A Simple Method to Prepare for an Interview at YCombinator — Updated 2020

Gareth MacLeod
9 min readOct 29, 2020

I originally wrote this method out in 2017 while working with a small group of interviewees for the W2018 batch. Since then it’s been read thousands of times, and I personally know many teams who have been accepted after using these ideas in their preparation. The old version was starting to age a bit, so I think now, just before interviews for the W2021 batch, is a good moment to refresh it. I’ve added some new material, revised some of the dated bits, and added some pandemic-era specific advice. Enjoy!

My startup was accepted to YCombinator in April 2014, on our third application and third interview in two years. In that period, through our failures, successes, and a whole lot of talking to successful interviewees, we developed an interview preparation method that I think is quite good. Over the years I’ve taught this method to dozens of other interviewees with a high success rate.

The goal of this method is to be able to efficiently communicate about your startup. To get into YCombinator you need to have a good startup, but we’ll take it as a given for now. Our focus is to teach you how to speak about your startup well, convince people that it is good, and do so in a high-pressure situation. Once you can do this, it will be useful for a whole lot more than interviews. The same skill can be applied for investor pitches, bringing on early hires, and can even help you refine your own plans and strategy for the company. These side-benefits are part of the reason I advocate for spending a good amount of time prepping for an interview. Even if you don’t get in, you’ll be a better founder afterwards.

The framework of this method is to build two documents about what you want to say to the YC partners during your interview, refine them by getting feedback from other founders and mentors, and then to do lots of practice interviews to develop state-dependent memory. These two documents are:

  1. Answers to the questions they will probably ask you.
  2. A short list of the best reasons to fund your startup.

Next, I’ll describe each component in detail.

The first component is straightforward. While preparing for an interview, my co-founders and I would start a document that listed all the questions we thought we might be asked. With a little curation, we’d end up with a list of 50–100 questions. Then we would come up with answers for each of these questions, expressible in 1–3 short bullet points. We’d refine these answers by having conversations with people about our startup, especially YC alumni. Once we were happy with the answers, we’d memorize the bullet points and quiz each other. Sometimes we’d do hours of practice over several days.

The reason why you should prepare question answers ahead of time is because it is extremely difficult to be both concise and understandable without putting in effort ahead of time. There is an apocryphal Churchill quote along the lines of “If you want me to speak for five minutes, I can do it in a week; if an hour, I can do it now”. In the format of a ten-minute interview, a prepared 10 second answer is better than an improvised 30 second answer with the same information content because the 10 second answer leaves more time for the partners to think, ask another question, or for you to tell them other good things about your startup. This doesn’t mean you literally memorize a script, but that you have the key ideas in the front of your mind so you can phrase them easily day-of.

It is really not that hard to anticipate the topics that are likely to come up in your interview. Ten minutes is a short amount of time, and the types of things the partners want to know find out about you in that period no mystery. Traction, progress, competition, details on your team, and other general topics like the ones that are likely to come up every time. The questions on the written application are a good starting point. There will also be questions that are specific about your startup. For these, simple brainstorming will get you far — what would you ask if you were evaluating your own startup as an investor? What might the partners have concerns about, or be interested in? When you reach the bottom of this well, you can ask mentors or peers what questions they would ask. There are also lists of questions other people have curated, apps that will quiz you with a timer. videos of real interviews on stage, and lots of other blogposts like this one out there.

It’s particularly important at this stage to think hard about what questions you hope they don’t ask, the subject areas you most want to avoid, and may have even been avoiding asking yourself. The YC partners are very smart and there is a good chance they’ll quickly find the weakest part of your story and dive right into that. Having rough edges doesn’t disqualify you — every startup has them — but if you are not prepared to talk about these things, it will look like you have critical blindspots in your thinking and you will probably fail the interview. At a minimum you should be able to to show that you are aware of your weaknesses and are thinking about how to address them. A starting prompt to answer a tough question might be: “Yes, that is a challenge, here are some of the ways we’re thinking about approaching it”.

Having answers on the page is good, but it is different from having *good* answers down on the page. We discovered this after our first interview. We were prepared to answer every question, but we still didn’t get in. I guess we weren’t as smart as we thought. To figure out whether you are giving good answers, ask for feedback from peers, mentors, and ideally some people who have actually interviewed at YCombinator before. The best format for this is to actually do a mock-interview: set a timer for ten minutes and have them pepper you with questions from your document (or publicly available sources), and afterwards, ask them what made sense and what didn’t. If you do enough of these, you’ll likely find some patterns in the feedback that will give you a strong signal about where you need to improve.

The second thing you need is a short list of the absolute best things about your startup. I’m going to call these “Sparklers” for short. Sparklers are different for every team and project, but a generic definition could be “things that are really important for a smart investor to know, that might make a big difference in their decision making, especially those that they might not think to ask about”. It’s critical to prepare your sparklers ahead of time because you cannot trust they will come up naturally in an interview, so you have to be prepared to actively introduce them. The partners are looking for reasons to fund you, but in the span of 10 minutes the conversation might simply miss the information that would get them excited about your startup.

Here’s a short list of things that might be a sparkler: traction numbers, growth rates, conversion rates, patents, founder track record, unique expertise, Ph.Ds, connections in the industry, hidden insights, unfair advantages, brand-name customers, LOI’s, daily actives, long waiting lists.

A good way to come up with your sparklers is to talk to experienced people about your startup and ask what stands out for them. A lot of the time new founders don’t understand the investor mindset well enough to know what sounds exciting, but experienced founders can fall victim to this too. I’ve done mock-interviews with founders and only afterwards learned a critical piece of information like “We have an exclusive deal with General Electric” or “We have a million dollars in ARR”.

Once you have your sparklers in hand, make sure you say them in the interview. That sounds trivial, but the pace of YC interviews can be so rapid that you enter a trance-like state and forget. You may also be able to answer some questions in your prep document with a sparkler. A brief pause in the conversation is a great moment to say “Something else we wanted to tell you was…”. Another option is to answer a given question and then just keep talking and introduce a sparkler. As always, this is a good thing to practice in mock interviews.

Pandemic Era Specifics

In the pandemic area, interviews are a bit more complicated because they are largely done over video chat. All of the advice in the rest of this post still applies, but there are a few more things I can say specific to this period.

First of all: nail the basics of a multi-person video chat. Make sure you and everyone member of your team on the call has the absolute best internet connection and video chat performance possible. I would go as far as to upgrade your internet to the top plan available for the month of your interview if you can afford it. Close other programs on your computer before you enter the interview to give your chat app free reign over your computer’s RAM and compute resources.

Plan ahead of time which team member is going to answer which questions, or which topic areas of questions. It’s good to have every founder talk at least once in the interview — this is one way to show the partners that you work well together — but if founders try to talk at the same time you can end up wasting valuable seconds in the “You go” “No you go” process. This is already true when your interview is in-person but is gets even worse in video chat where sub-verbal social cues are hard to read. The solution is to partition your questions document into topic areas, and make each founder responsible for answering the questions in one of those areas. For my team, the partition was something like tech vs. product vs financial operations, but you can figure out something that makes sense for your team. Sticking to your topic areas and not talking over each other is a good thing to practice in your mock interviews.

So that’s it. Figure out what they’ll probably ask in the interview, how you’re going to respond, and practice. Figure out your sparklers and make sure you say them. This isn’t a universal method. I’ve only used it when applying with software companies, so it might need to be modified for other types of startups. Your startup needs to be good to get into YC, interview preparation only helps with the small part of the process where presentation matters. But in general, I’ve brought this up with lots of alumni and they seem to agree it is a good approach. Maybe it will work for you or you can adapt it to your own style.

A final addendum is: if you have a good interval between when you get selected for an interview and when it actually happens, you can actually use that time to make your startup better! If you can meaningfully increase your core metrics, sign a big new customer, launch your product or finally get your first paying user, these are all things that can make a big difference in your interview, and can show the partners that you’re able to make a lot of progress in a short amount of time.

The great thing about interviewing at YC is that it can be a positive experience whether or not you get in. It’s a moment to think hard about your startup through someone else’s eyes. What is working? What are you worried about? What do you want this project to become? You might find you learn things in this process that stick with you a long time.

Good luck!

Thanks for reading! As a reward for reading all two-thousand words of that essay, I’ve set aside time next week to do one-on-one interview practice with a few teams. Send me a direct message on twitter if you’d like to be included.