Gabe: I knew, for instance, that
influencer content, um, not just UGC,
but like really people that are good
behind the camera or the phone and that
have a way of like attracting attention
and getting people to like, like, and,
you know, engage in content that works.
It's really about, especially
in paid social creative hits,
a lot of the creatives you
try, it's just not gonna work.
And it's only the small percentage of
creatives, like everything is the Pato
principle everywhere all the time.
So too with creatives and, and scaling ua.
We, when I came in, I tested,
you know, some creative that
they had in the account already.
We did contracts with influencers,
I saw that a couple of them had
really good, um, virality scale.
And so that was that early signal,
like, okay, there's something here,
and what if we chuck that into
advertising and we see how that goes.
And, you know, we tried a few of them,
some of them didn't work, and then
we kept trying and I kept looking
for those hits, the things that, like
a day.
see that meta is starting to scale
this up and it loves it and you're
getting great, um, CPIs great, cts great
cost per trials and you can see that?
like this is really strong performance.
so then I just started opening up the
budgets and really expanding out the
audience targeting and really like getting
more surface area for these creative,
this strong creative to get out there.
My name is Jacob Rufin,
CEO, and founder of Botsy.
Today on the Price Power Podcast,
we're talking with Gabe Kochi.
Gabe has been a leader in the
mobile marketing world for years.
He founded an agency called Inci.
Pia.
He co-authored the Advanced App Store
Optimization Handbook, worked with massive
brands, including Walmart, Coinbase,
Duolingo, and more, and is now CEO of the
number one sign language app in the world.
Lingo.
We talk about his journey to CEO of
lingo, the importance of creative
hits, experiment wins, how to
scale meta campaigns and more.
Let's dive in.
I.
Jacob: Hey, Gabe, how's it going?
Uh, welcome to the podcast.
Happy to have you here.
Gabe: Yeah.
Thanks Jacob.
It's, uh, it's good to be here.
I, just hanging out here in Vienna
and, uh, enjoying the weather here
and having some tea in the afternoon.
Jacob: Awesome, awesome.
Well, yeah, again, really
appreciate, uh, uh, joining.
Um, maybe you give a little quick
background on yourself for, for everybody.
Gabe: So, hey everyone.
I've been in, uh, digital
marketing since 2012 and mobile.
Um.
Apps and marketing since about 2015 or so.
Um, I co-founded an ed agency called
Inci Pia, and at that time also
co-wrote the advanced a SL book
with more at Stan from feature.
That was a great time.
I don't know how the heck we wrote
a book while also running and
scaling agencies and anyway, uh.
A little through that experience,
I started to actually get burned
out, so my co-founder and I decided
to, to shut down the agency.
Um, I went and spent a little bit of
time at Lawrence's agency, and then I
career pivoted and trained to be a life
and career coach, and then career pivoted
again when I moved out to Austria to
move in with my girlfriend in Vienna.
And there I came back to marketing.
I got a job as a marketer at this
sign language learning app called lno.
about a year in or so.
Then I was also asked to step
in to lead the company as CEO.
So that's been fun.
I've been, uh, CEO of lingo
learning and teaching sign languages
for the last year and a half.
Jacob: Yeah, it's so cool.
Link VNAs such a cool product.
Uh, uh, so, and I, it's quite
an interesting journey, uh,
pivoting one way, pivoting back.
I remember, um, early on when Apple search
ads were first released, uh, you were
writing content for incipient and like.
Essentially, you taught me how to
run Apple search ads like your, your,
your, your content was stellar there.
And I think that's when I first like,
I don't know if I, I had knew and sent
before that, but I think I, I, I came
across you then and so I think that's
when I first kind of, kind of heard your
name and, uh, so it's like awesome to k
kind of have you on the podcast today.
A after all that, you know,
one of the early, uh, uh,
early leaders in mobile growth.
Gabe: Yeah, well, it, it was,
it was a fun time and it was
everyone sharing everything, just
experimenting and nerding out on that.
Now I'm doing the same thing
with AI now, so it's another.
Another frontier to get into.
Of course, of course everyone
knows about ai and that's not like
a, a thing secret or interesting.
But, yeah, I've been putting together
this AI stack and it, it throws back to
the A A SO stack that we did, years ago.
So I'm, I'm back to the nerding out
and the content production again.
So it's good to be here.
Jacob: Nice.
I love it.
I love it.
Yeah.
Yeah.
We'll, we'll have to dive into,
uh, um, the AI stack a little bit.
Uh, really, really curious to
hear, hear more about that.
Um, yeah, I guess, um, do you wanna
start like also just sharing a
little bit, uh, about lingo, uh.
Gabe: Yeah.
Yeah, I'd love to lingo.
I'm a non founder, CEO.
Um, so again, I, I joined the company
two years ago, became CEOA year and
a half ago, but lingo, was, it's an
Austrian company, we were started in
2018 by the co-founders, Nico Mattias.
Who were doing their civil service
in Austria, so you have to volunteer
or go do military service for
a little bit after high school.
And they were working, um, with deaf
patients and they realized there wasn't an
easy way to learn sign languages online.
So they decided to co-found, uh,
lno and build the course themselves.
and.
Over the years, they added, they started
from Austrian Sign language and a website.
They went to an app.
They added American Sign Language,
British sign language, and so right
there is an interesting fact that
most people probably don't know.
There's actually.
Over 200 and maybe even over 300
different sign languages in the world.
There's not just one.
And, uh, so it's, it's
very interesting to learn.
And they're also quite different, like
British sign language, you spell the
alphabet with two hands and American
sign language, as well as Austrian
sign language you spell with one hand.
there's some interesting differences,
um, between all of them, and
it's a lot of fun to learn.
And, uh.
we now have almost 4 million learners.
Um.
Mostly in the US we're a team
of about 27 or so people.
Um, there's, Yeah.
there's a mix of hearing hard, of hearing
deaf colleagues on our team across
the US and Europe, and it's super fun.
Our learners.
Love learning with as well.
We actually have more than half of
the people that are learning with us
that I just have a general interest.
They'd actually, some of them don't even
know anyone who's deaf or who signs.
Um, but it's, it's great to, to teach
them more about not just sign languages,
but deaf culture and deaf people as well.
So it's, it's really enjoyable.
Jacob: Yeah.
That's awesome.
It is.
I've always found in my career
when I'm, I'm working on products
that, that actually help people.
All, all the work is so
much more fulfilling.
Uh, e everything gets a little
easier just when you hear reviews and
hear feedback from people that it's
like, you know, learning sign legs.
We're also potentially like
opening up communication with
children or parents that.
They couldn't speak with before.
And so it's pretty cool.
Um, and yeah, like I, I,
I, I learned about that.
Yeah.
Every, every country has
different sign language.
It makes sense, right?
You know, every country in the
world speaks a different language,
you know, I think as, as you know,
uh, um, Americans, we go, oh.
Stein language is the same everywhere.
It's like, no, we have to, you
know, uh, it, it makes sense.
It's different.
So that, that's awesome.
Um, and so, you know, you, you, you kind
of touched upon this in your little intro.
Um, you went, you started, you
joined Linko and, and now you're CEO.
Um, how does, uh, how does someone join a
company and then get, uh, promoted to CEO?
Gabe: Uh, it, it wasn't something that
I, um, uh, planned from the start.
It was, it was more of
a serendipitous thing.
I, I suppose that's how a
lot of my life has worked.
I, I met my girlfriend, uh,
serendipitously in the streets of
New York or the subways rather.
so that's a story for a different time.
But, um, Yeah, I joined and, um.
I moved to Austria.
I was looking for a place to actually
just freelance, 'cause we were
gonna go travel Asia for a summer.
And then I was freelancing and really
the, the, a lot of the pieces that lingo
to really scale up a lot in a year.
We went from a million learners,
um, a year and a half ago.
Um.
about almost 4 million now.
So a lot of that growth came from one,
just a great product, a great course and
content that was there before I came that,
that everybody had built out before the
product team, the learning community team.
Um, and
then, you know, there's, uh, actually
influencers were one of the earliest ways
that lingo really started to get traction.
And so,
Having, I, I think like there's a
saying out there of like, one of the
fastest ways to kill a bad product is
to just shove money at it and marketing.
Um, but one of the, you know, things
that's great about having a good
product that people like that has good
conversion rates, good retention rates.
Um, that, people leave, you know,
good ratings about is that it's,
once you find the right creatives,
then you can really scale things out.
So we had, um, you know, we have deaf
teachers, we have deaf influencers
and content creators, and so our, our
old head of social and partnerships,
she had had like a lot of great
influencers that I then came in and.
Took that content and then realized,
okay, we can run meta ads, um, we can
scale things up on that side of things.
And so we really opened that up.
We had a TikTok afterwards.
We opened up Google App campaigns more.
so a lot of, a lot of what would
happen was, you know, like using
the pieces that were already there
uh, taking the last mile excellence,
um, to, to make them work and scale.
And so then I, um.
Joined full-time, started to build out
a growth team as, as head of growth.
um, the co-founders, um, actually I, I
think like credit to them, they realized
that their skills and what they enjoyed
doing was more of the early stage, so
the last seven years at that point.
And so they realized, um, that they
thought it was time for them to actually
look to find different leadership that's.
You know, had experience in managing
this was actually their, their first or
almost first thing out of college, so they
didn't have previous experience scaling
it out and, you know, it, it worked.
Um, but.
At the point that we were at, then they
realized, okay, lingo probably needs
new leadership, um, new direction.
Um, and so we built out a
team of a leadership team of
heads, and I was one of those.
And then, uh, they said, okay, you've,
you know, been CEO before you also led
a team, you know, but other places,
you know, how to like build and scale
products and, and also companies.
And so they, um, we did a
little kind of like, um.
launch period where I started to take
over the reins and then, uh, we had a
transition period and then things have
have been great so far, but it was right.
Experience, skillset, right time, um,
serendipitous time, uh, rather, and also
just a, a reflection on what does the
business and the company need at this
point, and a willingness to, to change.
Jacob: Yeah, that, that makes sense.
I, I, I know it's been a couple years now,
but congrats again on, on, uh, uh, you
know, being CEO it is very cool to see.
I, I was very excited when I saw that,
uh, little LinkedIn announcement on the.
My feet.
Gabe: Yeah.
Jacob: Um, uh, yeah, and, and yeah, I
think so much of, you know, we can, we
can look back at our careers and go, oh
yeah, that happened because I did this
and this, but it's, you know, it, it's
all sometimes coincidence, serendipity,
sometimes just being open to new
opportunities and kind of keeping your
eyes open to kind of what's out there.
And so I think that makes a lot of sense.
You, you, you touched upon one point, um.
That they, they had good foundations
for the product, but also that they
had a lot of great, um, maybe it
was organic influencer marketing
content like, uh, uh, available.
So, you know, I think this is true in
a lot of companies that there's, um.
The seeds for great growth are there, but
sometimes we don't know what those are.
Like.
When you came in, how did you
know that was the right strategy?
How did you know, like did you have any
early signals that like, that influencer
content existed was gonna be valuable
or kind of what, uh, uh, did you just
have to take a bet and it worked out?
How, how did you think about that?
Gabe: Yeah.
Um, I was actually having a
conversation before this with, um, a,
a developer actually you introduced
me to who, you know, super smart,
built something that, that has a
great level of traction right now.
Um, but he was, he was saying,
he was like, you know, he
was just kind of peppering me
with questions and I loved it.
We were having this discussion and
I lost track of time and to go to
another meeting, but he was like,
you know, sorry, I'm asking so many
questions, but like, this is the first
time I get a chance to just like.
Pick someone's brain and growth
that like, has seen things,
that have tried things out.
And I was telling 'em all about you know,
what, what I learned at Lingo and before,
but to your, to your question, um, I was
able to recognize which pieces were there
that were important and what to try and
how, where to have confidence and what to
look for because I had run and scaled, um.
Campaigns with Coinbase, uh, with,
with, um, Walmart, with, um, people
fund, just like all these other,
um, companies back in the day.
And so I knew, for instance, that
influencer content, um, not just UGC,
but like really people that are good
behind the camera or the phone and that
have a way of like attracting attention
and getting people to like, like, and,
you know, engage in content that works.
And that we needed to find it.
It's really about, especially
in paid social creative hits,
a lot of the creatives you
try, it's just not gonna work.
And it's only the small percentage of
creatives, like everything is the Pato
principle everywhere all the time.
So too with creatives and, and scaling ua.
We, when I came in, I tested,
you know, some creative that
they had in the account already.
We tried making some, like
some static ads, some graphic
design ads, motion graphics.
I tried making a few videos myself.
Um, we had our, some of our teachers
who had videos and those were
like one of the earliest hits.
Um, one of our teachers who
was doing the, the colors, um,
in, uh, American sign Language.
Um, but then I saw that the.
Cr, the, the influencer videos
actually wasn't fully organic, or it
wasn't, most of it wasn't organic.
We did contracts with influencers,
I saw that a couple of them had
really good, um, virality scale.
And so that was that early signal,
like, okay, there's something here,
and what if we chuck that into
advertising and we see how that goes.
And, you know, we tried a few of them,
some of them didn't work, and then
we kept trying and I kept looking
for those hits, the things that, like
a day.
see that meta is starting to scale
this up and it loves it and you're
getting great, um, CPIs great, cts great
cost per trials and you can see that?
like this is really strong performance.
so then I just started opening up the
budgets and really expanding out the
audience targeting and really like getting
more surface area for these creative,
this strong creative to get out there.
And on the side we started up a
pipeline where we started to, to
work with more and more creators
and pull in more influencers.
And so constant testing,
um, was, was important.
But um, Yeah, it was realizing
that creative hits what drive paid
social, um, that there were some.
Promising creatives in the, in the mix
overall that hadn't been deployed from,
uh, Instagram organic to, um, Instagram
advertising and to just when it started
to work, have the, the guts to, uh, 10 x
the budget and the co-founders as well.
Like, I checked in with them,
you know, props to them.
They're like, I'm like, okay, you
know, this is the time to press.
The, budget, when it's working, go all in.
So can I, can I raise the budgets?
You know, 10 x in like a short
period of time, do we have the
cash flows to support this?
And they said, yep, we're good.
So, so then go in.
And that's, another thing like.
It's temporary.
Creative hits are temporary and
you get this excellent performance
period, but then it always at some
point saturates and comes down.
So you have to learn how to like
navigate these periods where you have it.
Press in, press your advantage, and when
you don't have it and you get into a
drought, pull back, realize that you've
just gotta work to get another hit.
And you're go, you're
gonna feel like a failure.
You're gonna feel like, ah.
You know, shit like why isn't it working?
Nothing's working.
Our performance is so bad.
that's the dynamics of modern paid social.
You just gotta get a strong
enough hit, um, that it works.
So.
Jacob: Yeah, that's, uh, amazing insight.
Uh, I think, I think a lot of people
find that valuable to just understand
that the mindset, but also, you
know, looking for those early gems.
And, and so like on that organic
social post, was it you looking at,
uh, um, a post going this, look,
this is pretty good, I like this.
Or was it like, Hey, this got a lot
of like, people were sharing this and
commenting, like seeing those early signs
that that kind of clued you in there?
Gabe: Yeah, it was the engagement.
Um, it was one just like views.
Um.
Absolutely.
It was make mostly views.
My, my view on, I dunno if they're
controversial, okay, using the search
ad, CPA goal or now whatever it's
called, CPA cap, is a controversial
thing that I do to this day.
Always use the CPA goal even
on your exact match campaigns.
Anyway, rant over.
Um, another one is like, when it comes
to creative hits, I just look for spend.
I just look for, for scale, because.
You're optimizing towards performance,
so you're not gonna let a, a
creative run away and spend a lot
of money if it's bad performance.
'cause you know, you, you, you're
gonna go outta business doing that.
So, yes, look for CTR.
Yes, look for CPI.
Yes.
Maybe look at CPMs, look at per,
you know, CPAs, but really like.
If a creative is scaling up,
that's how we define hits.
A hit level D is a 10,000
or more lifetime spend.
A hit.
Level C is a 25,000 or more lifetime spend
hit level B is a hundred thousand hit
level A is two 50,000 and hit level S.
It's like the Japanese games, like SSS,
super special, like you got the top score.
That's a million or more.
And so then I I looked over
at the, the influencer videos
that just had a lot of views.
And most of them also had
correspondingly a decent number
of likes, um, and engagement.
But it was scale.
Just took inventory over all the different
posts that we had done and, and picked a
few out, and they, most of them worked.
Jacob: And I think, you know, we
have all these, uh, we're, we're,
you know, have these, you know, CPA
goals and this, but at the end of
the day, we're always, we're paying.
It's always CPM, right?
We're just paying for
views, uh, and we hope.
That things convert.
And so I think that like it always gets
lost really sometimes, uh, uh, when, when
we're thinking about like spend there.
So that makes sense.
Um.
The other thing I wanted to ask.
Yeah.
With, with those kind of creative wins
and, and maybe creative lulls sometimes,
um, how do you minimize, uh, the lulls?
Do you have, um, and maybe this is
more kind of technical campaign set up,
uh, do you typically use a champion.
Uh, ad campaign ad group where 75%
of the budget goes there and you're
constantly trying to test, uh,
and get a new potential champion.
And then how do you, how do you integrate
that, um, potentially new winner into
the high performing ad set without
actually maybe harming your ad spend,
uh, uh, to that champion creative?
And, and then how do you
deal with all the like.
Sorry, I'm throwing a few questions at
once to you, but then how do you deal
with the, um, uh, just, you know, meta
algorithm of like, you put this in the,
the top performing ad set with most of
the budget or ad or ad group, uh, and
then like, does it not get reached or,
or, you know, I think there's like,
there's so many nuances of how you
kind of find those new creative hits
and then integrate them into your,
your top performing, uh, top spenders.
Gabe: Yeah.
I'll, I'll, I'll pick this,
the, the layers apart here.
Um,
Jacob: Yeah.
Gabe: with.
So one, I don't, I don't have a
clear strategy that like cleanly
manages hit transition to hit.
Like that's actually
Jacob: Yeah.
Gabe: right now.
Our, um, growth lead and I were
discussing like, how do we introduce
a new creative that seems like it has
potential into the mix where a current.
Favorite from Meta Creative Hit
is still hogging the budget.
Like, but it's, it's in the decline.
We know it's saturating.
So how do you do this
Indiana Jones kind of thing.
We're still trying to figure
it out, but to start like what
happens in a creative lull period.
What I've found, and this
is again going back to like.
Before SKA network, but
still after SKA network.
Um, you just have to have, you
just have to keep going until you
have a strong creative hit there.
I've found that there is nothing you
can do to make performance better if
you do not have a strong creative hit.
just that the reality right now, it's not,
you know, the same as search, but when
it comes to paid social, when it comes
to these like social, viral, organic.
like these platforms are
predicated on virality.
um, you'll be beating your head
against the wall trying to figure
out what, where's the next creative.
And our old one is
getting worse and worse.
Um, but you'll still run your old hits.
Like actually we have hit that's
still more or less running
two and a half years later.
Um, it's not as good as it used to
be, but it's still the best out of
the bunch if we don't have a new hit.
So.
Hoping that a new creative that like
looks okay in testing is gonna start
spending, I found that it doesn't happen.
It's just your hits are always
gonna be driving the majority
of spend, even if they're old.
So in the period of lull, you.
You can, you can try to reactivate old
hits that you maybe paused at one point.
can, um, do iterations, you know,
especially hook swaps on the creatives
to extend the lifetime of a hit.
But eventually they,
they're all gonna saturate.
so you just, you go and work with those
creators, those sources that produce the
hits, like we call them our hit bakers.
And we, like, we keep working
with those that are hit makers.
You keep, you know, marketing, you,
you keep doing a thing until it.
Exhaust and it stops working,
then you go on to something else.
Um, and it comes to, to, to the
testing, so we, we have a testing
campaign that's, you know, always on.
We're putting the new creatives into
it and we just let them fight it out.
For the most part.
It's not like.
There's a totally specific
scientific formula.
It's just like you dump 'em in, you try
to get you, you pause 'em after a hundred,
150 in spend, you evaluate performance.
If it's good enough, you flip 'em into
the maybe scaling or core campaign.
At that point.
It can be tricky, um, where
they go after testing.
But if a creative doesn't spend.
Kill it, move on.
Don't, you don't have to try
to force creative to every, or
spend to every single creative.
The hits will reveal
themselves 90% of the time.
So unless you feel really confident
about a new creative that's coming in
and it's not getting spent, you might
restart it, put it in a separate ad set,
but don't worry too much about that.
Then when you get a hit, um, then
we've tried a few things, like if you
have an old hit that's still running.
It is hard if you put
it in the same ad set.
The old hit is the alpha and it's,
it's not gonna allow any spend to go
to the new one, even if it's saturated.
what happens then?
We've tried, um, we actually
were working with an agency for
a little bit headlight, and they
brought this like scaling campaign.
There was an intermediary.
So you graduated from the creative
testing, you move it to scaling
and you try to, do, you try to,
you use separate ad sets to like
force spend to the new ones.
You try to get it to like a thousand in
spend and then try hope at that point that
meta realizes, oh, this is actually pretty
good, then it starts to pick it up more.
Then you could put it
in the core campaign.
But we still, it's, it still
doesn't work most of the time.
we're thinking like, okay, maybe
there's a core campaign where
there's a one ad set that has.
core hit in it, you set up another
one and they're both in core and yeah.
You like use ad set level budget and
maybe like over time decrease the
budget of the hits and increase the
budget of the newcomer, assuming the
performance continues to look good.
thing that we don't know
is like, I'd rather not,
I don't know.
It's, it's up to like style and
how you, how you do it, but.
not like turn off a hit
Turn on the new one, but
sometimes maybe that is it.
You just have to do that for,
you know, a week or two and just
try to force it to have a chance.
Um, and if it doesn't, a
performance gets way worse.
Okay.
Maybe you go back, you try something else,
but, um, yeah, don't have answers on that.
iOS, you know, makes it harder because
you have the delay of performance.
You don't have the true ad level.
Attribution, I guess Meta, a EM, they
have their own like, um, estimations
for that, so it's better than it used
to be, but yeah, it's tough still.
Jacob: I think.
One, like amazing tactical insights on, on
how you kind of think about these things.
I, I appreciate you sharing those.
And I think the, the major insight,
Al also is that, um, there aren't
hard and fast rules, right?
Uh, um.
You have to experiment and
iterate and, and, uh, um, it's
not like this happens, do this.
Uh, there's, it's a science,
but there's, there's also a good
bit of art, uh, uh, in it all.
Yeah.
It's super interesting.
You know, I'm not really on the, um, I
focus more on kind of product growth,
monetization, retention, so like, uh, uh,
learning kind of all the, the acquisition
insights is, is really interesting to me.
And I'm sure other people
will love that too.
Um, cool.
Um.
So, um, I wanted to also just ask
like broadly, you know, I think, uh,
it, it seems like paid acquisition,
um, you know, is probably driving
a lot of the growth at Linko.
Uh, um, are there other key initiatives
that you're working on right now that
are driving growth, you know, within paid
acquisition or, or kind of more broadly?
Gabe: Yeah.
Um, paid, paid marketing is still
the, the lion's share of our growth.
Um, but we, we have been working
a lot on like growth experiments,
so that's like the onboarding, the
monetization, um, side of things.
And in That, we've seen some,
some great improvements there.
So that's like an ongoing, um.
Uh, swim lane, I guess of, of work.
And we're also continuing to, to
experiment with core products.
You know, our head of product,
Elena is really, uh, trying to think
like strategically about product.
Like, you know, what are the, what
are the bets that we wanna take?
Um, how do we continue ex improving
and expanding on these experience and
adding new features and taking bets and
trying things where we don't know if
it's going to work, in the end or not.
But not only relying on.
UA and growth experimentation as
there's, you know, there's saturation
points to those where at a cer certain
point you can't squeeze anything else.
Um, out when you're hyper efficient, of
course always be testing, but, um, don't
only be relying on one or another pillar.
Um, and, uh, we're also trying new
kinds of like products and offerings.
So our core offering is a, um.
Is a subscription to our, um,
course in the app or the website.
you know, we're also, um, exploring
live learning events where people
can get on a zoom, uh, call.
And there's um, uh, a deaf
instructor or a, uh, mentor who
will like guide a sign practice.
So that's something we're trying as well.
We're revamping our
social organic strategy.
we're working on building a learning
community in Facebook groups.
we're doing some continuing, do some
like partnerships here or there.
Um, but yeah, and there's some other
interesting things, uh, in, in the works
here, um, in the, the beta side of things.
And we'll see where they go
before we share too much.
But try trying not to be too reliant
on one or another lever of growth.
Jacob: That, that makes sense.
And I think once you reach a certain level
of scale, there's diminishing returns,
you've gotta spread out a little more
to kind of keep on, keep on growing.
There are, are those, I'm assuming
those kind of a, a live, um,
training or teaching programs
or like a higher price point?
Gabe: Yeah.
Yeah.
We're, we're experimenting right now to
like a one-time purchase actually for now.
Um, but we're trying, like,
okay, you get one free and then
you can, uh, try another one.
Or maybe there's a sub
subscription at some point.
maybe we bake it all into like
one different price point?
subscription where you get, you
know, the core course and uh, live
learning events and another things.
So yeah, let's, let's see where the
experimentation is, is definitely,
uh, where we're at with that one.
I.
Jacob: Nice.
Very cool.
Very cool.
Yeah.
Uh, something that's become very.
Very popular.
And, and, and maybe you all
had a web platform for a while.
Um, but, but how is, um, are, are
you guys, uh, using much kind of
web to app conversion funnels?
Uh, and it remind me, do you
have both web and uh, uh, mobile
platforms like in terms of products?
Gabe: Yeah, we're already,
we're a React native app.
Um, and so we have.
Um, web app or web, iOS, Android, and,
um, yeah, we're actually trying to, to
set up some tests and, and see whether
it makes sense for us to, I mean, our,
is yes, we can, if we push more folks
to the, from app to web onboarding our
and payment flows that, that's, oops.
going to be the, um.
The optimal thing for us, you know, 'cause
our, of course the, it's, it's very shiny,
the share of revenue, net revenue that
you get from web payment, um, providers.
But, you know, we also see, you
know, we're setting up some,
like, doing the research still
and, and setting up some tests.
But we also see that, you know, it's not
as, it's only iOS US, for instance, and.
When you send people out of
the store, you have drop off.
Um, you can't use like a, an
in-app web model, like pop-up
model with Apple Pay for instance.
You have to send them out of, the app.
And so there's, sounds shiny.
It is shiny, but there's.
The conversion rate gets cut
down by these points of friction
that still get introduced.
So, um, we wanna move in that direction.
Um, but we've also seen other people
experimenting and they said, or
some folks have identified like,
well, it's, it's kind of a net net.
It's not a, a grand slam, to, to do that.
But again, always be
testing, draw things out.
Jacob: Yep.
Yep.
I, I, and I think my.
My hypotheses for, for kinda what
app to web, uh, conversion will look
like in the future is that, you know,
certain user bases, certain segments
and certain use cases will work well.
Where, um, I think that I, I've heard that
sometimes potentially older audiences, uh,
can convert better because they're, um.
Less, uh, uh, less frustrated with maybe
digital friction where, where younger
audiences go, where are you sending me?
Uh, uh, where older audiences may be
pay atten less attention to that kind of
like, or understand less the difference
between the app and the web experience.
Uh, and so I think, um, that will, will
probably, depending on your audience,
depending on your user base, that certain
segments you may learn, convert well,
and you'll take advantage of that.
I also think that, um.
Essentially promotions, um,
where you're offering a discount.
Because of that discount use, that extra
motivation and that extra motivation
kind of overrides that friction.
Uh, um.
From that experience, sending them
out to web and they're still excited.
Oh, well I get a discount.
That's fine.
You'd have a greater revenue
share, which is even more important
when you're discounting, right?
Uh, uh, to kind of take
home more of that revenue.
And so I think, um, that, like long-term,
that's probably what we'll see.
And maybe some apps will have.
A user base that it really makes sense
for that they, they send everybody.
But I, but I think it's, and generally
it's probably a later term strategy
where, you know, for, for most
apps, if you're under a million and
you're, you're paying 15%, uh, uh,
to Apple, don't even bother, right?
Like it, the, the you, you really
likely will see a lower conversion rate.
The math is that, is the conversion
rate drop, uh, uh, low enough.
The increased take home revenue,
uh, uh, makes up for it.
And so when you're at 30%, yeah,
it can start to make sense, right?
Uh, but, but again, it's probably not
just slap it, slap it in your app, and
apply it broadly across, across the board.
And so I, I think we'll.
Who knows how long, maybe, maybe
this will never come true, but like
we'll start to learn which use cases.
It, it starts to really make sense for.
And so that's, uh, that's my prediction.
Putting on recording it.
Well, you know, you can hold me to it in,
in a year to see what actually happens.
Gabe: Well, who, who knows?
Also, the, the whole space
changes so fast and, Yeah.
Jacob: Yeah.
And so, so,
Gabe: It's like, you
Jacob: right, right.
Gabe: Google came out and said,
okay, you can do whatever you want.
And then a judge was
like, wait, wait, wait.
Not so fast.
Jacob: Yep.
Yep.
Um, and, and then so that's, so
you're texting out app to web.
We'll see what happens there.
And, and so, um, do you, uh, do you
send paid traffic for web to app too,
to start on web conversion funnels?
Like do you have web subscriptions
where, where, uh, people
start on web as web right now?
Gabe: Uh, we do.
It's, it's a very small share of,
um, of overall new users acquired.
Um, and, and before.
Actually before we started to kind of,
uh, slow down on app advertising, uh,
scale and need to then look for other
areas to improve efficiency, um, web
wasn't even a, a consideration and
before, like all this happened with,
um, web payment flow possibilities.
And so, you know, we, we don't
have a web trial for instance.
There's some other areas where we could
like improve the web conversion rate.
Um, but right now?
it's, you know, it's much lower,
um, uh, than the app flow.
And so, you know, we've tried meta web
campaigns, we have some yeah, Google web
campaigns, um, running, but it's, uh,
until we make some further improvements,
it, yeah, it, it doesn't make sense
to, to shift a, a meaningful amount of
budget, um, into that, um, from app.
Jacob: Makes sense.
Makes sense.
Um, and so, you know, you, you've been
working in paid UA for years and years.
Like for someone just getting
started, like maybe they have,
uh, um, you know, a, a newer app.
They, they have some traction,
they have some organic traction.
Maybe they're doing a little bit of paid.
But for someone really getting started
with paid acquisition for your app, do you
have any like high level guidance tips?
Where to start, how to think about things?
Gabe: Hmm.
Um, well, one is definitely just Yeah,
experimenting, trying things out.
Um,
really like the, but.
Being realistic about which to focus on.
If you're an Android app, Google,
Google app campaigns, they're annoying.
They, they suck because you can't
get the, you can't target keywords.
You have all these limitations.
You can't actually control the ads.
Yeah.
But, but you get ads in the
play store, that's valuable.
apple search has, uh, campaigns,
if you're on iOS for sure.
Those are gonna be pretty easy to learn.
Um, and then meta, you know.
Maybe TikTok if, if in your
niche it really makes sense.
But start with meta and meta's.
Like, you know, for, for anyone who's
spending at scale, like meta's 50
to, to 90% of your, of your spend.
So just get really good at, at meta,
like go on LinkedIn, you know, just
consume all this content from people
that are writing about meta tips
and, and tricks and different things.
Um, and start trying them out yourself.
um.
mean, ex explore with, with ai
Personally, I haven't, I haven't been
able to, we haven't been able to make,
you know, it's also different for us.
Like you, we can't trust, for
instance, uh, chat GBT to create
a video of people signing.
Um.
That, that doesn't work for us, um, or
for our, you know, brand or our customers.
um, maybe try some, some AI ads.
Runway, see what you get.
But also creators.
Creators and influencers,
micro influencers.
Um, go out there and, and try to find
some people that you think could represent
your product well try influencer campaigns
with them as well as, um, try using their
content in meta and see how that goes.
and yeah, in addition to like consuming
content, just talk to people, send
people up, mess, like LinkedIn messages,
just randomly reach out to people,
go to events, meetups, conferences.
Like still, you know, like you're
referencing when, when I was, um,
writing it incipient, like, people
like to share, you know, their, their
tips and tricks and especially with
folks coming up in the industry.
Uh, it's great.
It's, it's fun for both sides, um,
to knowledge exchange and, um, and
then once you get, you know, to a
certain point you'll exchange knowledge
with, with folks coming up too, so,
Jacob: Yeah, that's something I love
about the mobile app industry of how open
and, and kind of like, uh, open to go
about sharing and supporting and helping.
It's a, it's a, um, a
good industry relatively,
Gabe: yeah.
Yeah.
Jacob: yeah.
Um.
Cool.
So, um, I, I also wanted to
maybe pivoting a little bit.
Um, I wanted to ask you about
your AI stack for startups.
Um, I, I saw, I, I saw you putting
that together and I know this is, you
know, kind of a, the stack element
as a nod to, you know, Andy Carve and
Morris, uh, uh, with the, the mobile
growth stack and the a SO stack.
Uh, yeah.
Can you tell me more about like.
You know, why'd you wanna do that?
And, uh, with everything else you
got going on and, and like how you
constructed and organized all that, I.
Gabe: this came from pure curiosity
and like, like what, what ended up,
I was writing all these blog posts
and, and why Moritz and I wrote
a book and, and came up with it.
The a SO stack was like, we just want
to try to concept, understand these
concepts and scaffold and organize
them in a way that it's like, okay.
For this, for, for keyword optimization,
here's the stuff that you're doing
for conversion rate optimization.
Here's the stuff you're doing.
Other tools and, and techniques analytics.
Okay.
so for me it's just like, I'm trying
to, to wade into the tsunami of e AI
concepts and things that I'm like, I'm
feeling like this tall again, which is
cool, but overwhelming at the same time.
So for me, I was just like, you know, I've
had, I don't know how many conversations
with Gemini and Grok and, uh, and Claude
and chat bt trying to just understand
what's the, what's the landscape.
AI is just so incredibly broad and so
just at least focusing on, generative
large language model, ai, chat side
of things or the, um, or image,
you know, creation or that, and.
I just tried to ask like, what are
the concepts that I need to know
about in order to operate in the
space and know how chatbots work?
How do they break down?
Um, how do you them?
Um, if you were to bring them into your
product, what do you need to learn about?
So like, okay, prompt.
That's, that's, that's the
simplest thing, uh, for like
the, for using AI as a product.
You tell it.
I guess even as a user, you, you, you
prompt it or you ask it something,
then, you know, there's top uh, top K.
So like, which number of responses
should it be, um, responding to you with?
What are some hallucinations?
You know, I, I have now at this 0.3
stacks, and they're again,
just like scaffolds.
Or like ways to organize
your, your learning journey.
So, okay, in prompting, you start with
prompt design and engineering libraries,
and then you look at, um, prompt templates
and variable injection and then system.
What's the difference between
a system prompt and assistant,
uh, a, a session prompt.
so it's just like.
Helping people to like figure
out where to start learning.
And it's organized into
like level zero learning.
Level 0, 1, 2, 3, 4, and
four is like cutting edge.
Like don't worry about that, for now.
Um, but yeah, I.
just did it as a weekend project to
help me organize, how I'm learning
and felt like sharing it too.
And hopefully someone else out there.
Jacob: That's so cool.
Yeah, I, I think teaching is often the
best way of learning yourself, right?
Like, uh, uh, learning something
enough so you can teach somebody else.
And that's kind of how I, I started
writing my newsletter retention
blog and, and like this was all
the stuff I was already doing.
I was like learning at these different,
learning at, uh, looking at these
apps, learning what they're doing,
analyzing onboarding flows, and
it's like, okay, well maybe I can
bring this to other people as well.
And that helped, uh,
uh, cement my knowledge.
Um, so it sounds like that's
kind of a similar, a similar
process to what you're doing.
Uh, any, um.
Any from, from doing that?
Any like counterintuitive learnings
you had while putting this together?
Anything that people would
maybe be surprised about?
Um, uh, yeah.
Gabe: yeah.
definitely.
Still, I, I, I put this
together, but it doesn't mean
I'm, I'm by any means an expert.
This is
Jacob: Okay.
Gabe: me to go from beginner
learner to intermediate learner.
It's my wishlist of, of
concepts I wanna learn.
one thing that, that has been
interesting and, and kind of.
Kind of counterintuitive is if you're,
if you're a technical kind of mind or
analytical mind, when you do the prompt,
you might wanna things down into like
bullets and sections and say, Okay.
here you are gonna do this and do this.
Then you're like, you instruct AI to
respond and operate in a certain way,
Actually, when you do that, you might
find that instructions are being ignored.
And even though you've organized
It, all and you're like, all these
things are important, how does it know
which one is most important when they
conflict, when the guidance conflicts?
Like what happens then?
Also there's context, um, windows,
and I guess if it's in the prompt,
it should all be within that kind of
context window, but it's, not all equal.
It's kind of like, well, I don't know
if it's exactly like, because the,
the way this information goes in,
you know, it's embedded in a vector.
Space and there's relevance,
um, considerations.
Um, and, and, what is it
like, um, semantic, like
closeness in the vector spaces?
I was gonna use the example of, um, SEO,
you know, the H one tag is more important
than the H two tags more important
than the H three tag, but it's, not,
I don't think the same in prompting.
So what you put at the
top isn't necessarily.
always most relevant, it depends on the
semantic relevance to the user's query.
So it's, it's kind of counterintuitive
'cause you wanna give it explicit
instructions to do or not do, but
then either it misses instructions
or it becomes robotic and it just, it
doesn't have, it loses its creativity.
And there's even a, for instance, a
thing called a catastrophic forgetting.
Um, where you try to teach it
something new and it learns that
thing and it forgets something
that was really important before.
So it's, it's black boxy,
it's like frustrating.
It's interesting 'cause
it's different every time.
Consistency, even if you're using,
uh, temperature that says like,
create, like always do the same
thing every time, but we'll still
be a little bit different each time.
So very much an art, I'd say more
than a science at this point of.
perfect prompt, and, uh, even
if you get it right today, it
might not be working anymore.
So it's, it's interesting.
Jacob: It, it's, um, it's, yeah,
I've had tons of fun exploring.
I think it reminds me of, uh, uh,
kinda what you're talking about.
Sometimes I'll, um, if I have this
long question, I'll break it up
into ask one part, have it respond.
Then follow up and break it up to make
sure there's like, no pieces missed.
But yeah, I'm, I'm, I'm excited
to dive in into your stack more.
Uh, uh, it, it's, there's,
there's so much to learn there.
Uh, and it, it's not gonna end.
There's always gonna be more and more.
Um, yeah.
Um, well, uh, this was awesome, Gabe.
I, I really appreciate you kind of sharing
all this expert knowledge, uh, and,
and, uh, um, you know, wishing you, you
success and kinda lingo growth journey.
Uh, and always having a new,
uh, uh, creative winner on the
horizon, uh, that's ready to scale.
Um, but yeah.
Anything else, uh, kind of you wanna
promote or have people go check out?
We'll, we'll link your AI stack
in, in, uh, the show notes
so people can go find that.
Um, yeah.
Anything else?
Uh, uh, you want people to go check out?
Gabe: No, people wanna
connect with me in LinkedIn.
That's cool.
If you have a question or
something, shoot me a message.
Um, I'm open for that.
But right now it's also helped me like,
um, with the AS stack, if there's a
concept that I missed, um, or I didn't.
Share it rights, like holes in
that, like, uh, I'm open for that.
And uh, also if you wanna learn,
uh, a sign language, American
sign language, it's fun.
It's cool.
Charlie lingo uh,
Jacob: Yep.
Gabe: it at that.
Jacob: Link bottom is an awesome map.
It's very well done, very well
designed, a great product.
Go check it out and then yeah, go, go
get if Gabe feedback on his AI stack.
Um, cool.
All right, well this is awesome.
Again, really appreciate you coming on.
Uh, yeah.
Talk to you later.
Gabe: Take care.
Jacob: All right, see ya.