Y Combinator

On "Hacking" Y Combinator Interviews and Successful Startups

(Originally published November 2012.)

Some advice to companies interviewing at Y Combinator: stop focusing on hacks and just figure out your goddamn business.

Six months ago, some friends and I made a somewhat last-minute decision to apply to Y Combinator. Paul Graham told us our product was “too much of a ‘startup idea’ anyway”, and Sam Altman started our interview by telling us “this is the classic example of a bad startup idea”. As you may know, we got in, and are now working on Framebase.

So, what was our interview “hack”? We did everything we should have been doing all along.

Y Combinator holds its place as the most prestigious startup incubator for good reason - the partners are the smartest group of people you’re likely to meet. As you’d expect, it’s obvious that they’re tremendously reflective on how to screen for good teams. Put this way, which do you think is more likely: you’re going to actually discover the most optimal number of times to say “we talked to users”, or the partners are going to notice that you’re trying to work it in without much meaning?

Moreover, the interviews, stressful as they might seem, aren’t actually hostile to you - the partners are trying to decide if the team is a good fit for the program. Unless you’re trying to “hack” them. In which case, you’re now in some sort of stupid intellectual, statistics-based battle with the partners. Even if you somehow trick them, how long can you go on finding hacks to get around not knowing who will use your product?


The Y Combinator interviews are about showing the partners that you can do what you need to do to have a successful startup. Along those lines, here are some of the things which made us successful: [1]

1. Know your product inside and out. You should be able to answer 99% of questions about your market without delay, because you should have already thought about them. iPG is a great way to test your knowledge. [2] If you can’t answer any question on that page in 15 seconds, you’re failing.

2. Don’t ever talk over each other - it makes you look like you’re confused and don’t work well together. (It’s going to do this to your users, too!) If you have problems with this, you should be working on your teamwork in general.

3. Know your co-founder. If you don’t already, it’s probably a bit late for this tip, but you should be aware that you’ve made a decision not unlike marrying someone you’ve been dating for a week.

4. Be clear and concise. [3] If someone wants more information, they’ll ask for it. If you still aren’t sure if something is concise, try to take out 30% of the words.

5. Be committed. When the partners ask you if you will continue to work on the project if you’re not accepted, you need to honestly answer yes. Have an honest conversation with your co-founder(s) ASAP if you haven’t discussed your actual feelings on the matter.

Seems simple, but these things are often out of the comfort zone of the founders, and never get done. It may be more comfortable to try to hack the process, but building up a strong business is a more assured path to success.


On evaluating advice: Whenever you see advice for yc interviews, an easy way to determine whether it’s good or not is to consider whether it would improve your startup ignoring yc. [4] As an example, saying you “talk to users” as often as often as possible won’t help you in any way, [5] actually talking to users almost certainly will.

To those giving advice: Please stop calling things hacks unless they’re actually some sort of unconventional edge-case - it’s harmful. Teams should know about most aspects of their business; calling such things hacks is misleading, in that it implies it’s something they shouldn’t already be doing.



1. Evaluating my own tips against my evaluating criteria:

1.1. Sort of goes without saying, but knowing your product is quite useful for figuring out where you’re going, knowing what questions to ask users, talking to the press, and talking to any investors worth talking to.

1.2. Having a team which trusts each other and an environment free of people trying to one-up each other is a fairly well-documented necessity for success.

1.3. See the large number of horror stories.

1.4. Useful for talking to users, talking to press, talking to family and friends.

1.5. By definition, continuing to work on the product without YC is beneficial to working on the product ignoring YC.

2. I’m not saying iPG will solve all your problems. Almost every product has its own set of questions. Any smart investor is going to ask about those. What are yours?

3. If you’re not concise, you run the risk of the partners cutting you off because they’ve formed a questionably-accurate picture in their mind of what you mean. The same thing will happen in the real world - it’s called tuning out. In the real world, it’s often difficult to get the right balance, because users are stupid. YC partners are not stupid.

4. Not all good “hacks” fit perfectly in this criteria - Instacart’s, for example, doesn’t. As with everything, some critical thinking will be required.

5. In fact, saying you “talk to users” as often as possible is probably harmful, in that you’ll only bring on awful investors.

6. For the sake of completeness, I’ll mention that it is possible that attempting to “hack” the interviews is a good predictor of success. This runs against the advice I’ve been given by many alums (as well as my common sense), however.