Subscribe
Share
Share
Embed
Founder Reality with George Pu. Real talk from a technical founder building AI-powered businesses in the trenches. No highlight reel, no startup theater – just honest insights from someone who codes, ships, and scales.
Every week, George breaks down the messy, unfiltered decisions behind building a bootstrap software company. From saying yes to projects you don't know how to build, to navigating AI hype vs. reality, to the mental models that actually matter for technical founders.
Whether you're a developer thinking about starting a company, a founder scaling your first product, or a technical leader building AI features, this show gives you the frameworks and hard-won lessons you won't find in the startup content circus.
George Pu is a software engineer turned founder building multiple AI-powered businesses. He's bootstrapped companies, shipped products that matter, and learned the hard way what works and what's just noise.
Follow along as he builds in public and shares what's really happening behind the scenes.
New episodes every Monday, Wednesday, and Friday.
George Pu (00:00.27)
Hello and welcome to another week of FounderReady Podcast. I'm your host, George Pu. This week, we're going to do things a little bit differently. I'm going to share with you some resources that I've read, I think are really, really helpful for founders and giving you a little bit of thoughts about how to run a company this year in 2025 and why building this year is so different from the previous years. So let's take a look at some stories. So three stories specifically caught my attention this week.
If you connect the dots, think it reveals something that are fundamental or changing about how you build businesses. The first story is that founders are actually walking away from traditional VC companies. Not because actually they cannot raise, but because the old playbook doesn't really work for those newer founders anymore. Mercury, if you don't know, Mercury is basically a neo bank company that helps startup companies to find an online reliable source of banking. So they actually have a good relationship with our founders and they surveyed about
1500 early stage startups about how they're actually funding their companies this year in 2025 and the results are actually coming in. It challenges everything that we thought we knew about startup founding.
George Pu (01:12.814)
So a few statistics, I'll read it out. 66 % of founders changed their capitalization strategy in the past year. So 2024 and 25. So only 22 % of companies have raised under 1 million in their last round. 73 % raised under 5 million total. And most important metric is actually that companies are using four plus, you know, more than four different funding sources. They're more likely, 40 % more likely to raise 5 million rounds.
And about 60 % say that market condition has changed meaningfully, then it has dramatically impacted their funding approach. So the biggest surprise is that 60%, 1 % of companies are actually very reliant on contractor talent and not actually full-time employees and full-time hires. So that's very different from the playbooks that we read in the past 10 years, whereas into scale, you hire employees, you don't hire contractors and have fine co-founders, right?
So apparently what founders are talking about this year is very different. And I think the main story here is that this is not really about founders who are getting increasingly more desperate quote unquote. It's actually about founders being more strategic, right? If we look back at the bubble and before the bubble, the old way of raising a big VC round and skill fast playbook is actually pretty much dead. It's being replaced by something that's more sophisticated and founders are using different approaches of raising. They're actually being more careful about revenue, which you most
You know, in past days, and even in some cases now, you know, VCs don't care about revenue, right? But that's not the story today. Today, VCs care about revenue. And if you want to raise funding, VCs is probably not just the only way to go for funding. Some companies raise for debt financing, which is basically taking out loans from different banks. And some companies take loans from their customers and they crowdsource, you know, crowdfund. There are different, different ways of raising capital, you know, now this year. And I think that has completely changed the game.
My take on this story is that I've actually lived this personally with regards to venture capital. So I tried to raise capital in 2022. I was almost raising the round, but eventually that round fell through. And basically what happened is that I actually bought a startup by company and we decided to be revenue first. So initially, obviously I carried out my own money to fund the business. But as we go a little bit, I think a year after self-funding, we were able to reach the breaking point and we were able to be profitable. So it's not just about God feeling it, but
George Pu (03:32.822)
It's, it's actually about like not giving out as much control to other VCs, other investors, angels or boards, right? It's about as a founder, I, if I do this, I have more control. I have more options, right? There are more ways I can go about building my business or than just focusing on that one, you know, route of running my business. So this is actually very important and I have something to share as well. So personally, I've been recommending this new, newer approach of basically building your startup.
two founders and I actually have an e-resource ebook. It's coming out for free I think in the coming weeks. So basically what a book it's about, it's about my experience of building on a contrarian way. Right. So if you have, if you started a company, you probably have read about the book Ling Startup. You probably have read about Y Combinator and their blog startup school. Right. And in it, basically there's like a huge prioritization about raising venture capital, right. Or raising some sort of capital for your company to survive. And that is.
because these companies, actually, they profit based on you raising more capital, right? So obviously they have their incentives in the game. And to be also fair, it was the playbook for the past 20 years, right? What has changed is the COVID-19, the federal interest rates and the bubble busted. So now everything being started over and also with AI, the playbook is very different. So in my ebook, I basically tell founders about this approach. So
When you start, most people are say this, like start a SaaS product as they are starting a SaaS product and you're selling this to different customers, right? That's the old way to do things. So my approach and my recommendation to founders is actually start a product, start an idea, know, start an idea, but don't build a product yet. Bring your product to the ideal customers and try to go for like the main and large cap customers and try to sell a service instead of selling a product.
And the reason is because I've lived through, I've been done doing this model myself for the past three years. And just by selling a service, it's something your customers will immediately understand, right? It's not trying a new product and say, oh, you know, this is like $29 a month, $90 a month. just like, businesses basically are wondering, okay, how much time, energy and onboarding and employee buy-in do I have to do for this product? Whereas for a service, they can, they know they can rely on you. You're the person, you're the company that's just going to do most of work for the business.
George Pu (05:49.408)
So that is basically like a short summary about my latest ebook. So if you're interested, will link that to this episode and you can also go to founder of reality.com, go to the menu and click on ebook and you'll be able to find that as well. So I talked to my friends, a few founder friends about this just the past weekend. And I think they actually agree with my approach. And I think maybe this is the way that we are going to be building companies in the future. And in many ways, it's actually a good thing.
because we're no longer going to chase VC model, which doesn't really work for 99.9 % of the whole population who doesn't live in San Francisco Bay area or the major VC centers. VC is not the best way for everybody. And it's not a way for everybody, right? There's obviously a sort of fit and not fit in this as well. If you're a certain demographic, you're less likely to get VC funding as well. So all that in short, I'm just not anti-VC, but I'm more of all just think that.
There's obviously a better way of building your business and you know, maybe the consulting way is the way to go. And as this Mercury study showed, you know, founders are choosing these different sources. So let's take a look at together. So the first one is actually consulting revenue, which I find it surprising because you know, I didn't, I didn't know about that. I didn't know about that when I wrote the ebook. So it's good to get some validations from this. So consulting revenue is immediate. So many of those startups are returning to consulting first, generate some revenue.
And the second is like grants, which is like non dilutive, right? So if you get a grants from a federal government, US government, or like different governments, they're not taking equity. So getting grants and third is called strategic partnerships. So you work with different partners and they pay you to deliver a service. So that's like more like aligned incentives. And fourth is like the small angel checks. So it's still checks, but angels are much easier to write checks than VCs. So I think this is probably like a survival tactics. And especially when you're between rounds, you're like, let's say you have a pre-seed round.
And you're looking to read a seed round. obviously sometimes it's smart to get a few small angel checks. And that's the way that many companies, many of those companies do. So that's pretty interesting. I think converting that to like actual insights thing, your first 100K should come from customers. Ideally come from customers and not investors. Right. Once you have revenue, everything else has become significantly easier. And for you listeners, I think the question to ask this year, if you're looking to raise capital.
George Pu (08:07.178)
is that ask yourself like what kind of service can you provide today? That solves the same problem and ask yourself, can I sell my product or my idea as a service instead? Can I do it in consulting and generate consulting revenue? Those are all the things that you can think about that's going to generate significant revenue at first. And I think that's a good way to go about it. The second story, I think it has already becoming a trend, but it's important to actually take it out because this is something that's very likely to happen.
So if you know perplexity, obviously it's the answer engine is the anti-Google tool that you can, you know, get an answer instantly without paying and without ads at the moment. So basically the founder came out and talk about, you know, being copied by big tech. And that's something he's concerned about for the rest of the founders as well. So obviously in December, 2022, Pexley launched your answer engine, which is like a, you know, a search that gives you direct answers, obviously with real time web crawling and citations.
And also obviously using the LOM instead of just linking to Google. And in early 2023, obviously just a few months after ChatGPT has been introduced, it has been gaining traction and users really love getting actual answers instead of clicking through 10 blue links. So that's me. I absolutely did not use Google that much. I actually use Google just for the AI overview. I did not click on any links anymore. And before that I primarily used ChatGPT for like the instant answers. like, you know, my own preferences aside, think Perplexity is a pretty cool tool.
Google still has a lot of like, resources to have because Google still controls, the ranking mechanism. basically like when you search through chat GPT, a lot of links actually get blocked. So they do not allow web crawlers like GPT is called to crawl information. So, but they cannot do that with Google. So if you go to Google and use AI mode is it's different there. So that's like, you know, one significant difference. And perpexity basically was being copied by Google, by Google, by chat GPT.
And also little bit by Claude because Claude also added web search and Gemini obviously also added web search and Google is being very aggressive. Now when you search, obviously there's going to be like an AI overview at the top. There's also like this AI mode that, you know, just like as an AI mode, if you can try it out, I think it's been pretty good so far. No ads. Add to white company, there's AI startup school. So the perplexity CEO, Arvind, I hope I'm pronouncing his name right. Go founders. If your company is something that can make revenue.
George Pu (10:30.796)
on a scale of hundreds of millions of dollars. Well, not for most of us. You should always assume that big companies will copy it. They will copy anything that's good and you got to live with that fear. So I think, I think that's pretty important. And I think it doesn't probably doesn't just apply to products that generate hundreds of millions of dollars. I think it does probably apply to any sort of products that have some sort of interest, right? If you're too small, maybe they'll come with the acquisition offer or there's a copy or attack. But if you're too big, obviously, yes, I think he's right.
that big tech will find a reason to copy you. So, but here's a plot twist. perplexity, as we all know, they didn't die. They haven't, they're not dead yet and they're still growing and they're still innovating. And personally, I was actually pretty pessimistic about perplexity when they introduced like the browser, because I was thinking like, who is going to use that? Like an AI browser. And I think many other founders, friends of mine, and also in the tech community, they kind of feel the same way. But at the end of the day, I think I'm actually very interested in.
Uh, at the moment using perplexity is comet browser. And I think they're innovating and they're trying everything they can to innovate. Right. And obviously user experience is something big for them. I know a few people in my life who have subscribed to perplexity, which is a huge thing, right. For something that's like simple and simple solution in AI. And also like they have just partnered with some big partners as well. So obviously perplexity is not dying. And I think the question is not, you know, obviously from, from his perspective, the question is not like.
whether big tech will copy you, right? You should always assume inevitably the big tech will copy you. And my take on this is like, first of all, I wonder if Google and Amazon and you know, these other folks, like, are they still considered big tech or is like Chach-PT XAI, you know, with Grok and Claude and Thropic, Bionthropic, are those guys becoming, you know, like the...
big tech now anyways, because their valuations are ballooning and they are obviously adding more and more tools. I think at the end of the day, these competition obviously benefit the consumers, but obviously for a startup company doesn't have as much as money as like opening an anathropic. heard recently that on secondary is opening is raising for like a valuation of $500 billion. I think that's probably like the most valuable private company in human history.
George Pu (12:44.3)
And obviously they have a lot of cash and they are going around signing different deals with different companies like Oracle. So it's basically very incredibly difficult for startups to compete with startups. If we still consider OpenAI and know, like Thraupik and others as startups, that's, you know, that's not even including like the competition with Google and with others. So I feel like it's, it's good to panic about it or even be like significantly worried or have fear about, you know, being copied and trying to
build most through complexity. personally think it's like backwards. So earlier in a few episodes before I talk about simplicity, right? I talk about like how simplicity is actually our only solution out of this like AI first world, because everyone's going to try to build more. Everyone's trying to build more features, build more like user user loops, have more pricing plans. But I think what perplexity did is like they actually stayed true to the original mission. They're not trying to like out-engineer Google.
They're just starting their own application and make it load really fast. think that's one of the applications I've seen that loads the fastest, right? There's nothing else I think loads that fast. And it was the fastest throughout as well. If you ask a question, it gets you the answer right away. Whereas like even using ChachiPT and others, it's still a bit more like desktop vibes. So when I use like mobile, I sometimes do use perplexity as much. Other times I use ChachiPT.
So they built a brand around being anti-Google, which I think is pretty important because we are all fatigued about clicking like blue links and going through links, reading those like SEO articles. And there's all, there's this like fatigue, right? Which Google bought it to themselves. It's like either you see the Google ads, which is like a little bit useless most of the time, or you go through the different links and try to read through those SEO marketed articles.
doesn't really have a lot of truth in it. So people are tired of Google and that's why people are using perplexity to remove that complexity. So obviously perplexity is great and customer experience and it's not just technology because competing on technology these days, we can see that the development of LLMs are plateauing. So instead of developing their own LLM, large language model, perplexity declined that and they did build in-house models, but those models are based on other models, open source models and other not open source models.
George Pu (14:52.268)
And just about like optimizing for speed instead of spending like billions of dollars on training their own new models. So that's a smart strategy, not going to the LLM game because it's super, super competitive. And as we can see from GBT5, I can see a lot of plateauing around that as well. think we have definitely hit a little bit plateau here and there. So yeah, my framework for this is like, you know, don't really think about competition that much because for me, for Simple Direct Financing, like when we started out, I thought about competition quite a lot. competitor actually win public.
I think about a year after we started and then they went public is called Green Sky. know, it's like, think it had like 10 billion valuation or something. So I was like, what the heck? Like our competitors are actually doing so much better in valuation. So let's like see what they're doing well and let's maybe like copy some of the stuff they did. Right. But at the end of the day, that's not the way it goes. And I realized, you know, through paying that that's not actually the way it goes. The Green Sky is not succeeding in their game. They didn't get delisted, but they basically got merged.
by another company, they got sold off to another company for multiple different rounds after being taken private. OpenSax, I think it's the one that acquired them. They lost a ton of money by acquiring Green Sky. So the lesson is like, just don't, don't just like copy or just don't like copy whatever your competitors are doing. If you're a startup, there's obviously a reason why people use you instead of like the competitors in this space, right? Like why people use Ficamai instead of Adobe. There's that, in the early days, there has to be a reason, right? Like find that reason.
and talk to your customers and find a reason to double down on that instead of trying to be the competitor. Right. So I wouldn't be worried about being copied by someone. think if they do, I'll just find out something new in my, in the same space and same product because, because I'm the extension of product. I think I can always focus on customers and build a better product. And for perplexity, I do think the future is actually bright because I do not see Google and the others taking bold, you know, things and basically.
taking as both steps as they can. think that's like a huge thing. So I guess we'll see exactly what they can do next. And the practical applications for all founders is that try not to be afraid of big tech and don't build features big tech can copy, you know, as much as it can, or just like build it without fear, right? Build a position, a brand, build a customer base it cannot copy. Perplexity itself is not just like AI search. There are AI search for people who hate how Google works. I think that's the important distinction.
George Pu (17:13.506)
And that's why personally I like perplexity. So I just thought the second story is pretty interesting that, you know, founders should probably know about this story. So third story is actually the, I personally relate to, that's why I picked the story, but it's the ARR theater problem, the ARR drama. So if you don't know, ARR stands for annual recurring revenue and basically MRR, which is the monthly recurring revenue, basically like let's say a hundred people subscribe my product for a hundred bucks. That's tenth MRR.
And sometimes it just can just directly translate to $120,000 in ARR, right? Even though you have just closed as customers, you can stretch it out and say, this is my ARR. So basically just like time something by 12. So Fortune magazine actually published a recent article and an investigation about, um, about certain founders and startups who have been abusing the word ARR. And I think that should terrify a lot of honest founders. Here are some actual examples that I have found.
Clueli, which is a very controversial startup by many means, because they say every issues add cheat on everything, which is like a marketing way to say something. I think it has. I think double our attention, but obviously also controversies. So in recent horror, where it's basically back then and they claim to double their ARR from 3.5 million to 7 million in one week. So not like even one month in one week, they doubled their ARR and there's another startup for the fortune magazine and another VC partner up to dig in deeper. So there's another startup, which is unnamed in this case.
They claim $325,000 in AR, but when the VC dig deeper, turned out to just be a two week pilot contract where the co-founder had good authority or the founder thought, you know, the pilot customers keep paying, right? Translation is just like a handshake deal and a prayer that, you know, this is going to transit into revenue. And also like the part, the patterns that VCs and other people are seeing these days who are in the decision of whether to make an investment, right? They're seeing startup companies are using
counting on actually pilot programs as recurring revenue. They're claiming they have booked this ARR based on contract that actually it's very complex with some Q clauses, which basically means that like if the customer is not satisfied, they can get out of the contract at all in a pilot program and they don't have to pay you anything. So the customer can cancel anytime. And the VCs are basically saying that you should not be counting this as annual recurring revenue, right? That is basically like a misrepresentation and it's not how it goes.
George Pu (19:38.86)
So basically a startup shouldn't treat this and all other future possibilities as guaranteed income. So VCs are basically calling this vibe revenue, which I find it pretty interesting. And I think, and I think, yeah, startups should be called out for when they're using ARR numbers. doesn't make sense. I personally, I'm being very careful by revenue number. I don't use the word ARR. I do use MRR for simple like financing quite a lot, because I think that's a, that's a good representation of where we are as a business.
For ARR, think I would count ARR if we go back to 2024 last year and we count all the revenue for that year, which ones are recurring, I will call that ARR and publish it if I have to. So I feel like those new startups, just launch, they do some ads and they get a few customers who signed up and they time their revenue by 12 and call it like, you know, things. So that's still okay, but I think like having a pilot program with some companies and like they haven't even paid.
with a sketchy contract and just call that ARR that just like make it so much worse, you know? So I think, I think that makes a bad rep for all the founders. Um, and I think on a founder's side, I do understand, um, there's a lot of like pressure on founders to show growth metrics because we're in a different time zone. We're in 2025 where everybody cares about revenue. obviously founders do care about revenue. They want to chase after revenue and they basically just inflated their valuations, um, because they want to inflate their ARR.
But I do think this makes me a little bit angry. Um, and because I, it hurts legitimate founders who are actually building sustainable businesses. And when those founders claim ARR now, VCs are going to be like, can you prove it? Can you show more documentation? Like we don't believe you sort of thing. Right. And I think it hurts those founders who are actually building it, who are actually serious. And I just don't think this is like the way to go guys.
And I think in the future, if we see sketchy AR numbers, at least call them out and be like, oh, is it like actual AR or just like making things up? You know, the real number, I think is not like the name AR, it's a good matrix to keep track of things. It's like the pressure, right? Everyone's trying to be the next cursor, which, you know, a cursor was really actually attractive. They went from zero to hun- zero dollars to a hundred million dollars in AR in one year.
George Pu (21:49.972)
But everyone's chasing after the successor of cursor and trying to become cursor without actually understanding what made cursor work. you know, do a case study of cursor. guess like why it worked. They were the first, they were better than the competitors combined. They were better than get up, co-pilot. They're better than chat, chat, BT, they're better than Claude. And they're just good at what they do. I use cursor sometimes, you know, I love it. Right. And instead of like digging on why it made them successful founders are using the AR route to, you know, take shortcuts.
So I recommend don't do it. And you know, even for me, for simple drag, our early revenue just came from real customers pay real money. Monthly had a re we had an actual contract with sign with customers. We actually renew terms. So some, some contracts we send, they are annual contracts. So we do count that in the ARR. But I think, you know, if you're a VC, of course, into this podcast, I guess like if you're making decisions, look at like the term I call locked revenue. So how many contracts are signed? How much money do they have in a bank from the revenue perspective?
And which ones are the probable revenue, right? Which is the second category probable revenue, which is like strong pipeline with clear next steps. you can count as like, probably you're going to close it. And the third one is possible revenue, which is like basically everything else. So if you're a VC or investor or an entry investor, probably look at the locked revenue and look at a probable revenue and count that in as like potential MR and ARR, but basically don't take the inflation number, inflated number and just take it as like, you know,
whatever is real. obviously we can see that founders are optimizing for fundraising to the building, which is, you know, frustration, frustrated, I guess I'm on our side, but I can see how difficult it is to raise money right now. So I can understand, but it does make it harder for everyone else. let me know in the comments, what you think. think it definitely not the best way to move forward. And I think now after hearing those three stories, let's combine and see like how the startup playbook has changed.
Obviously the old playbook is like raise money first, raise VC money, enter money first, build first, build fast and worry about competition later, right? Scale like throw money in marketing, throw money at sales, throw money at business development, throw money on ads and just worry about competition later because you be first, you're the Airbnb, you're the Uber, you go first, you expand first, you take the market share and everyone else can go screw themselves, right? And then the third step is basically show hockey stick growth and raise more money, raise more around series G, series H.
George Pu (24:12.078)
you know, series F all those different series, um, you know, be like the next door dash, right? So those thoughts happen for a reason. Obviously think about like the companies that did work door dash, Uber, Airbnb, those examples, they didn't make investors rich around that time, which is like basically 15 years ago. Um, they started 15 years ago. So I do understand why it work then, but now we're entering a very different time in 2025, you need to build revenue streams that cannot be copied. And it's, it's easy for me to say it.
Right. In reality, obviously comes from relationships, comes from content authority, comes from positioning, comes from speed. have to find your competitive advantage in your revenue stream that cannot be copied. So I actually recommend founders think about like ecosystem and community approach. So ecosystem approach is think about like how you can build adjacent products that can actually make your ecosystem stronger. Instead of building one product, one service, can you build two? Can you build three throughout and make those connected to each other? So you actually can solve two, three adjacent problems for your customers.
Um, which, which can be very valuable, right? So that's called the ecosystem play. The second play is like the diversification play, right? So, um, try to be more diversified, try to build a community. I think that's the one I was talking about building a community, have loyal followings, right? You can be the founder who has content authority. You can build a community who actually rely on your and your business. So that's the way you build branding, right? It's a little bit of a heavy early investment, not, in terms of money, in terms of time, but I do think this is something that's going to work. So that's my two.
My personal two approaches of how to do this. And as a founder, obviously if you're building now, have to stack different funding sources, grants, different like early traction revenue, even doing consulting, you know, to keep your options open. Right. That's really, very important. And lastly, I think if you can, um, and I will definitely 100 % recommend it's to report honest numbers. Don't make up numbers. Don't try to like be smart about numbers because at the end of the day, there's still going to be due diligence. Right. And, um, and, and VCs are very clearly.
They're catching up to this ARR quote unquote scam about this, right? So don't be attached to this ARR number. Try to report your new growth, your actual growth number, your actual traction. So when I was raising three years ago, I had to be like, you know, I was very honest about where we at. And you know, like, I guess like you will find the real investors who actually appreciate you for where you are. And that's the way actually to build the shit down the line, right? Don't inflate it. You can build urgency without faking anything.
George Pu (26:33.422)
Any matrix and you know, here's my prediction about like where we are for 2025, 2026, 2027. I personally do believe that companies who are striving in 2027 will be the ones that started with consulting revenue, which is basically the same as I said in my upcoming ebook. It's going to start with consulting revenue. It's going to be amplified with customer relationships and a community that treasures them. And they will use those things as a foundation to make better business decisions about where going next about funding and competition.
So VC funding will still be around, obviously still important layer of growth, but it will not be the only thing that startups count on in the future. So I guess takeaways of this is like huge for founders, right? Like defend every dollar of your reporting revenue, take a look at your revenue. Obviously like, you know, don't, don't report a phony matrix. Don't hurt your credibility, right? Fix those numbers.
And think about the competition a little bit as well about like, try not to think about competition, but if you think about as well, like, oh, what happens when big tech come after me, right? If what, if they come copy my product, what other things can I do? And my point of Aussie from like earlier, just be authentic to yourself, be authentic to your product, to your customers, build a community, try to build an ecosystem and do your best, you know? So all those news as we have picked so far, I personally really like, so we'll probably continue this format moving forward every week.
We'll all look forward to something new. And if you have any thoughts, any news that you want to tell me, can always email me at george at founderreality.com. And also you can find me on Twitter as the georgepoo on Twitter. You can also DM me anytime. DM is always open about sharing me some news that you thought. So whatever you think about these three news that's happened this week, feel free to comment below and we'll try to also build a community throughout from the comment section as well. So if you comment, you can probably connect with other founders, you know.
I think that's the ultimate goal about this podcast. So thank you so much for listening for this week. For this episode, the next episode is Wednesday and the one after is Friday. So I'll see you then.