speak to you and go ahead with it. But
to you and go ahead with it. But especially when your product or when
especially when your product or when your feature is early, incentivizing
your feature is early, incentivizing them. Not even incentivizing them
them. Not even incentivizing them actually giving them the hope that they
actually giving them the hope that they will get incentivized is a great way to
will get incentivized is a great way to you know get a lot of seriousness out of
you know get a lot of seriousness out of users and the users to speak to you
users and the users to speak to you properly. Uh so that really worked very
properly. Uh so that really worked very well for us. We ran it for like four or
well for us. We ran it for like four or five years um consistently and it used
five years um consistently and it used to give us like uh very strong inflow of
to give us like uh very strong inflow of people who wanted to speak to us. Now we
people who wanted to speak to us. Now we don't run it because now uh it's a
don't run it because now uh it's a situation where uh you know there is
situation where uh you know there is enough brand in the market of the
enough brand in the market of the product so that when we go and talk
product so that when we go and talk about the product people are
about the product people are automatically ready to talk about it but
automatically ready to talk about it but early on incentivizing them with this
early on incentivizing them with this gamification really helped uh for us. If
gamification really helped uh for us. If you just give blanket money, it might
you just give blanket money, it might not they might not, you know, have the
not they might not, you know, have the real motivation. They're just there to
real motivation. They're just there to take the money. But if you like make
take the money. But if you like make them do some effort, give them a scratch
them do some effort, give them a scratch card, disappoint them a few times and so
card, disappoint them a few times and so on. Then, you know, you actually think
on. Then, you know, you actually think that okay, people who are actually
that okay, people who are actually interested are coming. Uh, so that's how
interested are coming. Uh, so that's how you should do it.
you should do it. Okay. Rashmi is saying, could you
Okay. Rashmi is saying, could you explain how field interview candidates
explain how field interview candidates are searched and selected by PMS? And
are searched and selected by PMS? And so, uh, once again, it's it's the pretty
so, uh, once again, it's it's the pretty it's pretty similar. So most of the
it's pretty similar. So most of the products they track um uh locations
products they track um uh locations right uh most of the products track the
right uh most of the products track the locations which is why you know you
locations which is why you know you would know where your users are based
would know where your users are based out of plus most of the applications
out of plus most of the applications would also know some of their most uh uh
would also know some of their most uh uh you know um common hotspots. For
you know um common hotspots. For example, for us like a market in
example, for us like a market in Bangalore is a very small hot strong hot
Bangalore is a very small hot strong hot spot. For a company like Uber or Ola
spot. For a company like Uber or Ola like a big tech park in Bangalore or
like a big tech park in Bangalore or Pune or anywhere would be a big uh you
Pune or anywhere would be a big uh you know hot spot because they would know
know hot spot because they would know that a lot of people exiting are their
that a lot of people exiting are their users. Similarly for Zomato also that
users. Similarly for Zomato also that would be a big hot spot. So usually the
would be a big hot spot. So usually the candidates are selected based on this.
candidates are selected based on this. There are also agencies that will help
There are also agencies that will help you do this, but that's essentially you
you do this, but that's essentially you and the manager would not usually
and the manager would not usually confirm it because it will become too
confirm it because it will become too much of a micromanagement if they're
much of a micromanagement if they're confirming who you are meeting. But it
confirming who you are meeting. But it really depends on the kind of manager
really depends on the kind of manager also to be honest. There may be some
also to be honest. There may be some managers who may be a little too um uh
managers who may be a little too um uh involved and too picky to kind of go
involved and too picky to kind of go ahead and ask you to uh talk about it.
ahead and ask you to uh talk about it. But that that's that's most managers
But that that's that's most managers wouldn't.
wouldn't. Anab is saying Sugi and Zumato also
Anab is saying Sugi and Zumato also exactly same audience. Yes, absolutely.
exactly same audience. Yes, absolutely. Sugi and Zomato are exactly same which
Sugi and Zomato are exactly same which is why you see a lot of things that Sugi
is why you see a lot of things that Sugi does, Zomato does and then a lot of
does, Zomato does and then a lot of things that Zomato does, Swiggi does
things that Zomato does, Swiggi does because exactly same audience which is
because exactly same audience which is why there is no difference in their
why there is no difference in their behavior. So you can build on properly.
behavior. So you can build on properly. So saying how to do user research for a
So saying how to do user research for a B2B product. So uh for a B2D product
B2B product. So uh for a B2D product also there is a very similar pattern to
also there is a very similar pattern to what we uh kind of discussed uh in terms
what we uh kind of discussed uh in terms of uh uh you know going to users and
of uh uh you know going to users and looking at it. My personal favorite for
looking at it. My personal favorite for a B2B product is actually a survey. B2B
a B2B product is actually a survey. B2B users respond to surveys very well. Uh
users respond to surveys very well. Uh surprisingly well. They write very
surprisingly well. They write very detailed answers to surveys. So I always
detailed answers to surveys. So I always uh you know start with a survey to my
uh you know start with a survey to my users both the end user and also the
users both the end user and also the decision makers get some insights from
decision makers get some insights from there and then of course set up calls
there and then of course set up calls with them with my hypothesis and do the
with them with my hypothesis and do the same peak validation and all of those
same peak validation and all of those things that we just discussed in the
things that we just discussed in the questions.
Okay Mo is saying what's your preferred evidence mix for a go decision here
evidence mix for a go decision here percentage of users reporting severity
percentage of users reporting severity observed behavior change or projected
observed behavior change or projected revenue impact. Got it? So I am always,
revenue impact. Got it? So I am always, you know, very pro users. So I'm always
you know, very pro users. So I'm always very pro getting user validation for my
very pro getting user validation for my ideas and for my features. I personally
ideas and for my features. I personally feel revenue and u even behavior those
feel revenue and u even behavior those are things that happen over time. I mean
are things that happen over time. I mean there have been several features that
there have been several features that we've built because our users wanted it.
we've built because our users wanted it. It didn't really show us revenue
It didn't really show us revenue immediately because we didn't focus on
immediately because we didn't focus on revenue. Revenue was a non- goal for us
revenue. Revenue was a non- goal for us to begin with. But then now or of late
to begin with. But then now or of late when we've started charging as
when we've started charging as additional revenue for those features,
additional revenue for those features, those features sell like hot cakes along
those features sell like hot cakes along with the license that we have. More than
with the license that we have. More than like 30 35% of our users buy those
like 30 35% of our users buy those features on top of the license that they
features on top of the license that they buy. So I am you know extremely pro the
buy. So I am you know extremely pro the traditional method of PMing which is uh
traditional method of PMing which is uh you know you build for users you solve
you know you build for users you solve for your user problems and then the rest
for your user problems and then the rest align automatically. So that's why user
align automatically. So that's why user severity and user validation is usually
severity and user validation is usually my go-to to take a decision and because
my go-to to take a decision and because if that is there everything else kind of
if that is there everything else kind of falls in place uh usually
right okay I know we are over time but I'll just take 10 more minutes 1 minute
I'll just take 10 more minutes 1 minute let me just set that
Once you've completed the user value, the harder part is done. Uh now your job
the harder part is done. Uh now your job is to find the user value. Business
is to find the user value. Business value. Now a business value is where you
value. Now a business value is where you decide or you design something that
decide or you design something that we'll call as the business funnel.
we'll call as the business funnel. Right. Now, what is the business funnel?
Right. Now, what is the business funnel? A business funnel is nothing but
A business funnel is nothing but a stepbystep breakdown of the problem
a stepbystep breakdown of the problem that you have
that you have with the end point being a desired
with the end point being a desired business outcome.
business outcome. Right? So, for example, in the reviews
Right? So, for example, in the reviews example, the end point that we had was
example, the end point that we had was the order being placed.
the order being placed. Order placed without the re uh without
Order placed without the re uh without cancellation,
right? That is how it is. So what you end up creating to refine the business
end up creating to refine the business value that you have is you end up
value that you have is you end up building a business funnel where the end
building a business funnel where the end point is the most important. Right? And
point is the most important. Right? And now what you do is you simply trace the
now what you do is you simply trace the user's journey back from that end point
user's journey back from that end point to the entire thing. So when we were
to the entire thing. So when we were doing cancellation for Uber drivers,
doing cancellation for Uber drivers, your final point would be of course
your final point would be of course completing the ride. What is the step
completing the ride. What is the step before that? Picking up the passenger,
before that? Picking up the passenger, step before that viewing the
step before that viewing the destination, step before that accepting
destination, step before that accepting the ride, receiving request, starting a
the ride, receiving request, starting a driving session. Similarly, if you were
driving session. Similarly, if you were uh I'll just make a copy of this as well
uh I'll just make a copy of this as well so that I can edit it there. So when
so that I can edit it there. So when when you also get this, you will get a
when you also get this, you will get a uh a non-editable template. So what you
uh a non-editable template. So what you can do is just create a copy of it like
can do is just create a copy of it like how I'm doing right now and uh then you
how I'm doing right now and uh then you will be able to edit it as your own and
will be able to edit it as your own and attach it to sculp as well. So for
attach it to sculp as well. So for example for the Amazon one what I'm
example for the Amazon one what I'm going to do is uh the end point is order
going to do is uh the end point is order completed without refund.
completed without refund. The step four would be order received.
The step four would be order received. Step three would be completed payment
Step three would be completed payment for the order.
Uh just before paying I would add to my cart. Add to cart after evaluation.
Then step one would be look at product details page for uh
look at product details page for uh understanding
or actually sorry look at product reviews for deciding
look at product reviews for deciding whether this is something they want to
whether this is something they want to buy
and then finally looking at the products listing on
looking at the products listing on Amazon via search or explode.
Amazon via search or explode. Right? So what I've done is I've now
Right? So what I've done is I've now created a funnel. My outcome was my
created a funnel. My outcome was my business outcome which I was targeting
business outcome which I was targeting if you remember in the hypothesis where
if you remember in the hypothesis where my target is to complete orders without
my target is to complete orders without refund. Now when I say order received I
refund. Now when I say order received I should actually just clarify it. Order
should actually just clarify it. Order received by customer right. Um
received by customer right. Um then completed payment for the order uh
then completed payment for the order uh is done by the customer. So they or we
is done by the customer. So they or we can just simplify this and say placed
can just simplify this and say placed order and then we can simplify this and
order and then we can simplify this and say
say received order and this is refund
received order and this is refund uh
right um we can just simplify this as well we can just put it as act of cart
well we can just put it as act of cart uh we can call this as review product.
uh we can call this as review product. So all I'm doing is I'm just trying to
So all I'm doing is I'm just trying to give it easier names for you to
give it easier names for you to understand. Why I wrote down the full
understand. Why I wrote down the full ones is for you to understand my thought
ones is for you to understand my thought process of how I was coming to this and
process of how I was coming to this and um
finally landing on listing page. Right? So landing on the listing page,
So landing on the listing page, reviewing the product, adding to cart,
reviewing the product, adding to cart, placing an order, receiving the order
placing an order, receiving the order and not is showing a cancellation are
and not is showing a cancellation are the five points that you would look at.
the five points that you would look at. Right now once you have created this
Right now once you have created this business funnel uh essentially what I've
business funnel uh essentially what I've done is I've just created a user's
done is I've just created a user's journey. I've created six steps here but
journey. I've created six steps here but there can be more as well based on how
there can be more as well based on how long your product is. But try and you
long your product is. But try and you know keep relevant stuff because for
know keep relevant stuff because for example I could have tried to start with
example I could have tried to start with you know coming on the homepage of
you know coming on the homepage of Amazon searching and all of that. But
Amazon searching and all of that. But those are things that I'm not worried
those are things that I'm not worried about because at this point I'm just
about because at this point I'm just talking about cancellations. Similarly
talking about cancellations. Similarly for slack this the the outcome that we
for slack this the the outcome that we are going to look at is
are going to look at is the thread initiated right and I'm then
the thread initiated right and I'm then going to backwards calculate all of
going to backwards calculate all of this. Now what I'm going to do is I'm
this. Now what I'm going to do is I'm going to put in actual numbers over a
going to put in actual numbers over a certain time frame for each of these
certain time frame for each of these pointers. So landing on listing page,
pointers. So landing on listing page, reviewing the product, adding to cart,
reviewing the product, adding to cart, placing order, receiving the order, and
placing order, receiving the order, and no cancellation issued. I'm going to now
no cancellation issued. I'm going to now talk about the entire funnel of the the
talk about the entire funnel of the the the funnel that business funnel I've
the funnel that business funnel I've created and put in the numbers here.
created and put in the numbers here. Right? So here, as you can see, 1,000
Right? So here, as you can see, 1,000 people landed on just these are sign
people landed on just these are sign like indicative numbers. So 1,000 people
like indicative numbers. So 1,000 people landed on the listing page in Jan. 900
landed on the listing page in Jan. 900 reviewed a product. When I say reviewed,
reviewed a product. When I say reviewed, went on the listing page and saw the
went on the listing page and saw the reviews. 850 added to cart, 820 placed,
reviews. 850 added to cart, 820 placed, 780 received and there was no
780 received and there was no cancellation issue issue,
cancellation issue issue, right? Um I can I can fill in for
right? Um I can I can fill in for several months and then I can take an
several months and then I can take an average of this. This is my baseline
average of this. This is my baseline average and then go ahead and put my
average and then go ahead and put my percentages of each. Right? when I go
percentages of each. Right? when I go ahead and put percentages of each I can
ahead and put percentages of each I can put you know the 100% 86 82 79 75 75 and
put you know the 100% 86 82 79 75 75 and this gives me an understanding of where
this gives me an understanding of where there is the maximum drop off now what
there is the maximum drop off now what is a drop off drop off is nothing but
is a drop off drop off is nothing but someone going from a previous step to
someone going from a previous step to the last step someone dropping off there
the last step someone dropping off there right so what is my drop off rate is
right so what is my drop off rate is what I'm going to mention here where I'm
what I'm going to mention here where I'm going to mention that okay what
going to mention that okay what percentage of users are dropping off
percentage of users are dropping off there.
Right? This will give me an understanding of all the things that are
understanding of all the things that are happening here. Right? This is give this
happening here. Right? This is give this will give me an understanding of uh all
will give me an understanding of uh all the things that are that are uh that
the things that are that are uh that that are there. For example, in the
that are there. For example, in the first step there is a 14% drop off. In
first step there is a 14% drop off. In the second step there is five, there is
the second step there is five, there is three, there is four, there is zero. And
three, there is four, there is zero. And based on this you decide or based on
based on this you decide or based on this you know you figure out the value
this you know you figure out the value of your product. So for example in this
of your product. So for example in this case since we are trying to improve
case since we are trying to improve revenue what we are trying to we are
revenue what we are trying to we are trying to improve uh reviews rather what
trying to improve uh reviews rather what we are trying to do is we are trying to
we are trying to do is we are trying to solve this funnel drop off
solve this funnel drop off right there is a 5% drop off here right
right there is a 5% drop off here right second because we are trying to do this
second because we are trying to do this we are also trying to solve this percent
we are also trying to solve this percent drop off. Now the numbers here are a
drop off. Now the numbers here are a little weird. So there's only like a 0%
little weird. So there's only like a 0% drop off here. But actually these
drop off here. But actually these numbers would be slightly different. Or
numbers would be slightly different. Or maybe I should just change these a bit
maybe I should just change these a bit as well. Um yeah, for now this is fine.
as well. Um yeah, for now this is fine. So this is a 2% drop off. So now based
So this is a 2% drop off. So now based on the entire users's business journey
on the entire users's business journey or business funnel or whatever you want
or business funnel or whatever you want to call it, you will find which are the
to call it, you will find which are the pointers that your area can work on.
pointers that your area can work on. Right? Of course you you are not
Right? Of course you you are not building something that will help them
building something that will help them place more orders. You're not building
place more orders. You're not building something that will improve logistics.
something that will improve logistics. So these are not part of your scope.
So these are not part of your scope. You're not building something that helps
You're not building something that helps them explore more products. So you're
them explore more products. So you're building something that helps them look
building something that helps them look at the product and add to cart and also
at the product and add to cart and also helps them look at a look uh not cancel
helps them look at a look uh not cancel once they have received a product. So
once they have received a product. So those are the two journeys that you're
those are the two journeys that you're going to look at and then based on that
going to look at and then based on that you're going to you know kind of talk
you're going to you know kind of talk about the estimated business value. Now
about the estimated business value. Now here again you don't you cannot say uh
here again you don't you cannot say uh generic stuff like you were saying in
generic stuff like you were saying in the manager briefing business value
the manager briefing business value where you were just saying improving
where you were just saying improving quality and so on. Here you have to give
quality and so on. Here you have to give a proxy metric. So for example, if I
a proxy metric. So for example, if I have a 5% drop off here and a 2% drop
have a 5% drop off here and a 2% drop off here, I'm going to say that okay,
off here, I'm going to say that okay, there is a total of you know 6% or 600
there is a total of you know 6% or 600 basis points that I can uh you know
basis points that I can uh you know improve or 6 percentage points that I
improve or 6 percentage points that I can improve uh in this and there I would
can improve uh in this and there I would apply a discount of course because I
apply a discount of course because I would not be able to you know improve
would not be able to you know improve all the 6% of course uh it's going to
all the 6% of course uh it's going to take some time. So that is why they're
take some time. So that is why they're going to I'm going to focus on you know
going to I'm going to focus on you know just one of them uh and say that okay u
just one of them uh and say that okay u out of this six percentage points
out of this six percentage points available for me across two different
available for me across two different funnels I can apply like a 30% buffer
funnels I can apply like a 30% buffer and then go and say that hey uh based on
and then go and say that hey uh based on this buffer I'm going to only uh you
this buffer I'm going to only uh you know challenge and only find these uh
know challenge and only find these uh four things or these five things or
four things or these five things or things like that right so that is how
things like that right so that is how basically you can go ahead and create a
basically you can go ahead and create a very clear understanding of your
very clear understanding of your business value and then estimated
business value and then estimated business impact is what is the reduction
business impact is what is the reduction that you're causing in the uh in this
that you're causing in the uh in this case your increment in rights but in the
case your increment in rights but in the other case the reduction in the
other case the reduction in the cancellations and so on and so forth
cancellations and so on and so forth right so essentially you begin with
right so essentially you begin with creating an entire funnel so you start
creating an entire funnel so you start with the end point first where you'd
with the end point first where you'd look at the business action that the
look at the business action that the user is doing and that's relevant to
user is doing and that's relevant to what your uh thought process was and
what your uh thought process was and then step by step you break it down to
then step by step you break it down to the entire user's journey and then find
the entire user's journey and then find out which are the parts that you are
out which are the parts that you are going to be working on based on your
going to be working on based on your hypothesis and then based on that you
hypothesis and then based on that you start uh you create a estimated uh
start uh you create a estimated uh business value of the project or
business value of the project or estimated business value of the product
estimated business value of the product right all of this uh once done I'll just
right all of this uh once done I'll just attach this to sculp again so that uh we
attach this to sculp again so that uh we have this here so this is Uber business
have this here so this is Uber business value and save it and complete this
value and save it and complete this step. Once all of this is done, your job
step. Once all of this is done, your job is now to put it into a nice
is now to put it into a nice presentation and showcase it to
presentation and showcase it to leadership, engineering and so on. And
leadership, engineering and so on. And that's a presentation template that I
that's a presentation template that I have just given here. You can use this
have just given here. You can use this template very easily. Use a value uh
template very easily. Use a value uh business value, what are some of the
business value, what are some of the constraints, what is a summary and then
constraints, what is a summary and then a rough timeline. This is more of like a
a rough timeline. This is more of like a presentation of sorts for uh presenting
presentation of sorts for uh presenting it. The main work is these three steps.
it. The main work is these three steps. And once you have this, now you have a
And once you have this, now you have a proper business value document that is
proper business value document that is available to you to use. That is
available to you to use. That is something that you can use to um you
something that you can use to um you know, once again go back to your manager
know, once again go back to your manager and tell them this is what my estimation
and tell them this is what my estimation is. Uh get buy in from leadership based
is. Uh get buy in from leadership based on you know the amount that you are
on you know the amount that you are expecting this to do in business value
expecting this to do in business value and create a lot of value for your
and create a lot of value for your company from them.
company from them. Right? We'll take some questions now. Uh
Right? We'll take some questions now. Uh I hope you've understood. So any
I hope you've understood. So any questions on anything that we've
questions on anything that we've discussed today? So we started with uh
discussed today? So we started with uh discussing
discussing the manager briefing. We talked about
the manager briefing. We talked about how you're going to do that. Second, we
how you're going to do that. Second, we went ahead and uh within the briefing
went ahead and uh within the briefing talked about strategic fit, user
talked about strategic fit, user hypothesis and business hypothesis. Then
hypothesis and business hypothesis. Then we went ahead and talked about
we went ahead and talked about validating that user value either
validating that user value either through data or through surveys, through
through data or through surveys, through field of reviews, through virtual
field of reviews, through virtual interviews. Um and then we finally
interviews. Um and then we finally talked about the business funnel where
talked about the business funnel where how do you create the business funnel to
how do you create the business funnel to present it properly.
for a new feature we don't have the data to validate the user funnel. How can we
to validate the user funnel. How can we achieve this funnel analysis for new
achieve this funnel analysis for new features? See even though you have a new
features? See even though you have a new feature the end goal will be there
feature the end goal will be there right? So for example if you're working
right? So for example if you're working on Uber or if you when you're doing a
on Uber or if you when you're doing a new product this is valid and if you're
new product this is valid and if you're doing a new product then of course doing
doing a new product then of course doing all of this doesn't make sense you have
all of this doesn't make sense you have to first have a product and then build
to first have a product and then build features but if you're doing a new
features but if you're doing a new feature the end goal will be there right
feature the end goal will be there right like for example for Uber no matter what
like for example for Uber no matter what feature you do you want to contribute to
feature you do you want to contribute to the total rides eventually or for Amazon
the total rides eventually or for Amazon you want to contribute to the total
you want to contribute to the total orders placed eventually. So even if you
orders placed eventually. So even if you know you have a new feature that doesn't
know you have a new feature that doesn't exist but the funnel of the user will
exist but the funnel of the user will still exist right so you will still have
still exist right so you will still have some data to see what the funnel looks
some data to see what the funnel looks like maybe you wouldn't get like as
like maybe you wouldn't get like as solid validation from a business sense
solid validation from a business sense because you might not have 100% clarity
because you might not have 100% clarity there but still you would have enough u
there but still you would have enough u you know to go ahead and take a call
you know to go ahead and take a call based on that. Oh Abdul had raised her
based on that. Oh Abdul had raised her hand. Yes Abdul go ahead.
hand. Yes Abdul go ahead. >> Yeah uh thanks Mallay. Um okay so two
>> Yeah uh thanks Mallay. Um okay so two questions I have Mallay uh one is u can
questions I have Mallay uh one is u can this manager briefing uh uh topic uh
this manager briefing uh uh topic uh also and uh the validation of user uh
also and uh the validation of user uh value plus business value go together
value plus business value go together hand in hand.
hand in hand. >> So manager briefing is the first step
>> So manager briefing is the first step where you get your manager to align that
where you get your manager to align that okay this is something that I will work
okay this is something that I will work on. The next two steps is where you do
on. The next two steps is where you do the actual PMing of going and speaking
the actual PMing of going and speaking to users of going and analyzing data and
to users of going and analyzing data and looking at numbers and doing some
looking at numbers and doing some validation on it. So manager briefing
validation on it. So manager briefing you can just say that manager briefing
you can just say that manager briefing is actually getting the permission of
is actually getting the permission of your manager that okay hey can I can I
your manager that okay hey can I can I uh pursue this opportunity right and
uh pursue this opportunity right and then once the manager is aligned the
then once the manager is aligned the next two steps are more you know focused
next two steps are more you know focused PMing steps uh from there so that's how
PMing steps uh from there so that's how you should uh you know envision this
you should uh you know envision this >> okay this this is even before uh the
>> okay this this is even before uh the design and execution phase
design and execution phase >> correct correct so after this the next
>> correct correct so after this the next session what we will learn is design so
session what we will learn is design so in the next session we'll talk about you
in the next session we'll talk about you know how do you design it and then talk
know how do you design it and then talk about development and then talk about
about development and then talk about launch and then talk about measurement.
launch and then talk about measurement. So right now we are just at the phase
So right now we are just at the phase where we are deciding whether this is
where we are deciding whether this is worth doing or not even though this
worth doing or not even though this problem has come to us externally or
problem has come to us externally or maybe we have thought of it or whatever.
maybe we have thought of it or whatever. >> Absolutely. And we'll come to the final
>> Absolutely. And we'll come to the final part right business value validation
part right business value validation right
right >> um so once we derive user values u and
>> um so once we derive user values u and we go to business value validation uh
we go to business value validation uh and I I I was little skeptical right for
and I I I was little skeptical right for example we have implemented a new
example we have implemented a new feature called knowledge uh sharing for
feature called knowledge uh sharing for the client base for the client users
the client base for the client users >> uh so I I I high level did you know a
>> uh so I I I high level did you know a strategic alignment and the user value
strategic alignment and the user value business value with my uh perception uh
business value with my uh perception uh but I couldn't do the validation what
but I couldn't do the validation what you just shown through funnel right how
you just shown through funnel right how many users uh will come into first step
many users uh will come into first step and you know how many will be uh you
and you know how many will be uh you know uh will will will buy the product
know uh will will will buy the product uh uh at the step four right uh because
uh uh at the step four right uh because I don't anticipate this new feature uh
I don't anticipate this new feature uh is going to align with that funnel
is going to align with that funnel analysis because I don't even have a
analysis because I don't even have a previous data to show okay this is going
previous data to show okay this is going to be my business value if at all I am
to be my business value if at all I am going to add this feature to the
going to add this feature to the business is completely new business.
business is completely new business. He's not even Yeah. Huh. No, I
He's not even Yeah. Huh. No, I understand. See, if it's a completely
understand. See, if it's a completely new business line, if it's a completely
new business line, if it's a completely new product, then estimating the
new product, then estimating the business value doesn't make sense.
business value doesn't make sense. Absolutely. But for most of the time
Absolutely. But for most of the time when you're working as a PM, you will be
when you're working as a PM, you will be working on existing products that
working on existing products that already have certain things done, right?
already have certain things done, right? And then you would work on top of it. So
And then you would work on top of it. So that's how you would create that
that's how you would create that understanding from there. So in top
understanding from there. So in top products you would be working on
products you would be working on features where you would have this
features where you would have this context of all of this data and this
context of all of this data and this background of funnel and everything
background of funnel and everything available. So that is where this is more
available. So that is where this is more important. Of course when you're working
important. Of course when you're working at a startup when you're doing V1 of
at a startup when you're doing V1 of products MVP of products of course
products MVP of products of course doesn't it makes total sense to uh you
doesn't it makes total sense to uh you know not worry about all of this and
know not worry about all of this and first ship and let your first set of
first ship and let your first set of users come in. So so yeah that's that's
users come in. So so yeah that's that's absolutely true. Then then how can I
absolutely true. Then then how can I validate the business value?
validate the business value? >> So it is very difficult to validate the
>> So it is very difficult to validate the business value. You would have to rely
business value. You would have to rely on the user value initially. You can of
on the user value initially. You can of course look at competitors. You can of
course look at competitors. You can of course do surveys with users and
course do surveys with users and understand if they would pay or not. But
understand if they would pay or not. But to to properly get business value you
to to properly get business value you would need some numbers because
would need some numbers because otherwise it would be mostly a
otherwise it would be mostly a hypothesis only. So in thus such cases
hypothesis only. So in thus such cases what you do is you do a user value
what you do is you do a user value hypothesis validation and then you take
hypothesis validation and then you take a step back and you know the business
a step back and you know the business hypothesis is what you go ahead with
hypothesis is what you go ahead with apart from that. So that's how you uh
apart from that. So that's how you uh you know kind of build it out.
you know kind of build it out. >> Okay. Can I can I can I uh as part of
>> Okay. Can I can I can I uh as part of business value can I mention uh
business value can I mention uh something like you know converting uh I
something like you know converting uh I mean a monetization benefit right
mean a monetization benefit right selling selling uh uh the data via
selling selling uh uh the data via subscription for example. Yeah, of
subscription for example. Yeah, of course, of course you can. Of course you
course, of course you can. Of course you can. But you should have a logical
can. But you should have a logical funnel to it, right? So like you said,
funnel to it, right? So like you said, right, you should not have a feature
right, you should not have a feature that's not even remotely connected to it
that's not even remotely connected to it and you just give a hypothesis that this
and you just give a hypothesis that this is how it will work. You should have
is how it will work. You should have some clarity of okay, if someone does
some clarity of okay, if someone does this, then this door opens, then this
this, then this door opens, then this door opens because of that.
door opens because of that. >> So that clarity that funnel that I
>> So that clarity that funnel that I showed should be there in some some or
showed should be there in some some or the other form.
the other form. >> Okay. Okay. Understood. Understood.
>> Okay. Okay. Understood. Understood. Yeah. Sure. Makes sense.
Yeah. Sure. Makes sense. Makes sense. Okay.
Makes sense. Okay. Okay. Kumi is saying, "Shall we get?"
Okay. Kumi is saying, "Shall we get?" Yes, of course. Okay. Thanks. Thanks,
Yes, of course. Okay. Thanks. Thanks, Sujit for answering that. Um, yes, the
Sujit for answering that. Um, yes, the recording of this will be uploaded. Um,
recording of this will be uploaded. Um, okay. Rash Rashmi is asking if a new
okay. Rash Rashmi is asking if a new feature gets cancelled further in the
feature gets cancelled further in the product life cycle because it's deemed
product life cycle because it's deemed not necessary.
not necessary. Uh, because of market changes, priority,
Uh, because of market changes, priority, etc. One, does this happen? Yes, of
etc. One, does this happen? Yes, of course, it does happen, but very rarely.
course, it does happen, but very rarely. If it's gone past the validation cycle,
If it's gone past the validation cycle, product life cycle, very rarely have I
product life cycle, very rarely have I seen features get cancelled because
seen features get cancelled because market changes don't happen immediately.
market changes don't happen immediately. Yes, bandwidth is a big challenge. You
Yes, bandwidth is a big challenge. You may not get a lot of bandwidth from
may not get a lot of bandwidth from developers but uh then also it will
developers but uh then also it will still be in the backlog. It will not get
still be in the backlog. It will not get cancelled. So after the validation
cancelled. So after the validation getting canled is not that common in my
getting canled is not that common in my opinion. Right. Um the second is how
opinion. Right. Um the second is how situation h handled. I suppose I trade.
situation h handled. I suppose I trade. Yes. If at the hypothesis stage you find
Yes. If at the hypothesis stage you find towards the later half that okay yeah
towards the later half that okay yeah this is fine. I mean the users show
this is fine. I mean the users show value but business may there is no value
value but business may there is no value or there is business value but users are
or there is business value but users are not really showing interest in it. Then
not really showing interest in it. Then of course you reiterate and your
of course you reiterate and your original hypothesis was incorrect. You
original hypothesis was incorrect. You tweak that hypothesis based on new
tweak that hypothesis based on new learnings like we saw in that template
learnings like we saw in that template of saying that okay this is a new
of saying that okay this is a new finding and then do that entire thing
finding and then do that entire thing again. But usually after validation very
again. But usually after validation very rarely does the feature get cancelled.
rarely does the feature get cancelled. Okay. Mohit is saying if our funnel
Okay. Mohit is saying if our funnel shows multiple drop offs um it depends
shows multiple drop offs um it depends Mohit because you would have because not
Mohit because you would have because not every drop drop off would have a
every drop drop off would have a hypothesis right. So for example, you
hypothesis right. So for example, you may have a big drop off, but you always
may have a big drop off, but you always may or may not be able to have a
may or may not be able to have a hypothesis that supports that drop off
hypothesis that supports that drop off because some are natural drop offs,
because some are natural drop offs, right? For example, payment step is
right? For example, payment step is always a natural drop off, right? If you
always a natural drop off, right? If you have people adding to cart, 100 people
have people adding to cart, 100 people add to cart, 50 will always drop off. So
add to cart, 50 will always drop off. So you always have a situation where there
you always have a situation where there will be drop offs everywhere. So it is
will be drop offs everywhere. So it is not necessary that you always and always
not necessary that you always and always have this. You may be in a situation
have this. You may be in a situation where you don't have this as well. So
where you don't have this as well. So that's how you should look at it. You
that's how you should look at it. You may not have uh the ability to work on
may not have uh the ability to work on the highest drop off because you may not
the highest drop off because you may not have enough hypothesis there. But yes,
have enough hypothesis there. But yes, of course, if you have few hypotheses on
of course, if you have few hypotheses on the largest drop off, absolutely it
the largest drop off, absolutely it makes sense to focus on that because
makes sense to focus on that because this gives you the highest uh leverage
this gives you the highest uh leverage in terms of the impact that you will
in terms of the impact that you will drive.
drive. The second is if two steps in the funnel
The second is if two steps in the funnel look attractive, should we optimize for
look attractive, should we optimize for the highest marginal ROI per week for
the highest marginal ROI per week for the step that unlock? I mean it makes
the step that unlock? I mean it makes sense to to do something like that. But
sense to to do something like that. But again, it really depends on the
again, it really depends on the hypothesis that you have. If you have
hypothesis that you have. If you have hypothesis across different steps, it's
hypothesis across different steps, it's great. If you have the ability to create
great. If you have the ability to create hypothesis across different steps, you
hypothesis across different steps, you should definitely focus on this. But
should definitely focus on this. But usually in my op you know understanding
usually in my op you know understanding after a while generating these
after a while generating these hypothesis is the biggest blocker right
hypothesis is the biggest blocker right you you see a funnel drop off you see
you you see a funnel drop off you see 20% drop off here you've done a few
20% drop off here you've done a few things you've done 3 4 5 six iterations
things you've done 3 4 5 six iterations and then you have hit a roadblock of
and then you have hit a roadblock of okay why does why do people not uh you
okay why does why do people not uh you know move through this funnel and
know move through this funnel and usually the reasons are you know
usually the reasons are you know behavior communication and whatnot. So
behavior communication and whatnot. So that is probably the first cut and then
that is probably the first cut and then of course the second cut is if those are
of course the second cut is if those are there across the board then you can you
there across the board then you can you know use these filters to kind of pick
know use these filters to kind of pick and choose.
and choose. Yes go ahead.
Yes go ahead. >> Uh so Mallay my question is so if we
>> Uh so Mallay my question is so if we calculate the problem magnitude
calculate the problem magnitude >> is it in to that product prioritization?
>> is it in to that product prioritization? No, product prioritization is different.
No, product prioritization is different. We will talk about that uh when we talk
We will talk about that uh when we talk the next two next session actually as
the next two next session actually as well. Product prioritization actually
well. Product prioritization actually mostly comes into the phase when you are
mostly comes into the phase when you are already building something. What we're
already building something. What we're talking right now is pre-building which
talking right now is pre-building which is to validate if this is worth building
is to validate if this is worth building or not. So there of course if you have
or not. So there of course if you have multiple ideas you will have to
multiple ideas you will have to prioritize but prioritization is more
prioritize but prioritization is more handy when you have validated seven
handy when you have validated seven eight ideas and now you need to pick
eight ideas and now you need to pick which one to do which is what we will do
which one to do which is what we will do when we talk about tech and design and
when we talk about tech and design and development later.
development later. >> Thank you.
>> Thank you. >> Huh?
>> Huh? Yes. Go ahead.
Uh my question was uh see uh we discussed like three steps uh in the
discussed like three steps uh in the first we ideated the hypothesis second
first we ideated the hypothesis second and third in the business value when we
and third in the business value when we created this business funnel uh again we
created this business funnel uh again we uh were analyzing the data right uh I
uh were analyzing the data right uh I can have a different hypothesis that I
can have a different hypothesis that I created in the first step here uh
created in the first step here uh considering the drop offs when I'm
considering the drop offs when I'm trying to analyze the drop offs right
trying to analyze the drop offs right >> mhm you yes definitely see At the end
>> mhm you yes definitely see At the end you can have a different hypothesis but
you can have a different hypothesis but that hypothesis will still somehow be
that hypothesis will still somehow be aligned with the flow because the user
aligned with the flow because the user journey covers everything. So at some
journey covers everything. So at some point your hypothesis would be inserted
point your hypothesis would be inserted right. So for example in the Uber
right. So for example in the Uber example it would be inserted at the last
example it would be inserted at the last step where the driver has accepted but
step where the driver has accepted but they have not gone and uh picked the
they have not gone and uh picked the person uh which is because they canceled
person uh which is because they canceled in the middle and that is the hypothesis
in the middle and that is the hypothesis that we are focusing on. So, and of
that we are focusing on. So, and of course, this can be a case where your
course, this can be a case where your original hypothesis just doesn't feature
original hypothesis just doesn't feature anywhere in this business funnel that
anywhere in this business funnel that you create, but usually it will be in
you create, but usually it will be in some part of that entire funnel because
some part of that entire funnel because the funnel is very all-encompassing. The
the funnel is very all-encompassing. The funnel contains everything. So, that's
funnel contains everything. So, that's why it will usually fall somewhere in
why it will usually fall somewhere in the middle there.
the middle there. >> Okay. Or it will at least support it.
>> Okay. Or it will at least support it. >> Yes. Correct. Absolutely. Exactly.
>> Yes. Correct. Absolutely. Exactly. >> Okay.
All right, I know we are uh we were heavily over time already but uh thank
heavily over time already but uh thank you so much uh for joining. I will be
you so much uh for joining. I will be uploading the recording for this on the
uploading the recording for this on the LMS uh in a while and also uh a few of
LMS uh in a while and also uh a few of you have already submitted your
you have already submitted your assignments. So I will uh that's also
assignments. So I will uh that's also start I mean we'll start getting reviews
start I mean we'll start getting reviews of that as well in the coming days. um
of that as well in the coming days. um just try and unlearn a little bit um
just try and unlearn a little bit um because uh uh you might be in a
because uh uh you might be in a situation where you see a few things
situation where you see a few things that are very differently treated in
that are very differently treated in terms of how you think like a PM. So it
terms of how you think like a PM. So it will take a bit of time to kind of get
will take a bit of time to kind of get adjust that. So do attempt the
adjust that. So do attempt the assignment that uh is going to get due
assignment that uh is going to get due uh later tonight and um and yeah next uh
uh later tonight and um and yeah next uh time what we're going to discuss is
time what we're going to discuss is we're going to move to the next step
we're going to move to the next step which is design. So once you have
which is design. So once you have clarity on what to do, you then start
clarity on what to do, you then start working with designers on how you design
working with designers on how you design it and how it looks like. So that's also
it and how it looks like. So that's also going to be pretty interesting uh from
going to be pretty interesting uh from there. So so yeah, that's there and I
there. So so yeah, that's there and I can see a lot of you like the session.
can see a lot of you like the session. So that's great. Uh more coming your
So that's great. Uh more coming your way. Um I'll be uploading this there.
way. Um I'll be uploading this there. Apart from this, there is a small test
Apart from this, there is a small test as well. Uh I mean this is I think I we
as well. Uh I mean this is I think I we discussed last week as well that a few
discussed last week as well that a few tests just to check yourself. So that
tests just to check yourself. So that would also get live that would have
would also get live that would have questions on whatever we discussed uh
questions on whatever we discussed uh today and um and yeah that would be
today and um and yeah that would be great. Um apart from that of course in
great. Um apart from that of course in the coming weeks a lot of hackathons are
the coming weeks a lot of hackathons are planned and a lot of uh different kinds
planned and a lot of uh different kinds of sessions are planned so we'll uh
of sessions are planned so we'll uh we'll work it out and um and yeah great
we'll work it out and um and yeah great and really happy that uh so many of you
and really happy that uh so many of you like this um cool then I will see you uh
like this um cool then I will see you uh soon and thank you so much for your time
soon and thank you so much for your time and have a great rest of your uh
and have a great rest of your uh Saturday.
Click on any text or timestamp to jump to that moment in the video
Share:
Most transcripts ready in under 5 seconds
One-Click Copy125+ LanguagesSearch ContentJump to Timestamps
Paste YouTube URL
Enter any YouTube video link to get the full transcript
Transcript Extraction Form
Most transcripts ready in under 5 seconds
Get Our Chrome Extension
Get transcripts instantly without leaving YouTube. Install our Chrome extension for one-click access to any video's transcript directly on the watch page.
