Hang tight while we fetch the video data and transcripts. This only takes a moment.
Connecting to YouTube player…
Fetching transcript data…
We’ll display the transcript, summary, and all view options as soon as everything loads.
Next steps
Loading transcript tools…
How to Show Up in ChatGPT: HubSpot's Winning AEO Playbook | GROW EUROPE 2025 | HubSpot Live - AI Summary, Mind Map & Transcript | HubSpot Live | YouTubeToText
YouTube Transcript: How to Show Up in ChatGPT: HubSpot's Winning AEO Playbook | GROW EUROPE 2025 | HubSpot Live
Skip watching entire videos - get the full transcript, search for keywords, and copy with one click.
The core theme is the critical shift in how users discover brands due to the rise of AI-powered answer engines, necessitating a new strategy called Answer Engine Optimization (AEO) to maintain and grow visibility and demand.
Mind Map
Click to expand
Click to explore the full interactive mind map • Zoom, pan, and navigate
Good afternoon, Grow.
You ready to talk about AEO
or are you here to save a good spot for
Louis Theorough? Well, regardless of why
you're here, I'm glad you are because in 2028,
2028,
the majority of answer engine traffic is
I would have dropped my mic if it wasn't
physically attached to my person.
But let's just take that in for a
second. Google, the undisputed traffic
overlord of the past 20 years, and I do
mean overlord, is no longer number one.
That's not a small shift. That is a
tectonic plate movement under everything
we do in demand gen. And that means that
what you do now is going to determine
whether your business is winning in 2028.
2028.
At HubSpot, we saw this shift happening
early and we quickly pivoted our strategy,
strategy,
which is why I'm proud to say we are now
the most visible CRM in answer engines.
My legal team, Erica, our chief legal
officer, is here today. They didn't want
us to show you uh competitor names, but
I'll give you a hint. They rhyme with
Yoho and Sales. Whales course. Almost
gave it away there. You wouldn't have
gotten it if I hadn't made that mistake. [laughter]
[laughter]
[gasps] I'm Asia Frost. I am the senior
director of global growth and paid
advertising at HubSpot. I was born in
2016. I'm just kidding. I haven't gotten
that much sun. I started at HubSpot 9
years ago in 2016. And honestly, I think
that this is the most exciting era of
the company and marketing yet.
So, over the next 29 or so minutes, I'm
going to walk you through what is
changing about how people discover
brands like HubSpot and yours, what a
successful top offunnel demand strategy
looks like in the AI era, and most
importantly, your strategy for executing
it. And we're going to even have some
time for Q&A at the end. So, please
start thinking of your questions now,
but good ones only. Okay. The first
major change that I want to talk through
is where conversions are coming from.
Three years ago, highlevel educational
content like HubSpot's blog. Can I get a
whoop if you've ever read a HubSpot blog post?
post?
Okay, that was, you know, a little
anemic, but I'll take it. Uh, that was
driving the majority of HubSpot's top
offunnel demand, and the rest was coming
from paid and channels that put humans
front and center, like HubSpot Academy.
I'm afraid to ask for a whoop, but who
has taken a HubSpot Academy course?
>> Oh, stronger. Okay, thank you. And at
the time, we looked at this data and we
realized just how dependent on Google
search we were and how much of an
opportunity these personalitydriven
channels represented.
So that's why in 2021, we acquired the
Hustle. We built out HubSpot's
thirdparty creator network in 2022, and
we launched multiple YouTube channels
alongside our OG HubSpot marketing
channel, including How to HubSpot,
Marketing Against the Grain, which I
know y'all got a live uh recording of
earlier today, and more.
And in 2025, that bet has really paid
off. Human first content is making up
the majority of our demand and we get 35
to 40 million engagements every single
month across our media network.
Educational content meanwhile is just
28% of our demand.
And that's because of our changing
investments. But it's also because what
is happening on the internet as you may
have heard maybe on LinkedIn Google's AI
generated summaries have siphoned away
educational traffic. And not just HubSpots,
HubSpots,
AIOS now show up in nearly 6 in 10
searches. And when they do, clickthrough
rate is hald. Google is keeping people
on Google rather than sending them to websites.
websites.
Buyer behavior that's changed, too. And
I put the you and behavior for you. Europe.
Europe.
More than 50% of buyers now say they are
using AI to make purchasing decisions.
Armed with a short list, they turned to
other people, watching YouTube videos,
reading reviews on G2, Capterra,
browsing comments on social media. Shout
out to Huban420, our biggest fan. May
you never stop posting.
And by the time they arrive at a
vendor's website, they are far more
informed and they have far more
conviction in their choices than they've
ever had before,
which is why they convert at more than
three times the rate of traditional
visitors in a lot less time and they
generate a lot more revenue. It's a win-winwin.
win-winwin.
That means that the traffic that we are
able to drive from answer engines, not
only is it going to become one of the
most predominant sources of demand, but
it is incredibly valuable.
And that brings us to AEO, the content,
technical, and off-site tactics that you
can use to maximize your brand's
visibility in answer engines.
AEO belongs in the amplify stage of the
loop marketing framework and I am really
excited to walk you through the the
plays that HubSpot is using. But to
understand why those plays are working,
you need a crash course in how answer
engines formulate answers. I promise I'm
going to make this quick and painless.
The first concept that is very important
to know is called query fan out. When
someone asks chat GBT a question or
Google something that triggers an AI
overview like where's the best place to
go after grow AI doesn't just come up
with the answer to that one question.
It's going to break it into a bunch of
sub questions like bars in Westminster
that don't charge 22 for a martini.
Places to casually bump into grow
speakers without it being awkward. Best
late night food after drinking Westminster.
Westminster.
best place for drunk food that won't
destroy you.
And then it's going to go out and it's
going to find all the answers to those
specific queries and synthesize them
into one result.
I'm not sure that I have all four stages
in me tonight, but if you do, godspeed.
The second concept you need to know is
called memory. Every time you ask a
question, AI is going to factor in a lot
of additional information about you.
like everything you've ever asked
before, what you've clicked on, your
emails, your Google Drive files,
anything you've connected to Claude or
ChatGBT like a sauna or HubSpot. If you
haven't connected ChatGBT and HubSpot,
that is the one reason you can pull out
your phone right now. That won't hurt my feelings.
feelings.
And what that means is that every
response AI is generating incredibly personalized.
personalized.
Okay, those are the need to know
concepts. Congratulations, you have just
earned a very official badge for your
The first tactic that we are going to go
through under AEO is content.
And I'm going to give you a very simple
strategy to increase your visibility in
AI. Stop publishing the ultimate guide
to everything and start publishing the
exact answer to the exact question your
buyer is asking. In tragedy,
AI has completely commodified broad
educational content like the ultimate
guide to content marketing. So for every
101 guide that you used to publish, you
need to create 10 laser focused pieces
that each answer persona specific questions.
questions.
But where do you start? You could start
anywhere. At HubSpot, we have developed
a very straightforward process to figure
out who we should be creating content
for and on what topics.
And it's going to start like many good
things do with a 3x4 grid. You're going
to create a 3x4 grid for each product.
And across the top, you'll list your
buyer personas. The more specific, the
better. So, you don't want marketing
manager Margaret. You want marketing
manager Margaret at a 200 person
logistics company. Because AI responses
are so contextualized, the way you think
about visibility and content needs to be
highly contextualized, too.
Then you're going to put the stages of
the buyer journey down the side and in
each cell you're going to map the high
intent questions that that persona is
asking in that stage about that product.
So you're going to have a ton of grids.
At HubSpot we started with separate
grids for sales hub, marketing hub,
service hub. You get the gist. And we've
now created enough content that we are
going down to grids for the feature layer.
layer.
But how do you actually populate those
grids? How do you figure out what people
are actually asking in AI? Okay, there
are three methods and you're going to
use all of them. The first is pulling
keyword data from AHF, Semrush, Search
Console, Pick Your Poison. You might be
thinking, Asia, why are you talking
about keyword research tools? Is this an
SEO presentation or an AEO presentation?
No, it is an AEO presentation, but we
don't have prompt data and volume from
answer engines yet. Until we have that
data, then you are going to use your
keyword research as the best proxy for
what your audience cares about. But
you're not going to stop there. You are
also going to use social listening tools
like Meltwater or Common Room to pull
questions from Kora, Reddit, YouTube,
Instagram. Your buyers are asking
questions in these places already
formatted as questions. So, these are
some of your best clues as to the
questions that people are asking in AI.
And then you are going to layer in chat
transcripts and sales calls. Mine your
CRM. If you don't have these
transcripts, then actually talk to your
sales and customer success people and
ask them what questions people are
asking. These are your highest intent
questions and so they are the most
important to have content on.
All right. Once you have compiled all of
that data,
you're going to tag the queries and sort
and format them into funnel stages. I
highly recommend creating a chat GBT or
a cloud project to do this. And you can
use this very straightforward framework.
Awareness questions look like unbranded.
How do I do X? What's the best way to do
Y? Consideration questions are best
tools for ABC, top solutions for XYZ.
While Valuation questions compare
multiple options head-to-head. And
decision questions ask whether a
specific product or service can
accomplish a specific task.
Here's how that all comes to life.
Again, please use AI to do this. If you
don't, you will end up cursing my name
or your marketing associate will curse
my name, whoever does this work. But
either way, I don't need the bad karma.
Now that you have your questions, you
have to figure out where your brand is
and maybe more importantly isn't showing
up in those responses.
In March 2024, I built and launched
HubSpot's AEO grader with the help of
our incredible webdev and marketing
technology teams to help brands
understand just that. AOGrader will give
you a fantastic highle overview of your
visibility. It's totally free. We just
launched it in five more languages.
But to go a level deeper,
use a tool like Xfunnel. HubSpot
actually just announced that we were
acquiring Xfunnel last week. So when I
gave this presentation at Inbound, I
couldn't say that. Now it's all one big
happy HubSpot family. And Xfunnel is
really, really powerful because it shows
you visibility by question for all major
answer engines. So you can see here that
HubSpot is appearing in these responses
for this question across all the engines
that we care about. But there are some
questions we care about where we don't
show up and those are visibility gaps.
Once you've identified visibility gaps
for your brand, you can start creating
actual content. And here's where AEO
truly starts to diverge from traditional SEO.
SEO.
Let's take a hypothetical awareness gap
for HubSpot. As a sales representative,
I have no idea which leads are worth
pursuing. What are my options?
In the old world, we'd create a guide to
lead scoring. We'd optimize it for
Google. We'd slap an ebook CTA on it,
and we'd call it a day. But if you
remember query fan out from a few
minutes ago, to show up as a
recommendation in answer engines, we
need to create hyperspecific content
answering each subquery.
My team at HubSpot built an internal
tool to pull the exact sub questions
that answer engines break a query into.
You can use Cuporia by Mike King or
Dian's fanout tool. These are both free
tools. They work the same way. They will
take a query and they will give you a
structured breakdown of related sub
questions. Exactly the questions that AI
is looking for answers to when it
synthesizes a response.
So this question about how to
effectively prioritize sales leads
breaks down into the following queries
which we need to make sure our page addresses.
addresses.
And that leads us to the next step,
creating content that is structured for
answer engines.
These are the best practices that we
have learned and implemented at HubSpot.
I'm going to umbrella them all under
technical tactics because they're all
about the structure of a page. And fair
warning, I'm gonna go pretty fast here,
but there will be a very snapshot ready
slide at the end of the section, which
is the second reason you can pull out
your phone without hurting my feelings.
I will say nothing warms a speaker's
heart like seeing a row of cell phones.
It's very nice. [gasps] Okay, number one
on our AEO page checklist, put the
answer first. The first sentence should
answer the primary question completely
because answer engines are looking for
that quick validation that they're in
the right place.
So, here's what that looks like. The
most effective way to prioritize sales
leads is by using a lead scoring system
that ranks contacts based on their fit
and engagement. The lead is not buried, folks.
folks.
Okay, the next best practice. Your page
should then go a click deeper.
And this looks like two to three short
paragraphs that are going to expand on
your answer with context, definitions, methodology.
methodology.
This will signal to answer engines that
your answer is credible and complete.
Number three, reference original data
throughout. Answer engines are looking
for net new information to site, which
you can provide in the form of stats,
original data, and case studies.
This doesn't need to come from expensive
surveys. Usually, when I talk about
original research, people get kind of a
weird look on their face and they're
like, "Oh, we don't have a ton of
budget. How are we going to do that?"
Use your CRM, hopefully your HubSpot
CRM. Again, talk to your customer
success and sales teams for original
insights. Kip Bodner, our CMO, who is
here today, uh, got this insight from
talking to our marketing and sales teams
about how they improved our own lead
scoring systems. This was a very
inexpensive original insight for us to
provide in this blog post.
Number four, include an FAQ.
At the bottom of every page, you are
going to answer at least three sub
questions, each with its own mini direct
answer and short explanation. So, a
micro version of how you framed the
overall page. And you're going to format
each one of those questions with its own
H3 or H4 heading with the goal of
getting AI to really easily find and
lift it. And these FAQs are also a great
way to optimize for fan out queries. You
can basically knock out three to four at once.
once.
Number five, add structure up the woo.
Answer engines are lazy readers. So,
it's kind of like when I say I read an
article and I actually mean that I
watched a Tik Tok. So, you want to make
your content as easy to understand as
possible with bullets, super explicit
headers, tables, structured lists,
Number six, make sure each section can
stand on its own. Answer engines don't
actually read pages, they read things in
passages. It's called chunking. That's
literally the technical term, which
always reminds me of, you know, what
your dog does when it's not feeling
well. Which means for you that every
section needs to make sense, even if AI
hasn't read the section before or after
it. All comes back to this concept of
lazy reader.
Here is an anti-example from a little
newspaper you might have heard of, the
New York Times. It's a pop quiz,
requires audience interaction. What do
you think Mr. Strawmire is talking about
here? And you can just shout answers out.
out.
Sandwich. Okay, it's pretty good.
Any other guesses?
A wrap. Okay,
you're pretty clued in. These are good
guesses. Okay, as long as you have that
crunchy thing in the middle and you know
how to fold it, you can put anything in
there. So, you're circling around the
idea. The answer is a crunch wrap. For
anyone who has never had the glory that
is the crunch wrap, it is like the Greg
sausage roll for Americans. I did use
ChachiBT to find that reference. I hope
that's close. [gasps]
[gasps]
But if a human is going to struggle to
understand Mr. Strawmire's reference
about context, an answer engine is
certainly not going to be able to. So
when you are creating content, when you
are reviewing content, literally give it
the Taco Bell test and make sure that
every section can stand on its own.
I've saved the best for last, which is
why there is a helpful little sparkle
emoji next to it. Number seven, tie
every point back to the product.
You need to reinforce to the answer
engine why your product is relevant to
the topic so that it knows to recommend
you. If you simply give answer engines
educational and original insights
without tying those insights back to
your product, you are making them
smarter without reinforcing why that
concept is relevant to your brand and
therefore why you should get a
recommendation. So this part is key. Tie
every point back to the product. And
that means you're going to have a
one-two punch. Here is some good content
to get your attention. And here's why
that topic is deeply connected to what
my product can do. There should be a
product reference in every paragraph,
every other paragraph. It's that many
product references.
Okay, you can now optimize a page for
answer engines.
Last but not least, let's talk about a
very important part of AEO. How you
build authority online, your new
authority, and this is your off-site strategy.
strategy.
When this was a Google game, backlinks
were the most important way to grow your
site authority. But answer engines don't
really care about backlinks. They care
about mentions. Is your brand getting
positive shoutouts on Reddit, LinkedIn,
review sites?
And here is how HubSpot has generated
many, many thousands of mentions since
we began this strategy.
We started with the websites that answer
engines are already linking to in
responses for our priority questions,
especially the ones where HubSpot isn't
visible but should be. And you can use
Xfunnel for this. We can see here in
this example that HubSpot is mentioned
in a Tavo Q article that is being cited
14 times. Fantastic. But not in a useful
AI article that also has 14 citations.
And this is going to make it a priority
for our outreach team.
Getting positive brand mentions in cited
sources looks a lot like classic link
building. Publishing guest posts,
getting featured in roundups, asking if
the website will update an existing
article to include you. However, unlike
in the old days, we don't care whether
that mention is hyperl. We do care
immensely how HubSpot products are
described. The mention needs to
integrate our messaging so that answer
engines learn our positioning.
The third step is leveraging social
proof. That means review platforms like
G2, Capterara, Trust Radius. They are
incredibly sticky in answer engine
responses. Answer engines love UGC or
userenerated content. And so increasing
your presence on these platforms,
especially with specific feature praise,
can really shape how AI describes your product.
product.
Last, but certainly not least, seed
human first channels by partnering with
YouTubers, newsletters, LinkedIn thought
leaders so that their content includes
your product in context. And this is
really a two for one. You can meet
buyers where they are and where AI is
looking and training. And the goal is
that when AI models look around to
decide what to recommend, they keep
bumping into your brand in authoritative
high context mentions.
Now, this is kind of misleading because
I put this all in one slide and I've
made it look quick and easy, but make no
mistake, it is a lot of work. And as I
mentioned, at HubSpot, we have a
full-time team that is just dedicated to
that effort, which I know most companies
do not have. So, if you can only pick
one place to start, identifying AI
citations that you can influence is the
lowest hanging fruit for most companies.
That was your threepart AEO strategy.
Now, let's talk about how to see whether
it works. Measuring success
in traditional SEO. Then you would track
rankings, organic sessions,
click-through rate. We're still tracking
all of these things, but we have a brand
new search scorecard for the AI era.
In this world, we are also tracking AI
visibility, which looks at for each
query that we care about on the engines
that we care about, are we recommended?
We're looking at AI share of voice of
all the AI responses that mention a
solution. How often is HubSpot named
versus our competitors? And this is
really important. This tells us whether
we are gaining or losing recommendation
share over time. In September and in
October, Chat GBT turned down the dial
on brand visibility, which means that
brands across the board are being
recommended less. Now, if you're just
looking at your visibility, you would
freak out and you would say, whoa, what
did we just do? What went wrong? If you
look at your visibility coupled with
your share of voice, you know that,
okay, our share of voice has stayed the
same. Our visibility has gone down. It's
something that the platform has done.
The third metric that you're going to
look at is AI citations. How often is AI
using your website content to answer questions?
questions?
Now, visibility is the northstar metric
because visibility is what's going to
ultimately influence that conversion,
whether someone goes to your website,
whether it's right away or 5 days later
or 15 days later. But AI citations are
important too because what we have seen
at HubSpot is that when an answer engine
cites our content versus someone else's,
we are recommended more positively and
we are higher up in that citation, which
makes a lot of sense. We're always going
to say the nicest things about
ourselves. But if you care about owning
the source, which you do, you care about citations.
citations.
And then last but not least, we have AI
referral demand. Are you seeing visits,
conversions, and sales from people who
first encountered you in AI responses?
And I will be honest, this is hard to
track because most people are not going
directly from an LLM to your website,
which is why I recommend implementing
something like a post-purchase checkout
survey. A really simple back to basics.
How did you hear about us? Answer
engines are also now a referral in
HubSpot as of last quarter.
These are all illustrative stats, but I
do want to share some very legit ones.
Since we launched our AO strategy at
HubSpot, we have attained the highest
share of voice in our space. Citations
have improved 433%
and demand has gone up by nearly 2,000%.
And I very much believe that you can get
these kind of results, too. So by the
time we hit January 2028,
your brand will be dominating in answer
engines as they become the predominant
source of demand.
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.
