The First Version of Your Start-Up Should Embarrass You

Why you want to build the simplest possible solution to a problem, and launch it.

Once you've identified a real problem and done your research, start trying to solve it in the simplest way possible. Your first version should certainly embarrass you. "Minimum viable product" has become a startup cliché for good reason. Just build the simplest possible solution to a problem, and launch it.

This probably won't take as long as you might think. Each round of Y Combinator was designed to be three months long because Paul [as in Paul Graham, founder of Viaweb and co-founder of Y Combinator] wanted it to be a summer program, so students could decide to take time off from school if their company was going well. This happened to also be a reasonable amount of time to go from idea to a live product. If it takes longer than a few months to start testing your idea, it's because you're either trying too hard to perfect it (you never will, so don't bother) or there's some other bigger problem.

The first version of reddit was absurdly simple. We didn't have voting, and we certainly didn't have commenting or the ability to create subreddits. It was simply a place where one could submit links and, based on clicks, see them rise and fall on the front page. A new user would simply see a front page of interesting links to click on.

Hipmunk was a flight-only search when we launched (no hotels, car rentals, etc.), and even that was strikingly bare-bones. Thanks to Adam's hustle with the online travel agencies, we not only provided flight data but also started to collect a commission on referrals right away. We were making money from day one, which always puts a smile on investors' faces.

And bear this in mind: the first version of Airbnb, the startup that has more rooms available for rent than the Hilton corporation, started from a single apartment in the SOMA neighborhood of San Francisco. The founders began the site by renting out air mattresses in their own home to conference attendees looking to save money.

Once you're up and running, spread the word and start watching how users interact with what you've built. Listen to how they're talking about it. This is key. There's something incredibly satisfying about seeing the logs of the first users who try out what you've built. It's one reason why I can't encourage enough students to start building projects just for the experience of having real people all over the world use something you've built. Compared to dull schoolwork, learning by creating something relevant and usable is incredibly rewarding.

Once you've got something to show, use the growing number of available tools that allow people to share that great idea. We launched reddit in 2005, before "social media" was a phrase. Just five years later, launching hipmunk was drastically easier, because by then there were more tools than ever for people to spread the word about things they care about.

Word of mouth has always been the most powerful form of advertising-;and it spreads faster and farther than ever before. Make something people want, and people will find out about it. If you're not getting traction, it simply means you haven't solved that core problem of making something people want. But that's okay! Figure out what people are using. Talk to your users--those first hundred or so people who are willing to take a chance on a product they've never heard of are golden. Treat them well and get to the root of whatever problem it is that you're not currently solving for them.

Alexis Ohanian