How I Crashed and Burned in YCombinator

Kevin and I were walking around downtown Mountain View with Paul Graham (PG), one of the cofounders of YCombinator. It was a warm, breezy day in early July, three months after we had joined his startup accelerator. As PG strolled beside us, he gave us the news: “If this was a class in college, I’d be the professor warning you that you’re in danger of failing.”

We already had suspicions that our startup wasn’t going to be a huge success, but he confirmed them. We were in the most prestigious incubator in the world, and we were fucking up.

Let’s Rewind

Back in the spring, Kevin and I had been busily preparing for our YC interview. We had applied with a programming education idea, with the tagline “Codecademy on steroids.” I had countless friends who were nearly finished with their computer science (CS) degrees and had an idea for a company, but they struggled to move from concept to finished product. I wanted to help them bridge that gap.

I had known Kevin since the start of college, but it wasn’t until our final year that we started working together. We were a solid team, a designer/developer combo. We had a plausible idea, or so we thought. So we applied to YC because, well, it seemed like the logical thing to do. After spending nearly two decades as a professional hoop-jumper, my choices were a) go to grad school, b) work at a big tech company, or c) attend a prestigious startup accelerator.

The Interview

Despite the fact that YC interviews are supposed to be grueling, we emerged from our 10-minute interview relatively unscathed. The partners (Garry Tan, Kevin Hale, Michael Seibel, and Geoff Ralston) questioned our business model and our ability to execute, but we had convincing answers.

At this point most teams go home, and the partners make a decision by the end of the day. Not us. We were well on our way home when the partners asked us to drive back for a second interview. It was in this second round that I suddenly understood why people call YC interviews grueling.

With a new set of partners (PG, Jessica Livingston, Trevor Blackwell and Robert Morris), we had a completely different experience. We tanked. At one point, PG let out a long sigh and rubbed his forehead. “You’re halfway through your interview and I have no idea what your company does. This is not good.” I remember Jessica giving us a look that I could only interpret as total pity. We left dejected.

We left the YC offices with absolutely no idea whether we’d get in. Our odds didn’t look good, but just in case we went out and bought both a bottle of champagne (thumbs up) and a bottle of whiskey (thumbs down).

So you can imagine our shock when they told us we were in. We drank the bubbly.

(To this day, I remain convinced that our saving grace was a little-known fact that we demonstrated in the interview: when you bite into an apple-shaped plastic bottle of Martinelli’s apple juice, it sounds like you’re biting into a real apple. No, seriously, check it out.)

Impostor Syndrome

We were elated—briefly. What should have been a very positive mindset quickly turned negative. For the first time in my life, I had impostor syndrome: I convinced myself that I did not deserve the success I had achieved. I must have just been lucky, well-timed, or a particularly convincing charlatan. Surrounded by a cohort of people with more experience, confidence and traction, I felt like a total fraud.

From the outside, it might have looked like I was crazy to feel this way. I had graduated from Stanford, I had a CS degree, and I knew how to play the startup game. In college I had started an ed-tech company, ClassOwl, that raised nearly a million in seed funding. My two cofounders and I attended StartX, an incubator affiliated with Stanford, but ultimately decided to finish our degrees instead of pursuing the startup dream.

But on the inside I was anything but prepared. I was 21, I had somehow gotten myself into YC, and I had absolutely no idea how to get a company off the ground.

On top of everything else, our mentors had just impressed upon us that our idea was stupid. So we pivoted.

FanHero

Looking back, we should have tried harder to address the partners’ concerns. We should have found a way to use the code we’d written and the domain expertise we’d gained. Instead, we threw out all of our data, all of our work, and started over.

And I mean really started over. We researched Bitcoins, smart devices, 3D printing, biometrics, and a host of other industries. After several days of brainstorming, FanHero was born. We came up with the idea after discovering how little money most YouTube personalities make. Our thesis was that if you produced content that millions of people consumed on a regular basis, you should be able to make a living off of it.

Like many arrogant twenty-something recent grads, we assumed we could write a bit of software and disrupt an entire industry. We had no experience creating or maintaining an online following. And we had no YouTuber friends from whom we could get initial feedback. From the outset, it was an uphill climb.

We still had faith, however. If we could just gain a critical mass of users, iterate a few times on the product and get a TechCrunch writeup, we would solve all of our problems.

And then we “launched.”

Because the incubator is so prominent, TechCrunch winds up covering the launch of many YC companies. Few decline the coverage, as it’s free press. In our case, it’s unclear if we made the right choice.

In the words of the reporter:

So anyway, there’s this company called FanHero. It’s all about helping those YouTube guys make money in, like, non-advertising ways. Giving the community ways to support them through commerce — you know, selling stuff. It’s like the classic merch model, like how you go to your favorite band’s show and you buy a t-shirt or a CD.

The guys behind FanHero are these Stanford CS undergrads Kevin and Charlie, who like, grew up on YouTube idolizing YouTubers. These guys don’t remember a time when the world’s biggest stars weren’t on it. They’re in Y Combinator now because that’s where all the cool kids go to learn about the Internet and monetizing and stuff.

We had much bigger problems than the piece, to be sure, but launching with an article that roundly mocked us was one of the final nails in the coffin. That’s when we received an email, informing us that we were being summoned for office hours with PG.

The Meeting

He was brutal, but he was honest. Our company was in awful shape. We had less than five weeks to Demo Day, when we were expected to present our ideas before a massive gathering of investors—most likely the biggest such audience we’d ever have. So we pivoted. Again.

We went back to PG, and he told us to stop worrying about Demo Day. With the right idea, we’d be working on the company for the next five-plus years. It was foolish to sacrifice the quality of the company to optimize for the next five weeks. Instead, we should look for technologies that were interesting to us, as well as larger trends that we observed in the industry. To his credit, he gave us very, very good advice. When you’re a 21-year-old trying to create a startup first thing out of college, though, it’s hard to internalize that advice.

The last few weeks of the program are a bit of a blur. My morale plummeted, and my stress skyrocketed. I dreaded going to the weekly dinners, where I would have to talk to my cohort and confirm that no, we hadn’t found an idea yet and yes, we were still super excited about our “company.” I did little else but eat, code and sleep. I felt like everyone else was ready to raise millions, while we were desperately trying to find a product. There was no way we could compete with our batch mates on Demo Day.

A Way Out

The partners told us that we had the option to defer Demo Day, if we wanted. It wasn’t common, but other companies had done it. I felt an incredible sense of relief. Postponing Demo Day was a no-brainer. We didn’t have an idea we believed in, and although we could have cobbled one together and raised some seed funding, we didn’t want to do that. I didn’t want to do that, in part because it seemed disingenuous. But primarily it was because taking half a million dollars of other people’s money would have moved my stress level from unhealthy to crippling.

But skipping Demo Day didn’t fix our lack of conviction. The rapid-fire pivoting continued. Around this time, I started coming to terms with the fact that I was seriously unhappy. For as long as I’ve been interested in startups, I have never wanted to start a company simply for the sake of starting a company. Yet doing things “for the sake of starting a company” was where I found myself, as I scrambled for a plausible idea.

I don’t know if this is true for other people, but as a flailing founder I desperately wanted to believe in the startup myth—that success was just over the next ridge, that if we waited a bit longer, or had a slightly better idea, we would suddenly be riding a rocket ship.

I hated doubting myself, but I’ve never been good at blind faith.

So I decided to leave.

I Left My Heart In San Francisco

Since then, I’ve taken a break from startups. I still love Silicon Valley, and I know I’ll eventually end up back there, but in the meantime I’m exploring the world while doing remote software consulting. (So far I’ve visited Japan, China, Thailand, Greece, Costa Rica, Nicaragua, Argentina, and Brazil. Not to mention half a dozen states back home. Humblebrag, I know.)

I don’t regret my experiences in the slightest. Sure, I might have done things differently, but I’ve more or less landed on my feet. And I’ve learned something very important about myself: I don’t want to start another company until I find a problem that I care about. A problem that I eat, sleep and breathe. A problem worth solving.

P.S. Thanks for reading! I decided to find out what other founders did differently, so I’m putting together a book! It’s a collection of interviews with startup founders on the topic of “Do Things That Don’t Scale”, a fantastic essay by PG.

Huge thanks to Alexey Komissarouk, Casey Rosengren, and Ruth Marks for reading drafts of this.