Dalton Caldwell's Whale AMA

Right now I'm interested in things like food, transportation, housing; the stuff that every person spends a huge percentage of their paycheck on every month. Um, if you look at how much value has been unlocked by things like Uber and Airbnb, I think there's opportunities to build companies as large as those on other things we spend a lot of money on. One example is healthcare. Um, obviously if you're a consumer of healthcare, you may not have had great experiences, and if you've looked at the bills, you know a lot of money is getting spent. Um, I think there's an opportunity to build Uber and Airbnb scale businesses, um, attacking healthcare for instance, as well as definitely in food.

My advice for people building content businesses, uh, having built them before, is to have a clever plan on how to monetize. I think a lot of people think if they build it, people will come, and if they have millions of users, they will magically get rich, and it's not that simple. Um, CPMs are going down all the time, and you have to be working against that. Um, so if you find interesting ways to monetize via, I don't know, native advertising, or directly being supported by the fans, those are pretty interesting to me, and I think it's worth at least having a plan about that in the earliest days and not just assuming, you know, if you get a million dollars that VCs will pay for the whole thing. Um, maybe that worked a while ago, but I don't think that works anymore.

So, I'm the head of admissions at Y Combinator, um, can so can tell you a little bit about the process. The most important part about admissions at YC is filling out an application. Uh, sometimes people think they have to know us or network with us to get in, and that's not true. The vast, vast majority of people that get into YC are just founders that fill out an application. Um, a top percentile of the folks that, uh, fill out an application are invited for an in-person interview with us, and that's where we get to know you a little bit more. Um, a lot of times people email us and almost like ask for permission before they decide to put in an application, and my advice to anybody that's interested in YC is just to do the application.

Um, so much of startups just is just doing the thing you're thinking about doing and not looking for permission, uh, from people before doing it. So if you're interested, you should just apply. I think the null hypothesis is that, um, things won't change and the music industry is just inherently going to be smaller forever; that the fact it was as big as it was was a fluke of history. Um, physical media was a very good business, and because we don't have physical media anymore, it may be that there's no way to replace the revenue that will make it the size that it once was, and that's okay. I think there's several other industries that were, uh, that have gotten smaller with technology as they moved to pure digital, and it could just be that the size of opportunity in music, um, will never recover, and that's okay.

Um, so I don't think we should assume that one day the time will be right when that's the day to start starting music startups again. If I were on the team that was working on Whale, um, I think the biggest thing that I would be thinking about is less, 'Here are our metrics, and here's how we're tracking them, and let's just grow x% week over week.' I'd be more looking at the data to figure out what it is, and sort of existentially why do people want this? Why do they want to use it? And make sure that this is a thing. Um, like is this the next— is this a Quora competitor? Is this, um, some kind of Q&A app for your friends? Is this for tech people? Is this for celebrities? I think I'd be looking at all of the early data and try to get a sense of what my gut told me as a founder the future of the product is, and only then look at the metrics and make sure I'm growing in that specific, um, in that specific vertical or use case because if you're just blindly looking at growth and you don't have a vision for the product, you could get lost.

I was a founder for a very long time, um, way longer than I've been working in YC, and I raised I think eight or nine rounds of funding over the course of my career, and so I can really relate to what it's like being on that side of the table. And I would say one of the big things I've learned is how many decisions that investors make that are not personal. Um, I remember sometimes worrying if, you know, maybe someone was passing on my company because they didn't like me, or they didn't, you know, that it was somehow personal to me. And what I realized being on this side of the table is it's not. And I think my message, um, if I was going to start another company, and to the people out there, is to remember how unimportant doing things like networking and feeling like you're well connected to investors matter and that if you have a good company with good numbers, if you have a great product, um, that's what's going to make people fund you way more than trying to influence them and network with them.

Um, all right, cold emailing investors, what do I think about that? Two things: one, of all of the ways you can spend your time as a founder, of all the ways you can add value to your startup, cold emailing investors should be like those lowest priority. For every hour you spend on that, hopefully, you've spent about a thousand hours working on your startup itself, writing code, getting customers. Um, a lot of people spend way too much time talking to investors and not nearly enough on their startup. Point number two, um, the whole reason Y Combinator exists and we have an application process is that you can just apply. You don't need to know us. If you try to cold email us, we just tell you to go to our website and apply. Um, so really no need to email anyone in the YC ecosystem, just go to our website and apply.

Thanks Kanye West for president. Um, here's what I think: I was actually at the Kanye show that happened in San Jose, um, right as the meltdown was starting. He played a couple of songs and just completely stopped playing any songs and talked to the audience for, not exaggerating, for something like an hour. And he was just talking about Trump; he was talking about how he respects Trump, how he, if he would have voted, would have voted for Trump. And I saw all this stuff in person, and I think to him it was so crazy to see someone like Trump elected that it doesn't sound far-fetched that someone like Kanye could be elected. And you know what? Um, I don't think he's wrong.

Most of the time, a good startup, um, or a startup that's in YC, can build the entire product, do sales, do everything they need to do with just the founders. The founders between themselves have the ability to execute, and so they don't have employees. Um, so most of the time, um, I would suggest not hiring a founding team at all and only, um, start to hire when the company is going so well that that's your bottleneck, and usually that's after YC for a lot of folks, um, months after YC. And so whether or not you end up in Mountain View or moving somewhere else, um, that's kind of up to you. But the short version is if you're thinking about hiring very early in the process, it's usually a sign that, uh, you may want to add some more skills to your founding team.

There's a lot of variability on what makes a good YC application, and I would never want to put out advice that would cause everyone to try to make their application look a certain way, but what I would say, um, one underlying quality you see across the board is clarity of communication, where the description of what are we doing, why are we doing it, why are we the team to do it, and what are our insights. They're able to state those things very clearly, and clarity of thought on the part of the founder and clarity of communication, um, are very predictive of a company being able to get an interview as well as getting into YC. And even after YC, it's a good predictor of the company, uh, will be able to work if they're good at communication.

What I've learned about failure working at Y Combinator is that, um, most of the time, the vast majority of the time, it's not a big bang; it's not something dramatic that happens. It's just finally you capitulate, finally you decide to give up, you decide to go get a job, and that, um, failure is pretty normal and it's not the big crazy boom that you would expect. And sort of what I learn and the way I think about this is that failure is, if we think about the scientific method, failure is the null hypothesis. The null hypothesis is that this didn't work, and that you should assume by default that's what's going to happen, and it's your job with the time that you're given to prove otherwise. Otherwise, um, the null hypothesis is true.

As far as the why now question on the YC application, I actually think you're, uh, misunderstanding it. The reason it's there is why is this the perfect time to start the startup? If you would have tried to start Uber before the iPhone, it wouldn't have worked, and I'm pretty sure people did. Um, Instacart was an idea that was tried during the dot-com boom with Webvan and others and failed pretty badly for a variety of reasons, but the time Instacart started was exactly the right time to start Instacart. And so what we're looking for to answer on the question why now is what is happening in the world? What are the other factors that make this the perfect moment in time, not too early or not too late, to start the startup, um, that you're working on?

Repeat applicants have, um, actually a pretty good chance of getting in. I don't know if everyone realizes this, but, um, a pretty decent percentage of a batch is people that we had previously, uh, not accepted, and they came back, and then, you know, the next time or two batches later we took them. Um, my advice is to make sure that you're making progress. Nothing is more exciting than to see someone have an idea that's really early, uh, be unlaunched and then apply again next batch, and they're launched, and they have customers, um, or they have revenue. The more you can show that your idea is working and you're making true progress and not fake progress, um, the more likely it is you'll be able to get in the next time around.

And you know, it's super impressive to see someone come back next batch and the business is in such a different state, you know, the founders are working really hard on it and making tons of progress, and so we'd love, we'd love to interview those. The biggest wild card for knowing, uh, if you have a good chance to get into YC, it's actually team questions, and I know it's hard to talk about your team in the context of this, uh, whale question. But, um, who you are, who on the team is technical, how long have you known each other, what are your insights on the product? Those are honestly the most important criteria that we're looking for in a company. Um, unlike a later stage investor where you would have growth milestones or revenue milestones, um, we're much more focused on on the team aspects than hard specific numbers that I could give you guidance on.

There's this, uh, really common thing in Silicon Valley where after a startup is successful, a bunch of stories and myths are created, um, to sort of imply that was their mission all along. And I think in the case of Google, um, there's no way in the first few years of the company they said that their goal was to organize and make accessible the world's information. That was like a mission statement that came much later. I think in the earliest days, um, it came out of research they were doing at Stanford. They had a clever hack, um, by looking at backlinks to determine how authoritative a website is, and it ended up being a really good hack. They wrote papers about it; they didn't want to start a company, they tried to sell the technology, they weren't able to sell the technology, and then they end up starting Google.

And the real story to me is a lot more indicative of what the future of, uh, you know, building a Google competitor, other ways to make knowledge competitive, uh, accessible is. Is people will just fall into it; it'll be happening on accident. The best way to talk about what your insights are is to tell, tell the story, tell the narrative of here's what our initial idea is, here's how we talked to users, here's how we experimented with it in the market, and then here's what we learned. Um, here's something that happened as a result of user feedback, because often the best insights are things that were an accident or something unexpected that happened by you trying it. Um, insights that are gained that way are always way better than ones of just, 'Hey I read a book about this and I got an insight.' Um, you know, a good insight is something that you yourself has that someone else that hasn't actually tried to do the thing, um, would not have figured out on their own.