Latest / AI for Founders with Ryan Estes / Bootstrapped Icons8 into millions of users

Transcript

Bootstrapped Icons8 into millions of users · AI for Founders with Ryan Estes

0:00
Welcome back to AI for Founders. I'm Ryan Estes.
0:02
Today we're diving into the world of design, creativity, and
0:06
AI with the true original Ivan Brown, founder of Icons 8.
0:11
Ivan Story is wild. He started out as a UX designer,
0:14
making banner ads for 100 bucks a month, then bootstrapped Icons
0:18
8 into a powerhouse that now serves millions of creators with
0:21
icons, illustrations, photos, 3D models, even AI generated
0:26
people. From hustling custom icon packs
0:29
in 2012 to launching generated photos year before generative AI
0:33
went mainstream, Ivan has been way ahead of the curve.
0:37
And he's doing all this while living in Argentina, where he
0:40
built a guest house for founders and digital nomads that somehow
0:43
helps people learn faster and build better.
0:46
We talk about bootstrapping, the evolution of stock assets,
0:49
generative AI, and maybe even shark fishing on the Argentine
0:52
coast. Let's jump in but for some
0:55
business. This episode of AI for Founders
0:57
is brought to you by Kitcaster, the podcast booking agency that
1:00
gets founders talking. If you're building something
1:03
amazing, you need to be on podcasts.
1:05
Not just because podcasts are fun, though they are, but
1:08
because podcast build trust, authority, and distribution like
1:12
nothing else. Apple and Spotify are already
1:15
publishing transcripts for LLMS, which means that every time you
1:18
guest on a podcast, you're feeding AI the news about your
1:22
company. Want backlinks, brand awareness,
1:25
and qualified leads all in one shot?
1:27
Kitcaster makes it happen. We've booked over 15,000
1:30
interviews for more than 800 founders.
1:32
You should be next. Head over to kitcaster.com and
1:35
apply today because your story deserves to be heard.
1:38
That's Kitcaster by MO Burst. Now, you and I know startups are
1:43
built in in boardrooms. They're built in the wild during
1:45
3:00 AM. Bug fixes, botched launches, and
1:48
that one meeting where everything almost fall apart.
1:51
That's what Code Story is all about.
1:53
It's a podcast where tech founders breakdown the make or
1:56
break moments behind their products, What went wrong, what
1:59
went right, and what nearly wrecked everything.
2:02
Hosted by Noah Labhart, Code Story goes past the highlight
2:05
reel and into the honest stuff, late pivots, brutal bugs, and
2:09
somehow coming out alive with a product people actually love.
2:12
If you're in the builds stage, the scale stage, or the please
2:15
Let this feature Work stage, I can relate.
2:18
This show will speak to your soul.
2:20
Listen to codestory@codestory.co or wherever you get your
2:23
podcast. Just search Code Story.
2:25
It'll be right at the top. Because great startups aren't
2:28
born, they're forged. One hard moment at a time.
2:30
Code story.co. All right.
2:49
Welcome to the show, Ivan. I thank you for inviting me.
2:55
It's absolute pleasure. We're really excited to have
2:57
you. We've got lots to cover about
2:59
your really interesting business, the the, the dips and
3:03
dives it's taken through AI. So why don't you introduce us a
3:07
little bit to the product and tell us a little bit about
3:09
yourself? I come from used design.
3:13
I'm used designer and I used to run the used design agency for
3:19
maybe 10 years. And then we transformed it to
3:22
the product. We started creating the icons
3:25
and selling them in packs. And then we switched to
3:28
subscription business. And then we ventured to other
3:31
types of stock assets such as illustrations, photography, and
3:37
at some point we started creating AI tools based on our
3:42
background and container. We how we started creating nice
3:45
tools and we formed a company called Generated Photos for
3:49
that. Amazing.
3:52
And I, Ivan, I detect an accent. Where are you from?
3:55
Texas. Argentina.
4:01
Amazing. I've never been, but it seems
4:03
like an extraordinary place. Tell tell us about where you're
4:07
from. I live in a small town called
4:11
Carilo. I wish people know about it
4:14
because it's not a top destination for foreigners in
4:17
Argentina. But all the Argentinians they
4:19
know about this place. It's a forest on the ocean,
4:24
beach and nice designed houses. Architecture that is just style.
4:33
So I'm trying actually to make it a little bit more
4:37
international. I've built a guest house to
4:41
invite founders and helmets to live there and to work there,
4:45
and they usually achieve pretty impressive results.
4:49
Something there is something in the air that makes them learn
4:51
the framework they always wanted to learn, or 3D modelling or
4:56
whatever. Nice.
4:58
I love that. The forest in the ocean, you
5:00
know, I'm from Colorado and I am the definition of a Colorado
5:05
boy. I mean, you look up Coloradan in
5:07
the dictionary, you see a guy with a pickup truck, a flannel
5:10
shirt and golden retriever. And that's me, you know, So I
5:14
love hunting and fishing and that kind of thing.
5:16
What's what's Argentina like as far as like sportsman?
5:19
Is there good fishing? Do you got trout?
5:21
Trout fishing in Argentina. We have, yeah, yeah.
5:26
And we have ocean here. So I have a ranch close here and
5:32
this is the spot for shark fishing.
5:37
Really. Which Woodward species?
5:41
It's a small sharks, maybe couple meters.
5:45
The sharks I've seen, they're a couple meters long and you can
5:51
eat them, but you you can, you can feed them and let them go.
5:55
So cool. I'm going to want to come back
5:56
to that, but I want to dive into your company now.
5:59
It from the research I did, it looks like you founded Icons
6:03
Eight in 2012 and fully bootstrapped it.
6:07
Is that correct? Yeah, for 23 years.
6:12
Yeah, incredible. So walk us through now I know
6:16
that like early on, you know, I started working on in the
6:20
Internet projects, you know, designing Myspace pages and then
6:24
that got me into building websites and I kind of came back
6:27
back door through through the music industry And early on icon
6:35
sets, you know, were just like an absolute bear.
6:37
And it was products like yours early on that were just like
6:40
such a reprieve where it's like, Oh my gosh, I don't have to
6:44
worry about icons, custom icon sets that would take hours and
6:48
hours and hours. So early on, you know, in those
6:50
early days, how did you recognize kind of there would be
6:53
a market here? And then how did you bring that
6:55
product to market? You're right, it was rough.
7:01
Like in packs were rough. At the time I first of all, I
7:06
didn't think about working on icons precisely.
7:09
I was a web designer. There was the single
7:13
professional designer. So I was prepared to do web
7:15
design, branding and any part of my work where icons.
7:23
And it became apparent that icons is the only thing people
7:32
want from me. So they ignored all my offerings
7:35
but wanted icons. So I started creating icons and
7:38
I was greeting icons for maybe five years and I, I'm pretty
7:44
much sick of it already. And then I've noticed that stock
7:52
photography is sold by piece or by subscription later, but I can
7:58
see. And so I created a small a
8:00
single page website with just 40 icons and I was selling them for
8:05
$40 a piece and I had plans for that.
8:10
So we didn't have much time for engaging in this project, but we
8:20
used it to lower the total bill for our customers.
8:23
So we offer it Icos $460.00. But like look this and this and
8:29
this, we take it ready and we don't charge you an extra.
8:33
And then we created an icon pack for self promotion.
8:39
Vista came out some really difficult to draw icons 3D like
8:44
so it created a really nice icon pack for several icons.
8:48
This was a massive success. So it was very posted
8:51
everywhere, shared everywhere. And then we created more icons,
9:00
more icons. At this time Windows 8 was in
9:05
the air and it had such simple icons that everybody could
9:11
create them. So we I didn't see the future of
9:15
our custom icon business at this point.
9:17
So I created the icon packs for self promotion, waiting for what
9:25
will come next. And they were popular and one of
9:31
them customers downloaders, he said, like, I like the icons,
9:36
but I hate. So I, I required them to set the
9:39
link for the license was link where you you sent the link to
9:43
us in exchange for for this icons.
9:46
So he was like, I like your icons, but I hate linking to
9:50
you. How about 100 bucks?
9:53
So we'd started charging 100 bucks for no link requirement
9:58
that we still do it up to this point and then we enter it to
10:05
other stock assets, yeah. Describe those stock assets.
10:13
Where did you go to after 2012 and before you do?
10:16
I want to set the table a little bit so people remember 2012 just
10:21
like it was incredibly iconic. Ironically enough time in tech.
10:26
So you have the rise of Instagram was taking place at
10:31
this time. You have tablets are like the
10:35
new hot product. And as a result, you, what you
10:39
see is a lot of like this skew morphism that kind of dominated
10:42
early iPhone editions. We're talking about iPhone now
10:45
starts to give way. You see people that had these
10:48
three-dimensional shadowed logos that were going to flat 2
10:54
dimensional. Everything was simplifying.
10:56
They were iconifying absolutely everything.
11:00
So like icons, whether it's in the App Store, this all became
11:03
super, super important across the board as far as design.
11:07
So you very early and right on this early Crest of, of design.
11:12
That was like seeing the importance of having unique but
11:16
also variable icons that you could use in, in web design.
11:19
Because now we're starting to get like really cool web design.
11:22
You know, the we have high speed Internet now for kind of the
11:25
first time. So you can get things, you can
11:27
get animations to load, you can get different things happening.
11:30
So you were able to capitalize on icons early on.
11:33
Where did you go next as far as graphic offerings?
11:38
Are you right about this camera Fusion?
11:41
I, I, I wasn't able to draw them.
11:46
So they were incredibly nice, but they were 3D rendered and
11:50
now nobody of us use 3D Right now we have 3D department and
11:56
they create wonderful assets. But at this point we, I was
12:01
like, I didn't know where to stand.
12:04
So we're trying to grow them in Illustrator and it was all
12:09
right, but not awesome. And one of the directions where
12:18
I wanted to venture was custom developments or designing,
12:23
designing interfaces and developing them.
12:26
And my friend he gave me is the best business advice I've got.
12:31
He said developers need a company.
12:35
Before we hired single developers or freelancers and
12:40
they worked great for the first three months and then bad for
12:43
another six months and then found another job afterwards.
12:47
So he said, like developers, they need a company, they need
12:50
party, they need to exchange, exchange ideas, they need to
12:55
help each other. And so I had money for our
13:01
developers for three months. So I hired 3 developers.
13:06
I had money for three months. We needed to find something in
13:09
the three months, some revenue stream which will pay for their
13:13
existence. And I found a project and they
13:20
hired one of our developers for several months and they, they
13:25
then they stole came from us. But it was enough money to
13:30
sustain us and to get in the profitable pathways with icons.
13:35
And so now you have. 2 developers.
13:38
I had two developers but I had managed to pay them.
13:43
Great. And it was was the project they
13:46
were working on icons and how did that go as far as like
13:49
initially establishing the pricing going from like kind of
13:52
one off pricing to then a subscription which seems to be
13:55
the industry standard at this point?
13:59
Yeah, it it wasn't obvious at all at the point.
14:02
So I created the first stock icons, the first store of of of
14:07
stock icons. And I'm not sure if I was the
14:11
1st subscription based icon or illustrations service, but one
14:17
of the first we had the marketing person, a girl called
14:27
Anya and she was just from the college.
14:30
She worked for us for a year maybe.
14:34
And then she found another job in Arab Emirates and before
14:42
moving there she said move their subscriptions.
14:46
And I was like, I'm not sure. She said move.
14:49
And I was like, I'm not sure if I if somebody will pay $20.00 a
14:59
month and she said they will charge even more.
15:03
And this is the best pricing or marketing advice I've got in my
15:10
life. Heck yeah.
15:13
Now it's it's great because it's it's such a profitable legacy
15:18
business over the years. And you guys have made a pivot
15:20
into generative AI in a big way. So talk a little bit about
15:25
generated dot photos. Like what was the origin story
15:29
of that? And how did you kind of make a
15:31
decision then to kind of move the company into that direction?
15:39
It started with stock photography, so we venture it
15:42
into different assets like illustrations.
15:44
It's straightforward. Illustrations are just large
15:47
icons. We had processes for that.
15:49
We had skills for that. We just did it.
15:52
And there was photography which is really different.
15:55
And my idea at the time was majority before majority, we
16:03
took bunch of 10s of thousands of photos of people with the
16:11
same light, the same camera height, the same backdrop.
16:16
We traced, we removed backgrounds and we created the
16:20
backgrounds for them, such as different authors, home
16:25
environment, 3D rendered and we took photos with the similar
16:30
lighting and the idea was compose them into the scenes.
16:34
Because stock photography is normally 90% good for your
16:40
purpose. And so we wanted to create the
16:43
rest 10% with collagen of actual photographs.
16:49
In retrospect it didn't work, but we've got that large data
16:56
set of 80,000 images of people, same light.
17:04
Amazing. Why didn't it work?
17:06
What what what? What was the failure?
17:11
People didn't use it for long. They played with it and then
17:17
forgot about it. We had 5 minutes of fame when
17:30
like kind of designer communities, they picked up the
17:33
the tool. We had the composer for that.
17:36
So they picked the tool and they created some hilarious memes.
17:41
And this memes were published in all over the world in different
17:46
magazine newspapers, which I'm not designer, which not
17:49
technical, just in general audience and that's it.
17:55
And like we we we sound like we created the meme tool, but
18:00
apparently it was a one time thing.
18:04
Crazy. OK, so you end up then with this
18:06
gigantic database of photography and so did you use that as
18:11
basically training material to create the the generated
18:16
photography tool now? Yeah, there was NVIDIA library
18:25
in NVIDIA created the library for generating people and there
18:29
was a popular website which is called This person doesn't
18:33
exist. This is this Pok library.
18:37
The guy just installed it generated bunch of photos and
18:41
was showing them randomly from some folder.
18:45
We did about the same, but we trained it post trained it on,
18:50
on our data set on for our models.
18:54
And because it was such consistent data set shot in the
19:03
same, in the same manner. And it was also very, they were
19:08
like nice people. They are beautiful people.
19:11
And we did makeup and hairstyle and did retouching some really
19:16
beautiful photos. And this, this style gun trained
19:20
the library of NVIDIA style gun trained on this images started
19:25
producing really good pictures of people.
19:28
This is fewer defects. And they, they actually were
19:35
criticized for that. People said like, look.
19:38
Too good. Are they all beautiful?
19:40
They want someone like us. Yeah, that's funny.
19:46
That's how I feel. I feel like it's good for for
19:49
people like me because all the stock imagery now is so
19:53
beautiful. It's like people are going to
19:55
get tired of good looking people and then they can come deal with
19:57
me. You.
20:03
Also, look our first version, like the second version, it
20:06
already included the people not beautified.
20:10
We're called it, not beautified. Yeah.
20:14
So cool. So how, how then do you start to
20:17
incorporate these new, you have a new tool set with, with gener
20:22
of AI. How does that fit into the
20:24
subscriptions Now? How are you kind of
20:26
differentiating the different business?
20:27
Because I noticed there's multiple websites.
20:29
So there's like the the photos is kind of its own product in a
20:33
way, but it is linked on icons 8.
20:37
We created generated photos as a separate entity before for two
20:43
reasons. First, to separate it legally,
20:45
because when we started, we had no nothing to like in practice,
20:54
in legal practice to rely on and we wanted to to preserve.
21:00
I can say it for for for that. And 2nd, we, I wanted to be a
21:09
startup companies are cloud providers, GPUGPU providers.
21:15
They are nice to startups and they are not so nice to 1220
21:21
years old companies. We got some cloud credits, we
21:27
got some recognition, some programs for, for startups.
21:30
It was really useful. Yeah, 100%.
21:34
And you know, the before the degenerative assets, which now
21:40
are basically innumerable, there's 1.4 million digital
21:45
assets available to users@icons8.com.
21:50
How do you manage inventory like that?
21:53
What what is it? What goes into that?
21:57
It started there's thousand icons when we had thousand
22:00
icons. I once wanted to find for icons
22:05
for a mock up and creating and I spent 28 minutes by clock
22:12
looking for them. I realized we can't continue
22:17
just given the zip files, res images, we need some search and
22:22
for that we need some database, we need some application and we
22:29
need to tag icons. Title is not enough.
22:32
There are lots of collections of icons that just title them and
22:36
there is search. And if there is a search, I can
22:39
you, you have to type to to look for it like shorts you, you
22:43
won't find it as find or when they fired us or something.
22:49
So we we have to tag it. And for that, we try to keep our
22:58
internal team small and outsource.
23:02
Most of the work is there to mechanical Torque and now to AI
23:10
and it's getting to psych AI Is, is, is incredible now.
23:16
It wasn't before, but now it's incredible at tagging the icon.
23:20
So we will, we keep our tagging team smaller and smaller.
23:24
Right now it's for people. Amazing.
23:29
Cool, you know, and for people that want to check it out, it's
23:33
Icons 8.com and truly these are like very unique, beautiful
23:37
designs. Whether you want to look at
23:39
icons, illustration, 3D models, photos, music, and then there's
23:43
the whole suite of generative AI functions.
23:47
Now I speak to a lot of folks outside the States and their
23:51
target market is always the US because you know, that's, that's
23:55
how you make the big bucks. So how were you able to
23:58
penetrate U.S. market because it looks like you've done so very,
24:02
very well? We have about third of our plans
24:07
from United States and 3rd. From third.
24:09
OK, Yeah. And 3rd from everybody
24:13
everywhere else, it's in dollar value for that, I think, I think
24:21
the you have to create an international product.
24:25
Yeah, you create, you use English, you create
24:28
international product. You don't think about any
24:31
specifics. It's like for, for for folks
24:35
outside US, US is, is everything international.
24:39
Like you watch Netflix, you, you get some financial analytics,
24:45
you learn accounting, finance, you programming, you get VS
24:52
version of it. So I think the key is use
24:57
English and learning stuff in English and creating
25:01
international version with everything best don't make AUS
25:09
company for stripe, use stripe for billing.
25:14
Use some cloud provider from the United States.
25:19
Some your service must be hosted in the United States.
25:26
Yeah. So like, go this path and it's
25:30
going to be a viable US product upset.
25:34
Now, I'm very interested in Argentina now.
25:37
So my daughter's off at college and my son is a junior in high
25:41
school. So he's got two years left,
25:43
which means that we can start looking around at different
25:47
places to end up, you know, or at least to go to for a little
25:50
while. And we and, and me and Mrs.
25:51
Estes have really been considering that like of moving
25:56
somewhere else, trying something else.
25:59
Make the argument why Argentina is a great place to go for,
26:03
let's say digital nomads or startup founders or people that
26:06
just kind of want to have a new place for an HQ.
26:13
I think the major point is Argentinians like foreigners.
26:23
They, I mean you, you get support everywhere.
26:28
So you, you can get lost. You, there's like you probably
26:33
will have a problem of too much help with someone walking you in
26:39
all the places and it's already 4 hours.
26:42
But you, you, you will be helped.
26:45
And they actually like foreigners.
26:49
Everybody has some grandma, granddad coming from Italy, from
26:55
Spain somewhere, and they feel that foreigners are one of them.
27:03
It's, it's, it's people. People take you really well.
27:08
It's a friendly place, it's. It's a really friendly, yeah.
27:13
And besides that, it's a really nice climb.
27:16
It's like cuisine, which is Italian in great part.
27:22
This is like good infrastructure.
27:27
Buenos Aires is a great urbanistically, a great city
27:30
urbanistically. It's like New York was a great
27:32
of streets. And they have, we have the
27:40
street corners. They're like a little bit like
27:44
plazas and there are summer cafes on them.
27:48
And the traffic is slow and there are like this trees, this
27:53
tall trees on both sides of the street.
27:56
It's it's a really nice place. What about the startup
27:59
community? Is there a big kind of startup
28:01
scene? Yeah, for Latin America, it's
28:06
probably quite big. So we have some unicorns, most
28:09
unicorns I think really either from Argentina or from Brazil,
28:15
but it's also much bigger country.
28:20
Yeah. And if you want to like if you
28:23
check meet up dot com you will you will see lots of events
28:27
here. Yeah, you know, Brazil is great,
28:31
but like, I don't want to learn Portuguese.
28:34
Come on, everybody got time for that.
28:38
It's so cool. All right, Are you ready for the
28:41
founder challenge? Questions brought to you by
28:43
kitcaster.com. This episode is brought to you
28:48
by Moberst, the AI first growth engine trusted by iconic
28:52
startups like Calm, Robin Hood, Reddit, Uber and Google.
28:57
Moberst is a digital, full service, mobile first marketing
29:00
agency that helps companies to scale and become category
29:04
leaders. Whether you're finding product
29:06
market fit or ready to accelerate growth, Moberst
29:09
delivers creative, data-driven solutions that help with scale
29:13
and efficiency and impact. At Moburst, we're not just
29:17
guessing what works, we're predicting it.
29:19
As an AI first agency, Moburst uses cutting edge tools to
29:23
analyze user behavior, adapt creative in real time, and
29:27
uncover hidden growth opportunities.
29:29
The result? Marketing that's faster,
29:31
smarter, and built to convert from early stage startups to
29:35
global brands. We've helped hundreds of
29:37
companies overcome complex challenges, unlock new
29:40
audiences, and optimize every metric that matters.
29:44
At MO Burst, we believe that creative thinking, advanced
29:47
technology, and data are the core drivers of success at
29:51
scale. If you're serious about scaling
29:53
your business with precision and purpose, it's time to talk to MO
29:56
Burst. Visit moburst.com.
29:58
And let's build something category defining together.
30:03
All right, perfect. OK, we're ready for the founder
30:06
challenge question from kickcaster.com.
30:08
Now I'm going to give you a couple of categories of
30:11
questions that you can answer from.
30:14
So we have 4 categories. We have unfiltered, we have
30:19
intriguing, we have philosophical, and we have
30:24
risky. Which category would you like to
30:26
do? OK, let's do it.
30:34
All right. Let's do it.
30:34
Let's choose one. Let's go.
30:36
Let's go. Let's go.
30:37
Risky. Let's see.
30:41
Here we go. What's the most desperate thing
30:45
you've ever done to close a deal?
30:51
And it still didn't close the deal of course.
30:58
I mean what comes to my mind is recently there were 29 people
31:08
wanted to stay in my guest house I rented an Airbnb to and it's
31:13
for 16 people. So I offered them to make it
31:16
basically a shelter from the living room and it still didn't
31:20
work. They didn't rent it. 29 you got
31:27
it. You'd have to build it.
31:30
Yeah, you'd have to build a tree house or something Outback.
31:34
Yeah, yeah. OK, let me do another one.
31:40
I got it. We're going to do a couple of
31:41
these The Founder Challenge questions.
31:44
Here we go. OK, what's the most money you've
31:49
wasted on something completely useless?
31:57
That must be marketing. How so It's.
32:07
We have of the marketing model, it was the business wallet works
32:12
real well. This link link where license
32:18
where people set links to us. So we're a great on Google.
32:21
So when Google where people Google for icons, we get here
32:24
and people who don't want to set link, which is probably most
32:30
commercial software applications, they pay us $15
32:36
and I will. I always wanted to try something
32:41
within my control to some AdWords, some influencer
32:46
marketing, some any paid advertising basically and it
32:52
didn't work. Apples apples store advertising,
32:55
it didn't work. None of it worked.
32:59
It costed me, you know, hundreds of $1000 this experiments.
33:04
No kidding. What?
33:05
What do you think it was? Do you think it was the the
33:07
product? You think it was the audience?
33:08
What was it that misfired? I think we've sell cheap product
33:16
is a little rhetoric on advertising board.
33:21
So we compete with Better Business models such as
33:27
Shutterstock for example. Somehow they they know how to
33:32
get more dollars from the customers.
33:34
So they are outbeat us. On the keywords.
33:39
They think, yeah. So if if advertising not a big
33:42
driver for you, what has been the best driver for new, new
33:46
customers and growth for you guys?
33:50
Already said about liquid license but what else?
33:55
For example, Product Hunt. Product hunt, Yeah.
34:01
So I there is a website with top product, hunt planters,
34:08
creators, top product and #1 there it's stuck in time.
34:13
It stopped updating maybe couple years ago, but so I'm probably
34:19
#1 product hunter for for forever.
34:21
You know, it's not easy launching on Product Hunt.
34:24
How did you get such a big launch?
34:33
Like before Hunters really mattered.
34:36
So you, you want Chris Messina to launch you?
34:44
He was #1 Hunter. So you write him, you want to,
34:49
you pitch him, and if he chooses you, it's a great deal.
34:53
Then they were not really clear. Always pick, not really like the
35:03
product. Hunt is balancing between
35:05
getting free promotion for themselves and not letting
35:12
product is who stamped most people win.
35:15
So you should go within the their little rooms right now.
35:19
There are some certain boarding that is all right.
35:22
So you we said, man, we have a mail in for a million people.
35:26
So I was in the mailing to these people and we said without like
35:30
help the lunch party on Proton and it's OK with them unless we
35:35
don't say upvote us. Amazing cool.
35:39
Yeah it's definitely changed a lot over the years, hasn't it?
35:42
I, I still use it as like my my go to tab just to see I like to
35:47
see like what product under the particularly the UIS, the the
35:51
the front side of products to see kind of what's trending and
35:54
what what design that people are really resonating with.
35:58
There's always cool tools, you know, now it's just it's so
36:01
dominated by so many AI tools now that it's kind of hard to to
36:05
weed through it. But.
36:07
It it is. Great for inspiration.
36:10
Yes, thank 6 for for for keeping an eye on it and because project
36:15
hunt getting product hunt top places is one thing, but
36:19
another. The the next day comes the
36:21
second wave of press investors and bloggers who are spread.
36:31
Yeah. Totally love it.
36:34
Couple of questions here. If you sold Icons 8 for a
36:40
billion dollars today, what do you do tomorrow?
36:47
They have 3 passions right now. Besides I can say it's
36:51
motorcycles. Listening to classical music in
36:59
my own cinema. Nice audio files, less classical
37:04
music and GG. So I will enjoy this three for a
37:10
while. Maybe I will buy more
37:16
motorcycles. What kind of what kind of bikes
37:19
you like? I just started man.
37:22
I'm not like them all but I I would like some large adventure
37:28
bike. I would like some some really
37:33
heavy Harley-Davidson and Dura bike, maybe electric, electric
37:39
and Dura bike. And I want my garage to be I
37:45
would, I would, I would set the lights or I will switch on the
37:48
light and there will be like this, this this light about for
37:52
each of the motorcycles. Like Batman?
37:57
You just. Yeah, you know, I've always been
38:01
a big fan of the classic Indians from the 70s, but now they have
38:05
a new product line that's super cool.
38:07
I think they're they're called the scouts.
38:09
And these Indian scouts kind of a more of a medium smaller size
38:13
bike, but they are so cool. I mean, I was in high school is
38:19
actually 1 to get a motorcycle the first time.
38:21
And and my mom said that it was OK with her if I got a
38:25
motorcycle, but I had to save enough money to pay for my
38:28
funeral first. Now you have him running it.
38:33
So I kind of discouraged it. Cool.
38:36
Is there ACEO or founder that you follow or admire?
38:47
Probably most druke I I admire Peter Druker, but he's not CEO.
38:55
But there is Jesus Christ called high performance management, The
39:02
CEO of Intel, I forgot his name, but he was a it's a great book,
39:10
High performance management. I I really admire him.
39:14
What's the company called? Intel, OH.
39:17
Intel Yeah, yeah, yeah. Cool.
39:19
Yeah, he's a Co founder of Intel and he was CEO for lots of for
39:26
for many years. Nice let me find out his name
39:29
because I, I would love to look him up.
39:30
That's that is a great company. You know those those kind of
39:33
sleeper companies that are just baked into everything don't get
39:37
a lot of attention until like NVIDIA, you know, and then he
39:40
really they really blew blew up there founder of Intel.
39:44
Let me see. It looks like Gordon Gordon
39:49
Moore like. Yeah, yeah, yeah, yeah.
39:52
You know, he has also a nice great book Christians of Chasm,
39:57
but this one. Was a Robert Noyce.
40:02
In Hyper for hyper performance, I'll probably look for a book.
40:08
It's pretty amazing. Founded in 1968, It's
40:11
incredible. I love those stories because,
40:14
you know, you know, being a founder and as founders, it's
40:17
like every company has a beginning, a middle and an end.
40:22
And just depending on kind of like the longevity and the
40:25
management of that company, they can come pretty quick.
40:28
But you look at Intel, who's been dominant, I mean, and they
40:30
are founded in 1968, you know? Yeah.
40:34
And they rose 5% today, the stock.
40:37
Did they really? The end of growth and when I
40:42
opened Kindle. I read this book many times in
40:44
many years ago, but it's still one of three of my top books in
40:49
my Kindle. Amazing, amazing.
40:53
Ivan, I really appreciate it. You've been so generous with
40:55
your time. I got one more question for you.
40:59
Which one should I do? I want to say if you were
41:03
entering college today as a freshman, what would you study?
41:13
I was really thinking sometimes that being in Stanford in
41:19
computer science would give me a jump start in my career in life.
41:25
So probably something that I had to learn for boring for 20 years
41:32
would be given. So I would go and study computer
41:36
science at Stanford. Amazing, Ivan, thank you so
41:40
much. Thank you.
41:42
All right, now let's do that. The fired segment.
41:46
All right, Ivan, we're doing this little segment for
41:49
founders, help founders. So maybe you could walk a step
41:52
by step on a very interesting subject that you brought up,
41:56
which is knowing when and what to do when you got to let
41:59
somebody go. First of all, let him go now.
42:05
It would be the worst day of your week, but next day would be
42:10
the best day of your week. You already tried everything.
42:14
You already tried performance reviews, you already tried
42:17
coaching, you already tried resigning the task to them.
42:23
It didn't work. So this is the way to go.
42:26
And to do that, you have a frank conversation.
42:31
Actually I am always thinking about sending an e-mail would be
42:40
humane and get so you don't get the first reaction, the first
42:44
shock of the person. The person could process it in
42:47
private privately and then soon you have a conversation when you
42:52
confirm you are fighting this person in soft.
42:57
But for the words. If you want I can explain why
43:04
normally people don't want and they know.
43:06
So then you just let him go and you close all accounts.
43:12
Now it's also rough but it's needed Slack accounts CRM
43:20
system. In the best case he would be
43:26
talking to others and bed wipes spreading the bed wipes in.
43:34
In the worst case, he's downloading the list of
43:36
customers or or worse. That's it.
43:42
Amazing. And you know, it's, that's an
43:44
interesting technique you have you found that emailing them
43:47
first, just allowing themself is that kind of like to retain some
43:52
dignity so it doesn't hit him so hard.
43:57
I tried it couple times but normally I don't do that.
44:01
I think it's a good idea. I still need some Bhd to confirm
44:05
it's a good idea to start. Tremendous, Ivan.
44:10
Thanks again so much. Really appreciate it.
44:15
Hey founders, let me ask you this.
44:17
Are you building with AI or getting left behind by it?
44:21
If you're serious about scaling smarter, faster and bolder, you
44:25
need to check out AI for founders.co.
44:29
It's your go to resource for cutting through the noise and
44:32
actually using AI to grow your startup.
44:35
Each week, AI for Founders drops a sharp, no fluff newsletter,
44:40
dives deep with expert guests on the podcast, and hosts monthly
44:44
workshops where you can get your hands dirty with the real tools
44:47
and strategies. Whether you're precede or Series
44:51
B and beyond, launching something new or just staying
44:54
sharp, AI for Founders gives you the edge you need to win in 2025
44:59
and beyond. Hit AI for founders.co to
45:02
subscribe to the newsletter, tune into the podcast, and grab
45:06
your spot in the next workshop, and it's all free.
45:09
Build smarter, move faster, think bolder with AI for
45:13
Founders. Now back to the show.