This content provides a comprehensive guide to setting up effective cold email infrastructure, detailing essential tools and best practices for maximizing deliverability and response rates. It emphasizes a strategic approach to domain and inbox management, email sending, data enrichment, and leveraging intent signals.
Mind Map
Click to expand
Click to explore the full interactive mind map • Zoom, pan, and navigate
Hey guys, what's up? Kyle Dixon here. In
this video, I'm going be showing you
guys the exact process on how to set up
cold email infrastructure the right way
and some good tools you guys could use
and some things to know about each of
these tools. So, I'm going to be
covering everything from how to set the
inboxes, where to get the inboxes at the
cheapest places, domains, email sending
tools, enrichment tools, intent signals,
all that stuff. So the only the f I
guess the first logical place to start
with this whole process is the domains.
So for domains
kind of hopping straight into this is I
usually buy from GoDaddy or NameCheep.
Now you can get them um from wherever
you want. It doesn't really matter too
much on where you buy them. Um but I get
mine from GoDaddy or NameCheep. Okay.
Now, once I buy them, I wait 30 days uh
before I start sending emails. So, you
want to age your domains at least 30
days before you start sending emails.
You could push it and do uh 21 days.
It's up to you. Um I've done that before
and it it's fine, but
ideally you want to wait 30 days. Okay.
Now, for the next step, only have two to
three inboxes per domain. So, what does
this mean? This means for example an
inbox would be kyle@nike.com.
That's one inbox, right? That's one
inbox per domain. Now imagine I have
That's two inboxes per domain. Now you
know what you know what a third looks
like. It could be ryan@nike.com.
Right? So that'd be two to three inboxes
per domain. anything over that
and you're gonna have too many cold
emails being sent from one domain which
is going to destroy your domain
basically health. So there's three
different types of deliverability
infrastructure things you need to really
think about here when it comes to like
infrastructure wise. So there's
copyrightiting and stuff like that but
the three main things are inbox health,
domain health, and IP address health.
Now, Google and um
Microsoft take care of the IP address
health. If you do buy private SMTP
server uh inboxes,
you will have to take care of the IP
health yourself. Okay? So, IP you don't
really have to worry about too much
unless you're buying SMTP inboxes, which
I don't really recommend doing in the
beginning. If you want to do that in the
future once you get more experience, you
could do that. Uh but for Google and
Microsoft, you don't really have to
worry about that process. So now for
domain health, how you hurt domain
health is by blasting too many emails
from too many inboxes under one domain.
So we the way we take care of the health
of the domain is limiting the amount of
inboxes we have per domain if that makes
sense. Now inbox health is not sending
too many emails per inbox. Okay.
Now, another thing to keep in mind for
domains is only send cold emails from
the secondary domains.
So, what does this mean? If your primary
domain is nike.com,
you should buy secondary domains that
are go nike.com, get nike.com, and you
send cold emails from the secondary
domains. So, if you start sending cold
emails from the primary domain,
nike.com, and you get your primary
domain blacklisted,
then your website's not even going to
show up on Google because your domain
health for your inboxes is a is tied to
your website domain health. It's the
same domain. So, if you have a good the
same thing goes if you have a good
domain for website. So if you have a
good domain for your website um and you
start sending cold emails, your
deliverability is going to be really
good in the beginning until you destroy
it from cold email. So make sure you buy
secondary domains. Okay. Now for
inboxes, I buy my inboxes through scaled
mail. On average, each inbox is $3.50.
Now if you buy these through Google,
it's $8.40 per inbox. So, these are
going still going to be Google inboxes
as you guys can see here. Um, you could
buy Google inboxes. You guys could buy
Microsoft inboxes. You could buy, I believe,
believe,
SMTP inboxes. That's what I was talking
about with the IP address stuff. So, you
could buy them here. Now, I recommend
getting Google inboxes. They have the
highest deliverability right now. So, if
you're going to get started, get this
one for 140 bucks a month. 40 in boxes,
20 domains required. You can send up to
24,000 cold emails per month. Now, if
you want to level it up, I recommend
getting this one. So, you can send
42,000 42,000 cold emails per month, 245
bucks, 35 domains, 70 inboxes, and you
can keep leveling it up if you want to
get to 100 inboxes, 350 bucks a month,
60,000 emails. As you can see, you can
get whatever you want here. All right.
Now, what I recommend when you get your
inboxes is that you only send 20 cold
emails per day per inbox. So, if you
have 10 inboxes, you should only be
sending 200 cold emails a day across all
10 inboxes. Okay? Now, you want to send
25 warm emails a day. So, what this
means is inside of the tool that we'll
talk about later, instantly, they have a
warm-up automation where essentially
there's other people who also use
instantly, right? And it will send
emails back and forth from your account
to other instantly accounts and from
their instantly accounts to your
account. So, what these warm emails are
going to do is it's going to counter
interact the cold emails you send so
your domain health and inbox health can
stay healthy. Because if you're sending
20 cold emails and your your reply rate
is like 2%.
Google's going to see that and see that
you're spam. Because usually when you
send emails to people, you're getting
replied back to at least 30 to 50% of
the time, right? So that what the warm
emails are going to do is it's going to
simulate uh good conversations happening
on your in your inboxes and on a domain
level too, which is going to it's just
going to basically keep your domains and
inboxes healthy. Now I So you want to
send 25 warm emails a day, 25 cold
emails a day, which means each inbox
should be sending around 45 emails a day
in total. Okay? Now you want to warm up
your uh inboxes for at least 14 days
inside of Insta and you want to ramp up
the warm emails. So the first day of
warm sending warm emails should be like
two emails a day. The next day you
should be sending four warm emails a
day. The next day you should be sending
six warm emails a day. And you're going
to slowly build up to that 25 warm
emails a day over a 14-day period. And
the same thing with the cold emails.
Once you launch your cold emails, the
first day you're only going to send like
one or two cold emails a day per inbox.
The next day you'll send about, you
know, three or four cold emails a day
per inbox. And you'll slowly ramp that
process up over two weeks to like 20
emails a day per inbox. So that's
domains, that's inboxes. Now, the email
sending tool, Instantly.
So, inside of Instantly, basically what
this tool does is it sends emails for
you automatically. So, instead of you
having to email all your prospects one
by one manually, you could just put your
lead list in there with a S, you know,
CSV file like this. And you put your
lead list in there.
And then you could put your schedule in
here. Okay. So then [snorts] you just
press launch, resume campaign. So that's
kind of what that looks like. Um,
basically infrastructure wise, I don't
know why I'm on here.
Um, so that that's the automated part.
So it sends emails automatically. Now
also has an AI spam checker for
copyrightiting. So, kind of what that
looks like is if I go
um here and we preview some copyright
and then I test it out
uh spin tag. So, like basically what
this does, it looks for spam words in
your copyrightiting. So, it says this is
a spam word. So, there's over like 300
to 400 different spam words. And if
Google sees a lot of those spam words in
your copyrightiting, even if your inbox
health, your domain health, and your IP
address health are good. If it sees spam
copyrightiting, it won't even send it
through. So fast could be a spam word.
Now, it doesn't need to be 100% perfect,
but it needs to be as close to perfect
as possible. So like I would say around
one spam word or less, ideally zero.
Okay. And then make sure you press save.
Now, daily placement test. So inside of
instantly there's things called daily
placement test. So you basically this is
the most real data you could get on your
deliverability. So you could send you
know all these different emails from all
your inboxes testing out your copy uh
which essentially is test just testing
out um if your inbox health is good and
your copyrightiting is not spam. Now the
next thing is they have AI reply bots.
So if you get, you know, the biggest
thing to in cold email is when you get a
lead is we're responding back as fast as
possible. [snorts] So they now have
these AI reply bots where once you train
them up, they can essentially reply back
to every single lead you get
instantaneously. So think Harvard did a
study and said if you reply back to a
lead within like one to five minutes,
you have a 300% increase in conversions.
So that's why this AI reply bot is so
huge because imagine you're like taking
a poop in a toilet, right? And you get a
new lead and you miss that lead, your
percentage of closing them is going down
unless you have like a SDR or something
like that, right? Now, same thing I
talked about earlier, they automate the
warm up and uh warm up process for cold
and warm emails. So if I go to here,
you can see warm emails that are sent 20
a day, right? And you can see the
settings. So I can set a campaign limit,
increase slow ramp up process. I just
bought some more inboxes, so these
aren't set up properly yet. Um, I just
set that up right now. 20, right? And then
then
inbox placement test. Oh And then
so you can increase your warm-up. So you
you want to increase by one or two
emails a day. And then daily warm-up
limit, reply rate to the warm-up. So,
how many peop how many of your warm-up
emails are getting replied to. Now, your
open rate should really be around like
here. It just just put it in a higher
range. There's no really perfect number.
Uh, I turn this on. I send cold emails
on the weekends, so I disabled that. So,
that's kind of like the whole warm-up
process of instantly. Now, the unbox is
essentially a place I'm not going to
show this because I have all my
conversations in there. Um but
essentially it's this place where
imagine a unicycle, right? A unicycle
has one wheel. Um a bicycle has two
wheels. So a unib is essentially an
inbox for all your inboxes that are
sending cold emails. So when I go here,
all these in these all have different
inboxes, right? So if I get a reply back
to one of these, that means I would have
to monitor all 40 or 50 or 100 inboxes.
What this does is is right here, my uni
box, all my replies go in one place. So
any replies that I get go into one spot.
Now for analytics, analytics, you could
check your basically your reply rate, um
your positive reply rate, your
opportunities, your conversions,
uh emails sent, and stuff like that for
all your different campaigns and on an
account level. Now, there's also a CRM
in here. I'm not going to show that
because I want, you know, everyone to
see my data. Um, but there's a CRM in
here and essentially, um, it's like a
card-based system like kind of like
HubSpot and there's a C there's like
basically a free CRM in there. I mean,
now there's advanced stuff in that CRM
that you won't get unless you pay the
upgraded plan,
but it's good enough for what you need
to do. Now, there's AI spin tax.
So and instantly you could create
basically spin tax. So what this spinax
looks like is it looks like this. So for
example at random pick hey first name,
hi first name, hello first name. So
let's look at this here. See I have
different combinations. Let's look at
preview. So we got hi Mason, right? Hi
Mason, hello Mason. See how it's
randomly switching this up. So,
basically what this does is I have it
through bunch of different places here.
Random. Random. How this works is you
want to have as many different
combinations of your copyrightiting as
possible because if you send the same
email 10,000 times, Google will
recognize it and see that you're just
you're just blasting spam. So what you
want to do is you want to create
different combinations of your email as
many as possible that essentially is the
same email just written differently. So
every email isn't the same. If every
email isn't the same, Google's not going
to pick up on trends and you're less
likely less likely to be marked as spam.
Now lastly, lead database instantly has
a database. Believe it's up here. Yeah,
super search where you could filter by
location, job titles, industry keywords,
employees, revenue, lookalike domains.
So yeah, there's all different stuff in
here. You can even tell AI, hey, give
me a list of kitchen remodelers
remodelers
in Arizona,
and I only want CEOs.
So, I've actually never done this in
here before because I use clay. Okay.
Well, I guess it's not liking that. Uh,
Huh.
There's no way they don't have any
kitchen remodeling contractors in here.
Oh, there's industry.
Let's see what we got here.
Construction. Right. So, there's 34,000
results. Um, and yeah, you can look
through here and you can filter by
keywords, too. So, you can put like
kitchen in there and then these are all
going to be like kitchen remodeling
contractors. Okay? So, if I look up, you
know, all these different companies, if
it has keywords, it means they have
these keywords on their website. So, you
could also exclude keywords like uh roof
roofing, right? If you don't want to
reach out to roofing contractors, but I
don't want to spend too much time in
that. But that's basically the lead database.
database.
Make sure my computer isn't out here.
Okay, Clay. So, Clay is a tool that you
don't need because it is super
expensive, but it's it's extremely
helpful. So, in this tool, you can
enrich data, score data, and personalize
data. So, enrich data means you just get
more data on the prospect than you
already have. So, for example, right, if
we scrape a bunch of window, you know,
window replacement companies, all we're
going to have is the window replacement
company name, the name of the CEO,
uh, the website, and the domain, like
stuff. Basically, basic stuff like that.
But we don't know how much this
company's making. We don't know how big
this company is. So, an example of an
enrichment would be getting the
headcount of an organization,
getting their web traffic. we're getting
these enrichment, you know, enrichments
that we could use to score the prospect
against our ideal customer persona. So,
for example, if we're only trying to
work with window replacement companies
that have over 50 employees, we need to
enrich the headcount on all the window
replacement contractors that we scrape
to make sure they all have over 50
employees. So, that's what we do,
enrichment. We could then score it and
then we could then personalize it. So,
one of the examples is cleaning company
names, personal names, and domains. So,
when you scrape a list, the data is
going to be kind of messy. For example,
if I come into here, when I scrape this
data from Clay, this is Clay's database
that I scraped. This was the first name,
Jason M., that's not her first name.
Their first name is Jason. Look at this
one. I don't know how to pronounce it,
but hean Muhammad. That's two names in
the first name. Juan Manuel Manuel Anas
Wardy you know Ronaldo in all caps like
that's Dr. Panto you know you have it's
not a first name right so what we use
for clay is we clean the first name up
for some reason it's not showing it
properly but for example Jason
I told it to output Jason now the clean
name is Jason right so I can take tell
AI to go through all the first names and
return just the first name cleaned so I
can do that whole process same thing for
last name same thing for um their full
name, right? If you look at full name
here, we have CPA in it, CPA in it, all
caps. So, we could clean the names in
here with AI. Okay, so that's kind of
that process for cleaning stuff and then
you could score it. So, for example, we
actually we'll come back to that. Now,
we could find more emails in your total
addressable market. So for example, most
cold emailers only have one tool which
is usually like lead magic or something
like that. So how this tool works is
it's called a waterfall. So
as you can see here, I tried to find the
work email for these different rows. I
found a few, but I didn't find any right
here. If you look down here, I found it
with find email. Right? So basically how
this process works is it first takes
lead magic to try to find the work
email. Then it'll try ICPS to try to
find the work email. And if that ICPS
doesn't find the work email, it'll try
to find it'll try to find it with find
email. If find email doesn't find it,
it'll try to find it with Prospio,
right? And it'll keep going down this
list until it tries to find all the
emails and then it will output them
right here, right? And it will also
validate them. So you can pick your
validator. So mine's lead magic, right?
So that's what I mean by click find more
emails. You can do this waterfall
process that will help you find more of
those emails. Okay, AI research on
prospects. So how does this work? Okay,
this is a fun one. So
AI research on prospects.
I don't think I did any in here. But for
example, if I say,
Return me the
This is a [clears throat] really random
one I've never done before, but website.
I didn't create a website column, but
that should do the job.
Okay, it's creating the prompt.
So, there's certain prompts on the way
you prompt the web scraper. So, you
basically just tell what you want and
then AI will inside of Clay create the
prompt for you. As you see, like I I
don't have I don't know how to type all
this stuff out. It would take a long
time. So, I just tell you how to do it
for me. Okay. It's going to look at the
CSS variables
on the back end of the website.
Okay. It's almost done here.
Okay. Okay. Okay.
Now, the cool thing is about Clay is you
also can use your own API key. So, you
don't have to spend an insane amount of
money with Clay credits. So, my my API
key hooked up here.
We'll run this for 10 rows. This is a
bad example, but you could say, for
example, case studies, right? Or you
could say, if you don't have the c the
CEO's name, you could say, "Hey, scrape
the whole website and return me back the
CEO or the owner or the COO's name." And
then you could take the name and you can
find the email. See, you can use
basically all these different tools to
enrich the data.
Okay, of course, it's not going to work.
Okay, so this is for example, it
returned back
um six colors. So one of the colors was
okay, it's returning hex. So the next
step I could tell it to do is another AI
column. I could input all the hex codes
and say, hey, what are these hex codes
in, you know, words, like is this red,
is this green, stuff like that. I could
have also told Clay to do that, but um
not going to lie, I kind of got to take
a poop. So um just trying to be quick
with it. Okay. So there's also insane
insane amount enrichment data. So for
example, if I go to ad enrichment, I
could basically find monthly web
traffic. I could find the latest funding
round, their all their website tech
stack. So what technology is hooked up
to their website like HubSpot um instantly
instantly
uh RB2B. I can find their revenue. I can
find how many job openings they have.
Um, I could determine the revenue models
with the AI web scraper I just showed
you. Um, does the company offer a free
trial? We could check their website. You
know, there's all these different things
you could do inside of there to
basically enrich more data. And you'll
run it for all the columns.
Okay. Now, AI personalization. So, now
inside of Clay, you can basically set up
conditional things. I'm not going to go
too deep into this because this is a
whole video in itself where I could
personalize cold emails with all this
enrichment data that I was just showing
you guys. So, for example, hey John, saw
you like, hey John, um, saw you had a
new job opening at nike.com for this
position and that position, you know,
and that in that company name would be
enrichments that you got from Clay,
right? It's like, hey, I saw you had a
software opening position at Nike.com. I
don't know why they would have a
software opening position, but you could
basically just personalize the cold
emails based off the variables that you
enrich. Okay? So, every email is
different and personalized. If the
emails are personalized, they're way
more likely to respond.
And okay, now the cool thing about this
is you could automate the full process
of this thing. So, you could set up a
whole table with all these different
enrichments step by step. Hey, I want
you to clean the first name. I want you
to clean the company name. I want you to
do this. And you can create a template.
And then you just import a template, put
all your data in, and the whole table
will run from left to right. And you can
even send it to your instantly tool. So
uh, add to campaign. So you can
literally map everything out,
personalization, custom variables. So
once the whole automation runs, right,
you have the table set up like a
template. You put the data in, the whole
table will run from left to right. It
will clean all the names. It will find
the emails. It will scrape the
technology stack. It will find the
monthly web traffic on their website. It
will find any job openings. And then it
will score all the data based off all
the enrichments that I got. And then it
will personalize the cold emails. And
then it will automatically sent to instantly.ai.
instantly.ai.
Right? So that's the crazy thing about
Clay is you can have these automations
that enrich your data, score your data,
personalize your data instead of having
to redo it manually or, you know, one at
a time. Now lastly, I'm running out of
time. I'm at 3%. You guys can see here.
Uh, intent signals. So, Common Room,
UniFi, Clay, these are all different
intent signal platforms. There's
hundreds of intent signals. I'll show
you those in a second. Um, but
something I really want to tap on, then
we'll come back to those is RB2B. So,
what's RB2B? RB2B is something you could
put on your website. So, if you use G4
and you're familiar with it, you know,
the only data you really get is like the
age, uh, the gender of the person on
your website. you don't really know who
they are. You don't know their name. You
don't know their company name. So what
our B2B does is you put on your website
and then it will track the different
companies and names. So it will tell
you, hey, Kyle at root marketing was on
your website. So as you can imagine,
there's hundreds if not thousands of
people that go on your website that you
don't know about. RB2B tells you that.
And then you could automatically send
that data to Clay in a web hook, enrich
that data to find the headcount, what
industry they're in, and then you could
score them, and then you can send a
personalized cold email and do that all
automated. So that's the cool thing
about Clay once again. Now, triggery
basically tracks different types of
intent signals mainly for LinkedIn. So
it could attract who liked your LinkedIn
content, who comment on your LinkedIn
content, who is viewing your profile,
who are your connections, and they could
take all that data, send it to Clay. You
could have a Clay automation that
enriches every single one of those
prospects to find their headcount, what
industry they're in, and then you can
take all that enrichment data and score
it against your ideal customer persona,
who, for example, might be, hey, I only
want I only want to work with SAS
companies that have a 10 to 50
headcount. Okay, so it will take all the
enrichment data, use AI to score that
against your ICP to make sure they
actually are an ideal customer, ideal
customer, and if they are, it will send
automatically send them into an
instantly account and send them a cold
email and you can do that whole process
automated. Now, lastly, I'm 1%. All
right, guys. All right, guys. Just came
in clutch. Um, I had 1% left. My laptop
clutched that one up. Now, lastly,
intent signals clay. How does this work?
We'll hop into Clay real quick. We'll go
to here.
We'll go to See, there's different
signals here. There's job changes, news,
and fundraising. Uh different types of
10 signals here. Now, one of my favorite
is news because there's so much to it.
Um we'll just use [sighs and gasps] clay
table this. So, you can take your whole
clay table. For example, you could take
your whole total adjustable market. So,
if you work with SAS companies um that
are in the fitness industry, you could
scrape all those SAS companies, put them
in a table and and track intent signals
live. So, when one of your companies in
your total addressable market has one of
these intent signals, you could
automatically send them a cold email
instantly. So, for example, um
um
if one of the people in your market gets
a new award or they're recognized for
something, you could send them a cold
email automatically. So, the the cool
thing is is you could take your whole
market, your whole all your ideal
customers and you could track all this
different data on them and when one of
these things happen, you can send a
personalized cold email. And if an email
is personalized, they're like, I think
three times more likely, if not like up
to 10x more likely to respond. I've seen
cold email campaigns with 30 40%
positive reply rates, which means if
you're sending emails to a,000 people,
300 to 400 of them reach back positive.
So, intent signal campaigns are like the
highest positive reply rate campaigns,
which is why it's so huge and why I'm
going over it. So this is kind of more
advanced, but if you're at this stage
and want to implement this stuff, this
is kind of where you look at it. So did
any of the companies get, you know, a
new alliance partnership? Are any of
them going bankrupt? Are any of them
closing new branch or opening a branch?
You know, business expansion? You can
see this stuff. Are they getting bought
out? Are they cutting cost? Are is there
a new patent slashtrades trademark
dispute or infringement?
Um, did they just do a big sponsorship
and donation? You know, uh, downsizing,
headcount's going down, you know,
employee strike, fundraising, they're
going private, they're hiring, are are
they going IPO? Is there a new merger?
Uh, merger termination. All these
different intent signals you could track
on your ideal customer persona. um you
just got to put them all in clay and
then pick the signals you want to track
and then when they happen you send that
you send a very good personalized cold
email to try to sell them something.
Right? So these are really the five
things cold email infrastructure wise
you go you know inboxes I get them
through scale mail domains email sending
tool I use instantly.ai it's about you
could start with $37 a month but to
really make some damage you're gonna
have to spend around $97 a month on
this. Clay is around it's 140 but with
the API keys to save a bunch of money
it's like 344 bucks a month. Uh intent
signals RB2B I believe is like I think
around 100 bucks a month. This is around
like 100 200 bucks a month I think. And
then common room I think is like 700
something like that. Um I don't have
common room so I don't know but I see a
lot of people like it. I don't really
need it right now. Um, but yeah, so
that's intent signals, Clay, email
sending tools, domains, inboxes. That's
everything you really need to know about
cold email infrastructure. If you guys
have any questions or comments down
below, let me know. Also, check out my
other video where I talk about how to
save thousands of dollars on finding emails.
emails.
And yeah, guys, that's it for today. Um,
thank you guys so much for watching.
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.
