How to Prepare for the Interview
The first thing to remember about Y Combinator interviews is that they're only 10 minutes long. We want to give as many groups a chance as possible, and that means each interview has to be short.
Because interviews are so short, we prefer that you not prepare any kind of presentation. There just isn't time for slide presentations, screencasts, or prepared speeches. We only do two things at interviews: we ask questions, and we look at what you've built so far.
One question we often ask is: What are you going to do? This is the most basic question an investor could ask, and yet you'd be surprised how often we get to the end of the interview and the answer is still unclear. So please, please, figure out how to explain in a few simple, marketing-free sentences what you're doing.
Don't try to craft a convincing pitch. We don't need to be sold on your idea; the details are probably going to change anyway. We just want to feel that the general problem you're working on is a promising one, that you have interesting insights into how to solve it, and that you have the energy to make a lot of progress rapidly.
The way to prepare for this aspect of the interview is to learn as much as you can about the problem you're working on. Prepare yourself, not a presentation. Ideally you want us to be saying after you walk out "they seem smart and energetic, and they really understand the domain." You can't change how smart and energetic you are, at least not in a couple weeks, so the best way to prepare is to learn as much as you can about the domain.
It's particularly good to go talk to potential users. We're impressed by startups who've tested their theories on real users, and can tell us what they learned.
It will also be useful to think about obstacles in your path. We usually ask about those. And we tend to be more convinced by a candid discussion of the difficulties you'll face than a glib dismissal of them. You're going to face obstacles; every startup does; so if you act as if there aren't any, it will seem to us that you must have overlooked them.
You should be intimately familiar with the existing options, and what, specifically, is wrong with them. It's not enough to say that you're going to make something that's more powerful, or easier to use. You have to be able to say how.
In some domains, even that isn't enough. It would be easy to make something better than existing dating sites, or eBay, or most enterprise software. But these are bad because they're protected by high barriers to entry. So it's not enough to say you'll make something better; you have to explain how you'll overcome the barriers that allow existing options to stay bad.
We don't expect you to have all the answers. So if we ask a question you don't know the answer to, don't try to fake it; just tell us how you'd go about finding the answer. A smart person trying honestly to answer an unexpected question is usually much more impressive than someone delivering an evasive or canned reply.
You can get an idea of some of the more general questions we might ask from the application form. Here are some more: Who needs what you're making? How do you know they need it? What are they doing now? What, exactly, makes you different from existing options? Why isn't someone already doing this? What obstacles will you face and how will you overcome them? How will customers and/or users find out about you? What resistance will they have to trying you? How will you overcome that resistance? What are the key things about your field that outsiders don't understand? What part of your project are you going to build first? Who is going to be your first paying customer? If your startup succeeds, what additional areas might you be able to expand into? Why did you choose this idea? What have you learned so far from working on it? Six months from now, what's going to be your biggest problem?
If you're already launched, you should know everything you can about your users. Where do new users come from? What is your growth like? (Bring a printout of a graph.) What's the conversion rate? What makes new users try you? Why do the reluctant ones hold back? What are the top things users want? What has surprised you about user behavior?
It might be worth practicing by asking friends to grill you as if they were potential investors. Ideally you want people who won't take it easy on you because you're friends—people who will ask the hard questions. If you know any founders of YC-funded startups, they'll probably know what to ask.
There's one type of question your practice investors probably won't ask, and we probably will: whether you've considered various mutations of your idea. We spend all our time dealing with startup ideas, so we'll often see potential variations of yours that might be interesting. Your idea is almost certain to change as you work on your startup, and while it's not necessarily going to change into something that comes up in the interview (though it has happened), we're very interested to see how good you are at traversing idea space.
The most important thing to us is probably the demo. And by demo we mean a working prototype of whatever you plan to build. Mockups can be helpful too, but they're much more convincing if they do something. Even if you have nothing now, you should be able to build something in the time you have.
(Demo = live. Please don't make a video. We won't watch videos.)
The reason we like demos so much is that they reduce the amount of guessing we have to do. A startup needs to have (a) good ideas (b) implemented energetically. And while it's fairly easy to tell from talking to someone how smart they are, it's much harder to tell how good they are at getting things done. On that dimension we're practically reduced to guessing. So anything you can do to show us how good you are at getting things done will make us much more sure of you. A good demo multiplies the effect of however well you answer our questions.
One more thing: please don't be nervous. That's a real danger in
these interviews, because they're so short. You just can't afford
to be distracted. We could try saying that there's no need to be
nervous, that we're pretty
people and not
particularly intimidating. But while YC alumni know that, 10 minutes
doesn't seem to be enough for it to sink in. So you're going to
have to solve this problem somehow. Whatever techniques you find
work to keep you calm, use them. Matt Maroon recommends a couple
shots of Tequila. We recommend preparation: if you understand
your area better than anyone else in the world, what do you have