Expert Insights on Startup Ideas and AI's Impact on Employment with Andrew Wilkinson

You don't want to walk into the gym on day one and try and deadlift 300 lb. So, when someone comes to me and they're a

first-time entrepreneur and they say, "I'm going to make the next great AI company." I think that is the

equivalent. I feel like you've actually started and run more companies than maybe anyone else in the world. What is

your best advice for coming up with a great startup idea? Charlie Mer, Warren Buffett's longtime business partner, has

this amazing quote, "Fish where the fish are." The biggest mistakes I've made have been going into business models

where other people have repeatedly failed and thinking, "I can do this better." It's so funny to watch you on

Twitter. Clearly, you've become AI obsessed. It's like having the world's most reliable employee who costs $200 a

month and works 24/7. So many knowledge work jobs are going to change massively. I think the fundamental question is, do

all jobs just become a single prompt? Today my guest is Andrew Wilkinson. Andrew is the co-founder and CEO of

Tiny, a holding company that's often called the Berkshire Hathaway of the internet. They own over 40 businesses

ranging from dribble to we commerce to the Aeropress coffee maker. And they focused on buying profitable businesses

from founders and holding them for the long term. Andrew and his co-founder bootstrapped the business from zero to

hundreds of millions of dollars in value. And Andrew personally was worth over $1 billion at one point. In our

wide-ranging conversation, we cover a bunch of strategies for coming up with a good business idea, what common business

ideas you should avoid, his experience automating much of his work and life using AI and what that means for

employment in the near future. Also, what he's learned about happiness and money and how they are not directly

related and also how getting diagnosed with ADHD and then taking SSRIs was the thing that most impacted his happiness

in life. This is both a powerful and also a very tactically useful conversation and I'm really excited for

you to hear it. If you enjoy this podcast, don't forget to subscribe and follow it in your favorite podcasting

app or YouTube. Also, if you become an annual subscriber of my newsletter, you get a year free of a bunch of amazing

products including Bolt, Linear, Superhum, Notion, Perplexity, Granola, and more. Check it out at

lennisnewsletter.com and click bundle. With that, I bring you Andrew Wilkinson. This episode is brought to you by Sauce.

The way teams turn feedback into product impact is stuck in the past. Vague reports, static taxonomies, unactionable

insights that don't move business metrics. The result turn lost deals miss growth. Sauce is the AI product co-pilot

that helps CPOS and product teams uncover business impact and act faster. It listens to your sales calls, support

tickets, churn reasons, and lost deal surfacing the biggest product issues and opportunities in real time. It then

routes them to the right teams to turn signals into PRDS, prototypes, and even code that drives revenue retention and

adoption. That's why whatnot, Linkree, Incident, IO, and ZIP use SAS. One enterprise uncovered a product gap that

unlocked $16 million ARR. Another caught a spiking issue and prevented millions in. You can, too, at s.app/lenny.

Sauce, built for AI product teams. Don't get left behind. This episode is brought to you by

Interpret. Interpret is a customer intelligence platform used by leading CX and product orgs like Canva, Notion,

Perplexity, Straa, Hinge, and Linear to leverage the voice of the customer and build best-in-class products. Interpret

unifies all customer conversations in real time from Gong recordings to Zenesk tickets to Twitter threads and makes it

available for your team for analysis and for action. What makes Interpret unique is its ability to build and update a

customerspecific knowledge graph that provides the most granular and accurate categorization of all customer feedback

and connects that customer feedback to critical metrics like revenue and seesat. If modernizing your voice of

customer program to a generational upgrade is a 2025 priority like customer- ccentric industry leaders like

Canva, Notion, Perplexity, and Linear, reach out to the team at interpret.com/lenny.

That's e n t e r p re t.com/lenny. Andrew, thank you so much for being here. Welcome to the podcast. Oh, thanks

man. Great to be here. I've been wanting to chat with you for so long. There's so much that I want to talk about. And I

want to start with a topic that I know that you think a lot about and that's also on the minds of a lot of people,

which is coming up with a great startup idea. And uh this is something that a lot of people are thinking about right

now because AI makes it so easy to actually build your idea into something real. And I feel like this is something

you spent a lot of time thinking about. I feel like you've actually started and run more companies than I don't know

maybe anyone else in the world. I feel like you're in the top 100, top 10, something like that. I don't know. Does

that feel right? I've definitely had a lot of experiences and I've probably started or been involved with

75 different projects or businesses where I've been a primary contributor. And I wouldn't say that's anything to

brag about because I've been an inch deep and a mile wide. Um, so that's been in a lot of ways kind of my Achilles

heel is I get too excited about ideas and I start too many businesses. But as a result, I've seen almost every

business model under the sun. Amazing. Okay. So yeah, I think the benefit is for us is we get to learn from your

experience. So let me just ask you this question. What is your best advice for coming up with a great startup idea?

Ultimately, the best thing is something you're going to be interested in. But I think a mistake a lot of people make is

they choose something that everything is everybody is interested in. So, for example, they say, I I don't know anyone

that has had the thought, man, I would love to start like a really cool restaurant or, you know, I want to have

a cool cafe or something. And what in reality, what they think about is how cool would it be to come up with a an

amazing logo or, you know, all the fun stuff, design the lo the menu and stuff. But in reality, operating those

businesses is miserable. And it's also a very hard business because every morning millions of people wake up and they go,

I should start a cafe. But on the flip side, almost nobody wakes up every morning and says, you know what, I'd

love to start a funeral home or I would like to start a pest control business or I should start software that helps

people fill out forms for the government. And Charlie Mer, Warren Buffett's longtime business partner, has

this amazing quote. He says, "Fish where the fish are." And he gives this example. He says, "You know, if you're a

fisherman and you see a large pond and all around the pond there's a whole bunch of fishermen and they're all

elbowing each other out of the out of the way. They're all using the best lures, the best fishing line. Uh, you

know, they all have amazing strategy. You actually want to walk off into the forest and find a small fishing hole

with lots of fish and very little competition. And I think that's probably the most important thing in business is

actually to find those niches where you can actually make real money because competition equals lower margin. The

more competitors there are, the lower your prices have to be and the more competitive the business is. Ultimately,

it doesn't feel uh intuitive to go after small markets and to find niches. So, let's just follow that thread. Well, why

is that actually a source of some of the best ideas starting really small and niche? I don't necessarily think it has

to be something that is small forever, but it has to be something that like I think about if if you're a first-time

entrepreneur or a student or something, you don't want to walk into the gym on day one and try and deadlift 300 lb,

right? So when someone comes to me and they're a first-time entrepreneur and they say I'm going to make the next

great AI company or you know I'm going to launch uh a new bank or something like that something that is very very

rigorous and complicated and highly competitive and regulated. I think that is the equivalent. I think you really

want to take the baby weights and start slowly building your muscle. And I think about my own experience starting a

business. I was so lucky because the first business I ever started, my web design agency, which became Metalab,

that was so easy and it worked. It worked immediately because all I had to do was know how to build websites and be

able to talk to potential customers and then once they said, you know, yes, I will pay you $5,000, I just had to send

them an invoice, do the work, and that was it. It was a very simple business. And because of that, I got immediate

positive feedback. And I built my own narrative. And my narrative was, I'm good at business. I can do this. Keep

going. And then I went off and I started taking my money uh that I made from that original business. And that's when I

[ __ ] up. So I went out and I started a pizzeria and I lost all my money. I started a designer cat furniture

business, a online DJ school, uh a skin cream business. all of these things, I just lost all my money almost

immediately. But because I had that first win, I kept on going. And I just think it's so critical that people

choose a business where they get that initial win. Okay, this is a great topic. This is just like how do you

avoid creating a job for yourself you hate. There's a lot of business opportunities and ideas that like, yeah,

you can make this work and then you work and then you're stuck doing this thing, running a restaurant. Uh my wife tells a

story where a friend started a coffee shop and then he's like, "I thought I was starting a coffee shop, but I'm just

uh replacing milk and buying milk all day. That's my job now." So just kind of along those lines to help people avoid

creating a beast that they didn't expect. Any advice for just how to know this is maybe even though this may work

and make money, you probably don't want to be spending your life doing this sort of business. So like I just started a

pressure washing business. I was speaking at a local business school and after I spoke, one of the kids walked up

to me and he said, "Hey, I'm an entrepreneur. I've started two or three businesses in the past um doing

landscaping and I'm not enjoying school." And so, just on the spot, I said to him, "Why don't you drop out and

we'll start a business together?" And I'd kind of been cooking on this idea of a pressure washing business because I'd

studied that industry a little bit and I knew a few friends who' done it in other cities. And I had this unfair advantage

which is I owned a whole bunch of media properties where I could advertise it for free basically. And so uh we started

this business and and I think for him he kind of had this moment of do I really want to be washing people's driveways

for the rest of my life? And what I said to him is, look, if this gets to scale, you'll never touch a pressure washer

unless you want to. You know, if the business can get big enough where you can have employees, he can just focus on

sales or digital marketing or whatever aspect of the business that he really loves. I think that's the biggest thing

that a lot of entrepreneurs miss. You know, they say, "I don't want to do that for the rest of my life. I don't want

to, you know, be in the back of a dry cleaner dry cleaning clothes. And to me, it's just a question of scale. I think

the cafe is a great example because a cafe, if it doesn't get to a reasonable scale, is just a job. There's a big

difference between a business and a job, right? I think if you you could start a pressure washing business where you're

the only employee, yeah, that's a job. But if you can get to a scale where you can drive 10 leads a day, then you don't

have to do any of the pressure washing and you just do what you love. So, I think a lot of people have this kind of

Protestant work ethic where they think, well, I've got to be the person to do everything. And I think they really need

to lean into what I call lazy leadership, which is how do I get away from the things I hate as quickly as

humanly possible? How do I be teflon for tasks? It's interesting that this business is non-software related at all,

this pressure washing business. So I guess just for folks that are coming up with ideas and I imagine most ideas are

there's like a pull to make it software oriented something AI is going to help build for you and help you run. What's

your calculus on just going down a direction of like real world physical business like a restaurant power washing

business versus uh software just like what are the benefits of that direction? When should someone actually seriously

think about doing a business like a power washing business? I think if someone's listening to this podcast,

odds are they're someone who's kind of a digital native. And I think the question is, what's your unfair advantage and

what are you great at? So for me, I think I have reasonably good taste and so what I could do is identify and I

knew enough to identify great design and development talent, but mostly I was lucky because I was a talker. I was good

at meeting with clients and selling myself. And so ultimately my highest and best use, my superpower was sales. And

so that can be applied to anything, anything where you need to go out and you need to sell a customer. And so

really it just comes down to what do you get drawn to and then how do you find the profitable business within that. So

here's an example. A friend of mine, he owns a restaurant and he said, "Oh, like I love it. It's so I'm so passionate

about it. It's this stunning, beautiful uh restaurant in my hometown. And he said, you know, but it's really just a

job for a few different people and we can't make any money at it, but I've been noticing that there's all these

vendors that service the restaurant and those guys are making a killing. And so he told me about a business that cleans

grease traps, another one that cleans uh exhaust vents for kitchens. So I think looking around and seeing where your

passion is and then sniffing around within it probably somewhere within your passion there's an opportunity. So for

example I love movies. My happy place is a dark theater. You know that's the best way for me to de-stress is go to a dark

theater and watch a movie and get lost in it. And probably about four or five years ago, I was like, man, how cool

would it be to invest in movies in some way to be a part of that creative world. And so I started looking into funding

movies. And I realized that when you fund movies, like 90% of the time you lose all your money. And even if you do

make some money, you rarely make a good return. But I kind of spent a bunch of time understanding the industry and

learning about it. And then uh two years ago I was in New Zealand and I met the founder of Letterbox and I realized, oh

my god, this is a business where uh this has a moat. So it has a network effect. It's a huge social network for film

reviewers. It's something I'm passionate about and it's something that we can buy at a fair price. And so all those things

came together and I was like, "Oh my god, I can invest in film now." So and in the same way I used to be a barista,

we ended up buying the Aeropress coffee maker company. So I kind of follow my passions and spend a long time learning

different industries and then I find the profitable niche within. Okay, this is a great takeaway. Essentially, when you're

thinking about startup ideas, make sure there's some connection to something unique to you where you have some unfair

advantage. This makes me think about Brian Armstrong. He come he I saw him do a talk once and he gave this really

amazing insight about why uh why Coinbase did well and why he started Coinbase. And it's because if you look

at his background, he had this weirdly rare van diagram of background of computer science and I think it was

cryptography and economics which is the exact set of skills you need to start something like Coinbase.

And so I think the tip there is just what is that unique ven diagram of skills in your background and just

ideally what you're building connects to that and gives you an unfair advantage. Well, yeah. I mean, like I just met some

another UIC student at the local university student, you know, she is interested in marketing and so she's

gone out and she's found some local clients, so small restaurants and stuff and managed their social media. And she

kind of said, "Yeah, it's okay. I can make a,000 bucks a month, but it's a lot of work and you know, the owners really

want a lot for their money." And I said, 'Well, you know, if you just pivot that idea just ever so slightly and instead

of doing restaurants, you did realtors or wealth managers who have quite a large marketing budget and are used to

spending serious money and it only has to work a little bit to make a lot of money for them. Those people you can

charge $5,000 a month. So, I think often it's you find your passion, you know your skill set, you zero in, and then

you just kind of pivot and you find the most profitable way to do it. Like when I started my web design firm, I started

mostly with local and small projects, but very quickly I found a job board in San Francisco where startups would share

uh projects they needed help with. I could charge five times the amount for five times less work. This point about

fish where the fish are, I think, is really important. You can start something that's awesome, that you love,

that you're so excited about, but nobody needs it. Talk a bit more about just what that looks like. What have you seen

when you're thinking about ideas that tell you that there's fish but also not overfished as he pointed out? I think

it's hard like Warren Buffett has this great quote, I'm a better businessman because I'm an investor and I'm a better

investor because I'm a businessman. And I feel like in order to know what problems are valuable to solve, you kind

of need to have valuable problems. So, it's a bit of a hard thing because, you know, I remember when I was like 20, I

would say, um, I hate how ugly all the cat furniture I can buy for my cats is. I bet people would pay a lot of money to

solve this. But I didn't understand anything about the realities of that business model. I didn't understand how

little people were actually willing to pay. I didn't there I just I didn't have enough life experience to go yes that's

actually a worthwhile opportunity and I think that being able to know what people would pay to solve the problem

realistically is incredibly valuable. So my example of a realtor selling a house because I've studied that industry I

know that a realtor can make like 20 to $50,000 selling a single house. So, I can intuit it that they'd be willing to

spend a lot of money if I have a unique way of getting them clients that'll buy houses that'll convert at a high

conversion rate that is worth $5,000 a lead maybe, right? So, so I think you kind of have to like understand the

problem in depth before you really know because when I was starting out, I would go down every rabbit trail, you know,

every um infomercial idea I had, I would think was genius. For people that don't I don't know that don't have this

experience which is most people is there is there one thing you do maybe a heruristic that kind of gives you a

sense maybe there's something here maybe there's a lot of fish maybe it's a valable problem or no this will tell me

it probably isn't no unfortunately I think it's mostly gut I mean it's a it's a like you know Mer and Buffett talk

about having a whole bunch of mental models in your brain and they form a lattice work right and they all kind of

piece together and I feel like for me there's so much of this that is like it's almost like I'm an AI model and

I've trained on all this data of what works and what doesn't and what you know what's a good boat and what what's a bad

business etc. And when I see it, I just immediately know. I mean, I think all the all entrepreneurs are so lucky now

to be able to go into Claude or Chap GPT and just say, "Hey, I'm thinking about starting a Botox clinic. Can you break

down the numbers? Is this a good business? What's hard about it? What is the regulatory uh moat? You know, what

what would my payroll look like?" I didn't have any of that. And so, I started so many incredibly stupid

businesses, but I think there's no excuse at this point. you should be able to do it with AI. That's a really good

tip actually. Then there's this point you made about boring is good. That's a really good piece of advice. A lot of

people are going after flashy stuff, things they condemn on Twitter. Your advice here is boring is actually a

really good thing because fewer people are going after it. Is that the general tip? Totally. I mean, I think um so I

started this business uh one of my first called Flow. And Flow was basically a sauna before a sauna came out. So, was a

way to manage your to-dos and projects with your team uh in a web app. And we basically did the thing, we made the

mistake that so many entrepreneurs make, which is to go after an industry that everybody goes after, you know, just

like cafes, everybody has the idea, man, I wish I could design my own project management system or my own to-do

system. And not only that, but everybody loves new things. They love jumping around. And I know me personally in the

last 3 years I probably used three different productivity systems and I I love jumping around in them and I didn't

understand that and I ended up losing $10 million trying to compete with venturebacked businesses and

bootstrapping. It was all my own money. Poured $10 million lit on fire trying to compete with Asana because I didn't

understand how the business world worked really. You know, I was kind of like they they had raised, you know, hundreds

of millions of dollars and they were run by the co-founder of Facebook and it was a little little bit like I'm Fiji and

I'm deciding I'm going to invade the United States in retrospect. Just completely silly. Now, on the flip side,

if I'd taken that same amount of energy and I'd instead focused on, let's see, what's a really boring one I've heard

about there? There was a business I saw recently. They were making $30 million a year, and all they did was help people

fill out forms to get government assistance in certain programs. So, it' be like, uh, you know, your uncle is

disabled and you need to access government funding. They the process to do it is incredibly time consuming and

miserable and you have to fill out all these forms. They just have software that fills out the forms for you and it

says look you're going to get $20,000 as a grant. Pay us a,000 bucks and that is such a boring like nobody wakes up and

goes I want to make form filling software but I think they would if they could make $20 million a year. Along

these lines of just bad ideas things people shouldn't start. What are a few ideas that you think everyone thinks is

going to be a great idea and then they do it and then they always fail? Well, I think the biggest mistake I can speak

from my own experience. The biggest mistakes I've made business-wise have been going into business models where

other people have repeatedly failed and thinking I can do this better. So, for example, me and one of my best friends

about 10 years ago, we really wanted to start a bar. We just thought it'd be so cool to have a bar where us and all of

our friends could go. We thought it'd be great for the city. And so we and we and we thought, you know, we can run this

really high margin because we're tech guys. We know how to build systems. We're good at business. And we were

utterly humbled. You know, we realized that we didn't know anything about business compared to someone who runs a

pizzeria. If you think about it, like my example earlier of if you run a software company, what has to happen? You have to

hire a bunch of nerds. They need internet connections and computers and you need to pay them and you need to

coordinate online and everyone can work asynchronously. Nothing has to go right. You know, it's not that complicated, at

least at small scale. A pizzeria, it's like if the baker doesn't wake up at 3 in the morning and start prepping dough,

the entire thing is effed. And all along the way there's a hundred different failure points from front of house, back

of house, the deliveries arriving on time, all these logistics. And so I think it's been stuff like that. I I

mean I also, you know, I got into the news business, the local news business. I ended up buying a like a old paper uh

in Vancouver and it's the exact same thing. It's like I just can't you can't take a brilliant management team and

change a bad business model. You know, ultimately the business model wins. This is uh this is really good advice.

Basically, if there's a bunch of dead bodies in that space, there's probably something there that keeps killing them

that you probably not aware of until maybe somebody figures something out, right? Like once in a while, once in a

blue moon, someone's like, "Okay, here's how we do this." And then it works. Well, look at Instacart, right? When

Instacart came out, everyone said, "Well, web van failed. This is never going to work." And we still don't

necessarily know if Instacart is a great business. I don't I don't know. I haven't studied it but but I think it

was like okay enough technology has changed that it can happen but would it be easier to start Instacart, Amazon or

Coupang or would it be easier to start an enterprise SAS software company? Definitely the enterprise SAS software

company. I see. So this is I guess is the advice there. The thing that is easy to start is the thing you should not do

because everyone's going after that. Well, it goes back to my gym analogy of if you're going to deadlift 300 pounds,

train for 15 years, right? You should have already had three startups and they should have worked and you know, you've

really built up the reps. And I think that if I was going to start Instacart, you know, that's different than someone

who's 20 starting it. Not that I would be good at it, but at least I would know what I'm getting into. Let's talk about

something that is kind of this like endless debate on Twitter. Maybe it's a false dichotomy between lifestyle

businesses, kind of bootstrap businesses. This idea of not raising money, just making revenue, living off

the revenue versus ventureback venture scale companies. Feels like you're uh very good at the first bucket. And uh a

lot of people, this is their dream. I'm just going to start something. I'll start a lifestyle business, make a few

million a year, never raise money. Who needs VCs? VCs suck. Advice for deciding which route to go with a an idea you

have? Well, I think um there's this it's just not true that a quote unquote lifestyle or bootstrap

business can't get huge. I mean, we bootstrapped the entire business and now across all of our companies, we do

almost $300 million in revenue, right? So it doesn't there's no and the the whole time you know I had been focusing

on taking you know starting small businesses small ideas simple ideas often buying small companies and

watching them grow really big and I think that the only difference between what we do and what a venture capitalist

does is the level of tolerance of burning money on fire and I just haven't seen that if we ch if we choose the

right businesses that aren't in incredibly competitive markets or where they have some sort of moat. So, what I

mean by that is like a great brand or a network effect like a social network or something like that. There's no I don't

feel we're holding them back by not letting them light a bunch of money on fire because these things naturally

grow. The numbers I mean they're like, you know, balloons. They just go uh as long as we don't mess them up too much.

So I think the decision ultimately comes down to how hairy do you want your big hairy audacious goal to be. And if you

you know your big hairy audacious goal is I'm going to start the next satellite business that you know creates some sort

of crazy technological revolution. Yes, you're going to have to raise venture capital unless you're already a

billionaire or something like that. But if you're wanting to just tinker and solve a problem that you think is not

going to be hyper competitive. So for example, that form filling thing or software that just does a narrow thing

that doesn't require 20, 30, $40 million to be lit on fire before it can make money, then you know, you can have a

wonderful life and build a wonderful company that can scale into the hundreds of millions of dollars if you play your

cards right without ever raising money. per the story you shared with Flo where you were competing against a venture uh

venture funded company if there is a venturebacked company in the space is the advice like you're not going to win

competing with them most likely and try to do something else well look at um things do you know things

I do it's awesome so you know things still exists and things has existed for 20 years it's run by I believe one or

two or three people like it's not a big team, certainly under 10 people. It was, I believe, bootstrapped and it's just

consistently delivered an exceptional product and built up a loyal following. They have their, you know, 10,000 true

fans who use it and that's enough for them. But they don't do a lot. They've been very intentional. They don't do any

AI stuff. They don't have an API, you know, all they they've they're not on Android. They're very focused. And I

think they they have succeeded. But the question is what would an MBA or a business professor say about their

success. If the measure of success is did they maximize taking as much market share as possible and grow to be as big

as possible then the answer is no because Aana and other people have built multi-billion dollar companies. If the

goal is the founder has an incredible life, probably has three houses, flies all over the place, does whatever he

wants, has recurring revenue, and he gets to work with headphones on building a beautiful piece of software that

people love, then I think he's won. I am generally much more in the camp of the things guy has won, not Dustin

Moskavitz. Dustin Mos Moskovitz is just playing a different game than the things guy. And if Dustin Moskovitz was running

things, he'd be miserable. And if the things guy was running a sauna, he would probably kill himself because he wants

to put his headphones on and build. I wonder why nobody has come. It feels like just with lifestyle businesses, if

it's working and you would think somebody would come in with more money and more funding and just eat eat their

lunch is the key that the market is too small for a VC to ever be excited and so that's why no one's coming for them. I

think so. I mean, I think things probably I'm I'm just guessing. I don't know any of the numbers or whatever, but

I would assume it makes between five and $25 million of revenue, which to a VC is like not even worth considering. Like a

VC, they want to you for for a VC to invest in your business, you need to have a story where it's worth 300

million to a billion dollars or more. So, it's just not that exciting to them. Which, again, going back to fish where

the fish are, right? You don't want to be con uh fighting against the commercial fishermen, right? If you just

want to have a little business where you make enough, you know, you get enough fish to feed your family and your

village or whatever, find the other fishing hole. Don't go where the trwers are. This is great advice. If you're

trying to start a non- ventureback company is find uh this metaphor just keeps working for us, which is uh fish

where the fish are, but not where the professional fishermen are. Also, ideally, not where there's just like a

ton of uh I don't know, fishermen from all over the place. I don't know. Solo solopreneur fisherman. Okay. Coming back

to just starting a successful company, what would you say are the keys to an amazing business model and amazing

business broadly? Like what should you be thinking about and looking into? I know you spend a lot of time thinking

about this when you're buying companies. What are like a few bullet points that you want to focus on? What I do for a

living is basically buy businesses. So in my early career, I started one business. Then I started about 10 more.

And then I realized starting businesses was extremely hard. And I I was doing pretty well. I was making quite a bit of

money and I'd sold one of my businesses. And I was looking out at the next 10 years and really asking myself, what do

I want my career to be? And am I happy doing what I'm doing? And the answer was no. I did not like starting businesses

and experiencing that failure rate. It was incredibly stressful. And so I picked up a book about investing and I

got lucky. The first one I b I bought was about Warren Buffett. And for those that don't know, Warren Buffett is kind

of he's counter to every VC kind of hustle culture thing you might hear because he basically just sits quietly

in a room and reads all day despite owning 260 uh different businesses and being one of

the 10 wealthiest people in the world. His life is incredibly calm. He only does what he wants and he spends most of

his time quietly reading and once or twice a year he makes a big decision to buy a business. And so when I read about

Warren Buffett, I just thought, "Wow, I'm a sucker. I'm running around like a crazy person trying to run all these

businesses. Why am I not just buying businesses and letting them run?" And so when I'm buying a business, what I'm

really looking for is something that I can't mess up. Right now, obviously, we buy a business, we try not to mess it

up, and we're very intentional, actually, very odd in that when Tiny buys a company, we generally just leave

them alone. If they already have management in place, we say nothing changes. You know, no one should know

that any that we've even bought them. Uh obviously, people know that, but but really like there should be no change

whatsoever. The only change that we generally make is if the founder wants to leave the business, then we'll bring

in a CEO to run the company. And that's probably the most important decision that we make. So I'm looking for a

business though where it is so good that it's hard to mess up. And most businesses are very easy to mess up. You

know, like one person leaves and the whole business falls apart because it's held together with dental floss and duct

tape. And so I'm really looking for what Warren Buffett would call a moat. So a moat is basically a brand. So that could

be a brand like Coca-Cola, Tylenol, Advil, something like that. So something that gives you pricing power where you

can consistently charge and you have loyal customers or uh where we usually find a lot of opportunity is in network

effects. So typically we're looking for a community that's gotten so big where people don't want to go elsewhere. So

for example, we own Letterbox, which is the largest social network for film lovers. And the question is, if someone

else wants to compete with us, why would someone go to their social network that has a small number of users when all

their friends are already on letter boxed? And so you see that same kind of thing with Instagram or Facebook or

similar. And so we're looking for businesses like that, something that has staying power that is hard to compete

with and that we can hold over the very long term. It was interesting when Elon bought X, how that was such a test of

the power of network effects. If you think about it back then, he changed the name. So the brand completely changed

the team, 80% of the people left. Like what was left? It was the network and the network effect. And the the simplest

way just for folks that aren't super familiar with network effects, the way I think about it is just network effect is

where every additional user that joins the network becomes more useful for everybody. There's another mode which is

high switching costs which I don't really like because it's it's not really very consumer friendly, but an example

might be Salesforce where no one wants to they've they've spent years and they've done all the implementation and

trained everyone on it and it's kind of the standard. Everybody hates it, but ripping it out and switching is such a

pain in the ass that they just won't bother. You know, that's another form of sometimes a great business. But again,

that's more depressing. I don't love that one. What's the last company you guys bought? Uh, we bought Sorado. So,

Sorado is the largest DJ software company in the world. So, if you ever see a DJ playing and they've got a

laptop in front of them, probably using Sorado to DJ. Wow, that is super cool. What like a fun thing to to I used to I

used to DJ so I I was aware of them and it was one of those moments we talked about earlier where you know I

understood the business. They had uh a really interesting mode of this huge passionate user base and deep hardware

integration and uh and I also love DJing and I was able to analyze the business and you know luck comes the prepared

mind and it worked out. This is another great call back to your excellent piece of advice of just ideally what you're

working on whether you're starting the company or buying the company is you have a unique unfair advantage or some

kind of unique background that connects to that idea. Let me ask you one more question before we shift to a whole

different topic. Uh, hint hint AI. Something that I know you talk a lot about, something you believe strongly,

and something you deal with a lot is when you buy companies is just is people and the the challenges of of management

and hiring and people. So, what have you learned about just uh how to successfully I don't know find amazing

people, keep amazing people, help people be more successful within the companies that you operate? My business partner

Chris has a really great quote. He says, "There are no problems. there is only people problems and so we've found that

you know you could tell us that there's a disaster that you know one of our business units is about to fail or

something like that as long as I'm surrounded by really great smart people I feel fine right but when we have a bad

actor we have a psycho or a narcissist or a really horrible difficult person we're dealing with all bets are off and

life becomes incredibly stressful so I'd say I spend probably 20% % of my time just trying to make

sure we're filtering people very carefully and ensuring we're working with people that we enjoy. Uh, and I

could talk about it in the AI. I've done some really interesting stuff to help me screen and identify difficult people.

But I've just found that the biggest mistake I made in my early career was I would hire people because I like them on

gut and I would think that I could change them. So, it's kind of like in romantic relationships. It never works

out well if you, you know, get married to someone and you go, "Okay, well, this is going to be a great relationship as

long as I can convert them to Christianity or I can change their parents or their personality or whatever

it is." And so, I've just found that I've never been able to change someone. I've never You can never mentor someone

out out into being a good employee. And the the kind of heruristic I have is if I ever think should I fire this person

even once I should fire them immediately. Uh it has always been a signal at least for me that when

someone's a superstar I can't imagine firing them. I think it's it's impossible the I'd be lost without them.

But when I repeatedly start going man is this person gonna work out almost always within six to 12 months uh it doesn't

work out. And so I really try and be really direct about stuff like that and just cut when it's not working. So I'd

say like the lesson really is hire for what you need. Don't hire just for potential, which is counter to what a

lot of people say. A lot of people say hire for potential and coach them into being whatever you need. I just haven't

found that. I I think you kind of need to have hire someone who already is fully formed and can do what you need.

But again, everyone does it differently. This is big advice. Powerful advice. Is this is this true? Not just for like the

CEOs of the companies you you run is for folks under lower down the ladder too. Well for CEOs one of the most

interesting things we've observed is so so I'll give you an example. So we had a CEO come in and we were interviewing

them and he my in my opinion the business needed to grow via organic marketing. In his opinion it was

enterprise sales. And so when we're interviewing him, he keeps going back to his experience building with enterprise

sales. And there's this great saying to a man with a hammer, everything looks like a nail. And so for him, he's

looking at this business and going, "Enterprise sales, that's the way to grow." I hire him, but I say, "Look, if

we're going to hire you, we need you to do the organic marketing." Lo and behold, of course, he goes off and he

does the enterprise sales thing. And I've just found that hiring CEOs is like you're they're an elephant and you're

the rider. They're going to go wherever it is that they want to go. And so listening incredibly carefully to

people's words and their experiences because generally people will do the thing they tell you. So what I look for

when I interview a CEO now, I want to be nodding along. I want to go that's exactly what I would do or that's way

smarter than my idea. and then I just leave them alone because I've found that anytime I try and pull pull them in a

certain direction or coach them or whatever, it just doesn't work. And again, this could be my problem. I might

be the world's worst coach and mentor, but for me, that's been what's true. I imagine there's also an element of they

won't love the job if they want to be doing say enterprise sales and then they're like creating viral Tik Tok

videos all day and tweeting. They're like, "What the hell?" Well, I've also found people will shoot themselves in

the foot, right? if they if they if I tell them an idea in a board meeting and I say, "I really need you to try this,"

it never works because usually they kind of sandbag it, right? They don't really put their heart into it or they they're

just kind of placating me and they don't want it to work. They want their idea to work. And so I've learned not to do

that. This episode is brought to you by Miro. Every day, new headlines are scaring us about all the ways that AI is

coming for our jobs, creating a lot of anxiety and fear. But a recent survey for Miro tells a different story. 76% of

people believe that AI can benefit their role, but over 50% of people struggle to know when to use it. Enter Miro's

innovation workspace, an intelligent platform that brings people and AI together in a shared space to get great

work done. Miro has been empowering teams to transform bold ideas into the next big thing for over a decade. Today,

they're at the forefront of bringing products to market even faster by unleashing the combined power of AI and

human potential. Teams can work with Miro AI to turn unstructured data like sticky notes or screenshots into usable

diagrams, product briefs, data tables, and prototypes in minutes. You don't have to be an AI master or to toggle yet

another tool. The work you're already doing in Miro's canvas is the prompt. Help your teams get great work done with

Miro. Check it out at miro.com to find out how. That's mirror.com. Let's talk about AI. Uh, it's so funny

to watch you on Twitter. Clearly, you've become AI obsessed, AI pills. I don't know what the term is. And I love that

you're not just the kind of person that just talks about it and tweets plantifications. You're like clearly

using it in like every facet of your job and looking for more ways to use AI. So, I have a lot of questions here because I

think this is where more and more founders are going to be where you are today. First of all, just what's in your

AI stack these days? like what are the tools you find yourself coming back to most or finding more most useful? So the

primary tool I use is a tool called Lindy lindy.ai. Uh and what it basically lets you do is

build workflows and agents. So it's very simple. You might say when I get an email in the Gmail API, uh I want you I

want to add an AI agent that reads the email and labels it based on that. Right? Or you can build a crazy Rube

Goldberg machine that sends different emails all sorts of different places. So for example, uh when I get an email and

it's related to my kid's school. One of the big problems I have is that my kid's school sends so many emails, you know,

the field trip is on Thursday and you need to bring the following things and meanwhile there's parent teacher

interviews and all these things get added to my calendar. the AI agent automatically takes that, puts it onto

my calendar, makes sure there's notes for all the things I need to prepare and stuff. So, like I have just basically

tried to take every single thing a human could do in my inbox and automate it with Lindy. And you're sitting there in

Lindy doing this yourself. Uh, or do you have people that you've kind of trained to help build these sorts of things for

you? No, I do it myself. Okay. How many agents, if if that's the term you want to use, do you have running for you? Oh,

man. I think I probably have not not a crazy amount within let's see each well each workflow probably has four or five

agents. So for example my inbox has four or five agents and then I have a whole bunch of other you know I've got a

calendar agent, I've got a meeting agent, I've got an email agent, I've got uh a scheduler, I've got a CRM and

basically they're all different workflows with a whole bunch of agents within. Okay. And you're not even you're

not an investor in Lindy, right? You're just a fan. Super easy. No. No. Fun fact, I actually am. So, this is great.

That's awesome. A tiny angel investor. So, just want to put that out there, which is I love that this came up

organically. What are a couple other uh really interesting use cases, workflows you've built that people might be

inspired by? Let me let me go through the email one a little more. So, basically, emails come in and the first

agent says, "Does Andrew even need to see this?" Like let's say you're on a thread and you've already chimed in and

everyone's just saying cool that works. I don't need to see that. So it just archives that never never. So that

immediately reduces my email about by about 20%. Then it decides is this something that is time sensitive? Is

this something that needs to be dealt with in the next 24 hours? And if so, it labels it in a special 24-hour label. So

when I go to my email, one of the biggest stresses I find is I go, [ __ ] there's 200 emails in here and if I

don't go through all of them, I don't know if there's an emergency burning somewhere. So now I do. Then it takes

any email uh or every single email and it decides is this a simple decision. So for example, let's say you email me and

you say, hey, do you want to get lunch? So it emails me privately and it says, hey, Lenny wants to get lunch. Do you

want to say yes? Do you want to say yes, but in a few months? Do you want to say no? How do you want to say no? And it

gives me multiple choice. I just can say four, like number four. And then it'll email you as me and it'll write out a

nice thoughtful email or whatever. So stuff like that is like so freaking cool and helpful. And I'd say that it's

replaced stuff that my full-time I used to have a full-time assistant just working on email. That's all completely

automated. And there's all sorts of other cool stuff I'm doing there. Other agents, like I'm working on one right

now to manage my calendar, which is really complex. I do a lot of scheduling and a lot of rules. I have another

really simple one. It adds emojis to every single calendar event. So, when I look at my calendar, I've got beautiful

little emojis for every single calendar event. Very silly. What are the emojis? Like, do they represent something? Yeah.

So, if I write weightlifting, it'll just do like a, you know, a guy weightlifting or whatever. I've got one that this

one's cool. Basically, what it does, every time I email someone, it looks them up online. It figures out what city

they live in. It checks my CRM in Air Table. It says, "Okay, I don't know. Where do you live?" In uh in Marin, in

the Bay Area. Okay. So, it' say, "Okay, you email Lenny, he lives in Marin. It puts in the database you live in Marin."

And then next time I travel to the Bay Area, the agent will see two weeks before and it'll say, "I saw you're

going to to San Francisco. Here's all the people in San Francisco you should try and have coffee with. Right. So,

just like all these things that Yeah. Yeah. Exactly. Like all these things that I've dreamed of having my

assistant do that my assistant could just never consistently do because she gets distracted. We're doing an event.

There's something going on. It's like having the world's most reliable employee who costs $200 a month and

works 24/7. Okay. I want to follow that thread. But first of all, what other tools do you have? What are Lindy is one

what else do you find really useful? So the other one is um Replet. So Replet is basically a vibe coding platform. You

can literally go into it and say you know I want to make a website for my SAS software business. Here's you know a big

document with a bunch of details and it'll go and design a pretty impressive website. But then you can also build web

apps now. So you can you can literally be like, "Yeah, build a Python web app that does XYZ." And the design is

getting so good because it uses Claude 4. You can start saying, "Do it in the style of Stripe. Um, think this through.

Rewrite all the copy and the tone of David Oglevie or Malcolm Gladwell or whoever you like." And it does a really

incredible job. So what I'm finding is that things that previously would have frustrated me because I would have had

to rely on a team of five, I can just do entirely on my own when it comes to web projects and stuff like that. So I'm

having a blast with that. That's cool. So you're team replet versus lovable bold v0ero. Is that roughly? I mean

they're all great. I I just am the most familiar with replet and it seems like it has the most functionality. like it

has a lot of uh beef to it, whereas Lovable and Bolt seem a little more basic. You might have to deal with uh

deployment and stuff in a fiddlier way with them. Awesome. Okay. Anything else? Um, another one I love, so Limitless, I

don't know if you've seen this. I've I've got one of those. It's really cool. So, you know, clips to your shirt. I

actually just put it on my pocket so no one actually notices that I've got it on usually and I just have it on all day.

And what it does is it just records everything I do. And then I can say, you know, the other day I had a coffee with

someone and I'm the kind of person where I go a mile a minute and I say, "Oh, I'll I'll send you that and I'll do this

or whatever." And so at the end of the day, I can just say, "Hey, what did what did I promise to people today or me even

better, everyone loves this use case, I have a fight with my girlfriend." And she says, "You didn't say that." And I'm

able to say, "Well, actually, you know, you did." and and you can query it as an LLM. So you can say you're a couple's

counselor. Look at this fight. What did each person need? How did it start? What are the key lines? Where did it change?

How could you do this better? And so honestly, from a relationship standpoint is probably where it's been the most

useful. What's cool, too, is that has an API. And so eventually, I haven't gotten there yet, but eventually you'll be able

to just have it record your entire day and then automatically go into your to-do list or whatever it is and add

to-dos or send emails or whatever you need it to do. Wow, this is so fun. Uh or just per your point with the

relationship, ask it, what could I have done better today? Where did I where did I where did the day go downhill? Yeah, I

saw a guy on Twitter. He said, every day I ask it, how could I have been a better dad? And it'll be like, "Oh, your, you

know, your daughter tried to get your attention about this. You, you know, you should have paid attention or whatever."

And that was with the Limitless data. Oh, man. I love that with the relationship you didn't go where I

thought you were going to go where it's like, "Oh, but who actually said it? What did you actually say?" I love that

it became much more wholesome of just like, "How could we have communicated better holistically?"

Okay. My wife actually is like, "Don't don't wear this around. You just got to wear it really covertly." It honestly

honestly I know it sounds kind of cheesy because you think like oh it'll tell her right but in reality usually you're both

guilty in the fight and my girlfriend has actually appreciated it because there's been half the time it'll be like

oh yeah you know Zoe kind of said this and it triggered this but a lot of the time it's me too and I think I'm totally

right. So I I find it actually really helpful. other other tools. I mean, otherwise it's kind of the basic stack

of uh Claude and Chad GPT and Gemini. And Gemini I use mostly for things that are really um large. So, for example,

all my medical records I have trained on Gemini 2.5 just cuz the other ones can't really go through it all. Claude I use

for writing very extensively. I find it's the best for that. And then Chad GBT for everything else. In terms of

cool things I do recently, like I went through my entire medicine cabinet and I just took big photos that showed all the

different uh medicines and stuff and now I can say something like, "Hey, um you know, I'm really tired and stressed out

today. What supplement should I take and what dose?" Uh or I'll say, "Remember remember all the medications I'm taking.

Whenever I ask you a health question, always remind me in the context of I might be taking that medication." And so

a few times it said, "Hey, you shouldn't take this other medication because your genetics say you're not compatible with

it and you're already taking this other one." So stuff like that's been really helpful for that use case. Do you come

back to a specific conversation where you've given it that context? Use a project. When I say remember, I just say

whenever remember these supplements, this is what I have in my medicine cabinet or this is what I take every

day. I see. So just tap into the memory of chat GPT. Yeah, this is awesome. This makes me uh remember that I I've started

to use and this connects to other points you've been making. I've started to use Chad GBT deep research to prep for these

conversations. I used to have a researcher who did this for 400 bucks. They did research on every guest that I

had 4 to 500 bucks and gave me a whole doc. Here's everything about them. Here's their background. Here's

questions you might want to ask. Deep research is better just as of recently, especially with 03 Pro. We were talking

earlier about agents. I forgot one of my favorite agents. So it looks on my calendar 30 minutes before I meet

anyone. It goes on to perplexity and it does a deep dive on the person and it sends me it goes into my email and it

gets the context of the meeting. So often I book meetings 3 months out. So I usually look at my calendar and I go who

is this person? I don't even remember booking this meeting. It'll text me before and it'll say here's who you are.

Here's your background. Here's the email context of why we're meeting. And it can and it basically does all the research

for me. So, and I I do the same thing. I also use deep research and say like give me if I'm meeting them, you know

everything about me. What are our commonalities and what are all the fun things we should talk about? Oh my god,

I just I just want to plug and play all these agents you built to use them myself. Uh, interestingly, I tried to

create an agent to do this research kickoff for me, but I don't think you can automate chatbt deep research. Okay.

Only Yeah. Okay. Okay. Perplexity is great, too, but it's different. Okay. Okay, so let me follow this thread I was

going down which is around job displacement. I know you think a lot about this. So with me, this researcher,

it was a contractor I was working with for a few months. I no longer need them. Your assistant, you no longer need.

Thoughts on just the impact we're going to see on jobs as a result of AI. There's this great uh William Gibson

quote, the future is already here. It's just not evenly distributed. And I think that we are in the Palm Trio phase. Do

you remember the Palm Trio? No. So, so when I was, let's see, in 2007, I think was that when the iPhone came out, 2007,

2007, something like that. In 2007, there was this device called the Palm Trio, and it was like a little PDA with

a stylus and you, it had a little modem attachment. And so, what you could do is you could get your email anywhere. And

that was this shocking thing. It had a you know, black and white or barely color screen and you could go in and

email people. And I remember being walking through the mall buying shoes while answering business emails and

going, "This is the ultimate freedom. This is incredible." But the problem with that device was it sucked. You

know, you had to use a stylus. It's not a good user experience. And then the iPhone comes out and you're like, "Okay,

this is this is what I was waiting for. This is the real thing." But I got a little preview with the Palm Trio. And

so right now, people like me who have the time to the time and skill set to nerd out and build the agents can do

this. It's just not accessible to everybody. And I think that if it was accessible to everybody in terms of if

you could just open chat GPT and say, "Hey, ChatGpt, I run a business. Can you help me?" And it started asking you

questions and it said, "Oh, well, you know, you do sales. Do you want me to set up a sales agent?" Okay, great. Plug

in your HubSpot API. Now, tell me a few things and then give me access to all your data and then before you know it,

it's just a digital employee that's on the other end of the phone on chat GPT just like in the movie Her that you can

talk to and it can go do stuff. I think that is kind of the iPhone moment that we're going to have in sometime in the

next 5 years. And so I think that if AI doesn't progress we will see some serious job displacement like what you

already mentioned the translators the researchers the admin um some assistant jobs depending on the kind of work they

do those jobs will definitely be massively affected but I think pretty quick all knowledge work jobs could be

affected if the models scale in the way that they say they're going I mean if you trust Sam Alman and Dario Amade I

mean Daario Amade said by 2027 our models will be smarter than all PhDs any so in any subject that is a

staggering statement and he's been a very conservative kind of almost fearful voice in the AI world very cautious to

say things like that and so when he said that I kind of perked up and started listening carefully. I don't know that

that's going to h how it's going to play out. He might be trying to fund raise. We have to take it with a grain of salt.

But if that's true, then I think so many knowledge work jobs are going to change massively.

Okay, there's a lot here. I actually had Mike on the podcast, their CPO, Mike Rger, and he pointed out that Daario,

every prediction he's had so far has been right over the past few years. And so there's reason to listen to his

insights. Okay. Andre, I know you're not going to have all the answers for people, but let me ask you a kind of a

two-part question for new grads and for people currently in the workforce, say like I don't know, senior product

manager, advice on what they should do, where should they focus, what skills will matter most, what jobs do you think

will last? What advice can you give us? You know, everybody 10 years ago, do you remember there was this whole movement

that we should teach kids how to code? Remember that? So now teaching kids how to code is like teaching kids basic and

we have a guey right you don't need to learn punch cards and basic programming because or MS DOS because we all have

keyboards and mice so we don't have to do that and I feel like coding has really gone away because of that and

I think now people are saying people need to learn how to prompt which I think is very true for the tool set

right now but I don't know how relevant that is in the future because I think we're still again at the palm trio

phase. I think that the AIS will be very good at eliciting what the actual problem you're trying to solve is and

reinterpreting it in a way that it can optimize the query to get the right answer and solve the problem. And so I

think the fundamental question is do all jobs just become a single prompt? So for example, does a CEO just become uh grow

the business while making the customers happy and turning a profit or or something like that and it is able to

actually be an omnisient presence that can run a whole company. Now that's a big I'm someone building AI agents. I

can barely get it to reliably do you know calendar entries and that kind of stuff. So I don't think that's totally

imminent. But I can certainly see a world where that's coming. And so the question is what is one to do when that

reality is barreling down at you? For people who have resources, I think there's a lot of things they can do in

terms of investment. They can invest in the companies that'll benefit from this. Companies that have a lot of compute or

energy or that sort of thing. So that's one thing, but that's not really the question. The question is what do you do

if you're a smart 18year-old or 19year-old or whatever. In my opinion, I think the best thing to do right now is

to get incredibly good with these tools and utilize them to build enough wealth that you can put the money into

diversifying into uh compute and energy. And I do think that while everybody thinks that, you know, all jobs will go

away in 5 years and robots will be everywhere, people generally overestimate things in the short term

and underestimate them in the long term. So, I think there's going to be this long window where robotics is not

anywhere near good enough to even do like raking leaves and stuff. And I think there's going to be a ton of

opportunity for people to just spin up new businesses that never existed before. Like I was talking to some

friends and we were like, you know, what do you do in a world of abundance where, you know, everything is really cheap and

and companies operate almost autonomously? I think there's a lot of weird skill sets that we can't imagine

right now. Like for example, just being funny, right? Being funny to hang out with. You know, think about Only Fans,

right? Right now, there's this whole idea of Only Fans or or where people are paying for comfort and connection, but

in a romantic sense. Imagine people did that dayto-day. They just love hanging out with funny people. So, you have a