Y Combinator Startup Podcast - How To Change The World? Get The Small Things Right

Episode Date: October 6, 2022

Dalton Caldwell and Michael Seibel talk about the importance of understanding incentives and doing research when it comes to building a world-changing startup.   To create Rookies Mistakes we asked YC founders: Is there a simple fact you wish you knew when you started your company or a rookie mistake you wish you could take back? Apply to Y Combinator: https://www.ycombinator.com/apply/

Let's say that changing the world is like uprooting a tree, like a big old, tall tree. Imagine there were two founders. One founder knew that trees have roots and the other founder had no idea, right? Like the trees with roots person, they have an advantage. They have a very large advantage. The tree without roots person, they're going to have a lot of very horrible, how do phrase this on the job learning. Yeah, it's not moving.

Like, we don't get it. Like, we're pushing it and it won't fall over. Yeah, we've done all the math. Like, if the tree ends right where the ground begins, like pushing like this, it should, it should work. Hey, this is Michael Seibel with Dalton Caldwell and welcome to rookie mistakes. We asked YC founders for their rookie mistakes so we could share them with you and help you avoid them.

Here's the next comment that was written by a YC founder. We built a product because we thought the world should be a certain way, not because we had a customer who actually wanted to buy the thing we were building. This is coming from a pure place, and we look at our heroes, right? We look at Elon Musk. We look at Steve Jobs, and we have this model of people that forced the universe to their will. And Elon, electric cars of the future, I am going to impart electric cars. I'm going to make it happen. And again, he kind of did.

good for him, right? But for most of us mere mortals, it's pretty hard to force people to want the thing that you're selling or the thing that you think is good for them. It's really hard if someone doesn't want something to like debate them and give in some they do actually want it when they don't actually want it. Most people don't have organic experience with sales. And so they kind of, their vision is like, oh, I walked onto a used car lot to look around and this genius salesperson convinced me to buy this car. That's like, that's not even how it works in real. Like, you didn't just randomly walk on the use car lot of nobody who doesn't want a car walks on the law.

You have a great point. Do you know what kind of idea we always come up, like I would come up with? Is guess what? Buying a car sucks and dealing with car salesman sucks. Buying a house sucks. Let's just like build a website and we'll get rid of car dealers. So this is an example of that. It's like you kind of run into something that's annoying.

And you're like, yeah, let's just get rid of that. And again, I'm with you. Let's get rid of that. You know, I don't like buying a car either. But I think what happens when you, again, haven't done the research. Yeah. Is you were like, oh, actually there's a reason car dealers exist. And oh, and actually there's incentives for these, you know, this is like, like the more you dig into that, you realize that you can't just wave a magic wand and make car dealers go away.

You can't just sprinkle software on the problem. Yeah, or like realtors, like buying a house. Why is it so hard to buy a house? Well, you know, I don't even know how to summarize that. But like a lot of founders want to believe they can just like start a company and revolutionize the home sales process. Right. So I feel where it's coming from. But in practice, the simplistic engineering efficiency arguments don't always work when applied to the real world.

I mean, you know about this from hiring, right? Like, haven't you seen this with hiring? I see this hiring all the time. I see the hiring startup applies to IC. And the thesis is always the same. I'm an engineer. I'm a genius engineer. And I am being assessed by a non-technical recruiter.

This is a crime. Like this is a crime against humanity. Like it's dehumanizing that I have to go through this experience. And if only the hiring manager was empowered with software, the entire recruiting team and all recruiters would just go away, disappear off the face of the earth through the hiring manager and software. And what's weird is that, like, I kind of believe the premise. Like, I believe an engineer is better at determining whether another engineer is good or bad.

However, I also look at the way the world is organized and I ask myself, well, if it seems consistently organized that these massive teams of recruiters exist at all these companies, there must be some efficiency going on that I don't understand, right? And the one core thing that I've learned is that even though. engineers are better at creating other engineers, there are two other facts. One, they don't tend to enjoy doing it. And two, it doesn't tend to be the greatest way to extract value out of a great engineer. Yeah, it's not the best use of their time. Like if you're employed engineers, you don't really want them to spend all their time doing this. You want them to be doing their jobs. And so what's funny is that like hiring companies inevitably run against this, like you aren't going to defeat HR. You aren't going to.

to get rid of all the recruiters, even though you don't believe they should exist in the world, they are going to exist in the world. Either you're going to accommodate them, or they will kill you. Like, they tend to actually have way more power than you might think.

And at the end of the day, the hiring manager is like, you know, this comes up a lot. Like, the hiring manager is your, like, talk friend, but not your action friend. Like, it's the person you talk to. It's like, oh, yeah, I wish I never had to use these recruiters. Like, I wish I didn't have to do this.

I always got all these bad candidates. But they never do, the thing you want them to do? Like, user parked all the time. To push back on ourselves, hey, what are Dalton and Michael saying? They're not saying you should try to change things. Like, shouldn't, aren't we, you know, shouldn't you be trying harder to make the system better? Let me, let me, let me answer that first. I don't know what you say, Michael. All we're saying is you can make huge changes in human behavior. If you really understand the subtle pressure points.

You've got to really understand the motivations of all of the key parties and cater to those motivations, cater to human institutions. Like the people that do really change the way the world works got a lot of small things right in the earliest stages. Like ride sharing, like the idea of getting to a stranger's car and them driving to the airport was like bizarre when it came out. younger viewers may not remember what it was like. But at the time, it was like, yeah, here's my idea.

A stranger comes to your house and you get in their backseat of the Toyota, right? That's normal. We would have, that was insane. And so to actually architect something like ride sharing, you had to really understand all of the people involved. They actually started off with those hired cars, those like limo cars, which was normal. Yeah. So there's all these like little things that had to happen for that idea to actually work.

and to normalize ride sharing that if you didn't understand all the core constituents, you're just like, hey, cars are inefficient, we should have more, we should let people share rides. How many times is YC fund at that idea before Uber and it didn't work? This was actually a really common bad startup idea was ride sharing. Isn't that hilarious? Yep. And so it's, I guess the point we're making here, and I, again, I want to hear what you think of this, Michael. You just need to understand the little things, folks. You need to understand the small motivations of all people in involved in an ecosystem and design the thing well, rather than just think you can burst in this,

burst out of the scene with something and like everyone will magically use your solution and boom, you know, hey, there's no more realtors. We can just buy and sell a house on a website. Like it's more complicated than that, man. You have to have a really, someone will solve us eventually. Maybe someone watching this video, but you have to really look at the intricate motivations

of every person involved and have an answer for why they're incentivized to use your thing. You know, I'm going to try to come up with the bad analogy on the spot. Let's say that changing the world is like uprooting a tree, like a big old, tall tree. Imagine there were two founders. One founder knew that trees have roots and the other founder had no idea, right? like the trees with roots person they have an advantage they have a very large advantage the tree without roots person they're going to have a lot of very horrible uh how do i freeze this on the job learning

yeah it's not moving like we don't get it like like we we're pushing it and it won't fall over yeah we've done all the math like if the tree ends right where the ground begins like pushing like this it should it should work that's a great way to say it there's so much under the There's so much another surface to understand with a lot of these ideas. And if you don't understand it, yeah, you're going to learn. You'll get some on-the-job training. That's a great way to say. Here was the next comment that was written in.

YC founder wrote, we did not do enough research before deciding what to work on. We now realize the more research you do in your space before you write a line of code, the better. It seems like people go to the effort of starting a company, of trying to raise money for a company, of applying to YC with a company, of making it to YC interviews with a company. And we can tell you haven't Googled it. You haven't typed in like the idea that you're working on into Google or read any articles about folks that have tried it before. And this is such a self-owned. It's free to do research on those that came before you and failed. cost you nothing. Doesn't take very much time. And to spend years of your life or months of your life

or any amount of your life on an idea that you've not done just basic cursory research on is really dumb. Do you understand? Like maybe you'll do it better or different, but to not even understand what came before you is kind of a willful ignorance. And I don't get that. Right? It's almost like, do you think, Michael, do you think it's that they don't want to know or that they just don't think it applies to them or, oh, this time it's different? Like, why would you obstinately refuse to research related companies in the past? I would say, one, this is very commonly a young founder issue. And one thing I see about young founders is their, like, internet knowledge doesn't go way back. So, like, I think the first mistake is, like, well, if I haven't heard of a company trying to

do this, that means a company has never tried to do it. I think a second thing that people do, is they ask their friends here. This is a situation where like Googling is probably a hundred times better than asking your friends. Because like if you're a young person whose internet history is like four years old, your friends are probably in the similar position. And then I think the last thing is a bit of a psychological thing. I think that's what you were getting to, which is if I'm so excited to do this, why would I want to learn something that would like defeat it before it started?

Yeah. You know, like can I live in the illusion in the dream for a little longer? And again, let me be direct on this one. During the dot-com bubble in the 90s, a lot of people started companies, and a lot of them tried pretty much every idea you can imagine. Right? Every idea that you see today has some analog from the 90s. So dig in, look into it. Like, see if people have tried this before, right?

Same deal. In the 2000s, people pretty much tried every idea. And again, maybe in the example of something like Instacart, people had tried Instacart before it didn't work. But the founder of Instacart knew that and they intentionally designed it to be better. You understand what I'm saying? Like they stood on the shoulders of those that came before them, smart. Versus if you're the founder of Instacart and you didn't realize that webvan existed, how is that hope? Like, you're not helping yourself to not know about it. Same in the 2010s. Think about how many startups we funded. They've tried stuff.

before. And so from our perspective, we see patterns and the ideas people start constantly. And we get it. You know, you're better, smarter, faster. You have a better programming language. You know, whatever it is, you know, we get your better. Daltin. I got better you. I got better you. You know, oh, mobile. I remember in our in our area when we were young founders, it was like, oh, we're an app. We're an iPhone app. iPhone changes everything, which is both true, but doesn't mean we should have discounted history. You know, Dalton, as you were saying this, something occurred to me. The platforms change faster than the human problems. So like you and I have lived through web 1-0, Web2O, mobile, and whatever the hell you call what's happening now. The Facebook platform,

you know, we've seen many of these. But turns out that like I want to go grocery shopping, but I don't want to go to the grocery store. That's been a problem the whole time. Like, or I want to get good food delivered to me in where I am, so I don't have to go get it. Same human I want to get a job. Same problem. And I think that's maybe the logical reason why you should do the research because it's almost guaranteed you're not solving a new problem. You're just using a slightly new technique or a new technology or something new to try to attack

an old problem. Let's study how everyone else tried to attack the old problem. And it's weird because, you know, sometimes, rarely, but sometimes you're doing something on the internet that hasn't been put on the internet that hasn't been done on the internet before. You know, with Justin TV, live video hadn't really been done on the internet before. What was interesting was almost inadvertently we got to learn about the old way because we started getting a lot of press coverage. And we started getting a lot of these old-fashioned video guys doing live video news coverage of us. And we would see them with their like van would pull up.

And it's like we beam this from the camera to the van to the microwave thing on that hill over there. It's got to all line up and we're like, oh, shit. And that was like state of the art. And so to me, it's like, wow, watching that, we were like, well, you know, our thing isn't as good as theirs, but it's way easier than having a van and a microwave dish up on a hill. So, like, now that plays devil's advocate a little bit. Dalton, I have an expert advisor. I have an angel investor. I have a seed investor.

I've got a series and investor. They're the experience ones. and they believe they're working with me, like, isn't that validation enough? Don't they, didn't they do this research? Kind of. I mean, let's go out of this two ways. One, it's best to do your own work on this because ultimately you're responsible for your company in it working or not.

And this is what I tell people in the batch. You know, I'm like, well, just make sure, you know, you have to live your life. So if your start doesn't work, it's not really my problem, you know. like I'll be okay. And so feel like you've done all of the diligence and research and you know the risks of what you're doing. But, you know, to the extent I have relevant advice, it could be helpful. I think the other thing is sometimes experts can be too biased positively or negatively based on a small data set. So if you're working on a startup idea that someone had either personally worked on before or lost lots of money investing in before, they're going to have opinions.

In my case, you know, media and entertainment startups are really. hard. So I, you know, I know a lot about them. I can advise people on them, but I'm not usually the most optimistic on that startup idea. And so when you're talking to experts, just make sure you get a wide variety of them, I think would be my reaction to your pushback and synthesize them. Just in the same way I said earlier, it's good to listen to lots of different kinds of music to synthesize together. I would just be synthesizing a lot of different opinions or a lot of different pieces of history as you're forming an opinion. You know, in the, in the Brex example, they went and spoke to every person that ever tried

to start a Challenger Bank before. And guess what they all told them? They told them two things. One, useful information that they wrote down in their notes the founders of Brex did. And two, they told them, don't do this. Are you crazy? So the Brex founders ignored that part. But this is similar.

When you speak to experts, you just need to know what bits and pieces to draw from what they have to say, then to blindly examine. what they say as facts. And that includes us. Totally. Right. You know, I think you brought up

something so counterintuitive about investors. Like, I get so many emails from people saying, like, I'm doing social this or live video of that.

Can you help me? And like, I'm one of those people who's like, wow, I took a lot of hits during my social media live video days.

Like, I think sometimes counterintuitively, investors who know slightly less about your space can be a little bit more grass is greener, like optimistic than folks who know more about your space. And I think that's not something that I realize. I don't mean that something it's obvious to founders. I think that like people assume expertise equals optimism. Yeah, right? Haven't we observed the opposite of that? Yeah. I mean, yeah, there's something nice about not knowing much about something. It's much easier to see it as a

simple question or a simple problem. Like, imagine asking the founders of DoorDash about food delivery and what they would have to say about it versus me. Versus a customer of food delivery. Yeah, exactly. Yes.