Yesterday I attended YC’s Startup School. It’s an annual day of awesome talks and interviews with tech luminaries.

This was my second time going (I went in 2010). Overall it was very entertaining but I don’t think I learned anything new. The speakers all had great things to say but it was very much silicon valley consensus. I still enjoyed myself, though, and came away energized.

Someone wrote a good summary here. A reoccurring theme was attempting to distill the formula for successful startups. Some of the key points were:

  1. Build something you love or something for yourself
  2. Stay laser-focused on product
  3. Get paying customers as fast as possible (it validates the idea)
  4. Choose a big enough market
  5. Have a good answer for ‘why hasn’t this idea been done?’
  6. Know some secret that the rest of the market doesn’t

While I think these are all correct some of the time, some of them are highly dependent on the business. The second point is often true for consumer startups but matters less for enterprise startups. All day, speakers didn’t disambiguate advice for enterprise startups vs advice for consumer startups; I think that’s dangerous.

The first point could be refactored to ‘be authentic to the opportunity’ - you don’t need to love it or use it but you need to have an incredibly deep understanding of the problem, customer, and ecosystem. Loving / using the product is a great way of getting that authenticity but it’s not necessary.

The corollary to ‘be authentic’ is ‘choose a problem you can stay excited about for five years’ - this is more entrepreneur dependent: some founders need to love a space to stay motivated, others just need to see a business opportunity.

Points five and six go together well and align with the Consensus / Non-consensus 2x2. They’re also probably true across consumer and enterprise. The answer to ‘why hasn’t this idea been done’ is almost always a change in technology, social behavior, or regulatory environment. Again, nothing here counter-intuitive for anyone that’s spent time following startups or hanging out in silicon valley.

One reoccurring theme was the survivorship bias in the speakers. There was a lot of “we did X and succeeded” where it was unclear if X was essential to the success. Lots of correlation / causation confusion because we were just looking at success cases and didn’t see counterfactuals. These types of talks were entertaining but not useful.

I appreciated Mark Zuckerberg’s honesty / humility during his interview. PG was pushing Mark to reveal a formula, an early strategy, or a thought process that destined Facebook for global success. Mark kept dismissing PG and emphasizing that they had no grand plan, didn’t think of it as a startup/company for a long time, and didn’t think it would be the global identity/network platform until well after they expanded beyond colleges. The laser-focus on product helped them succeed but the degree of success was a function of luck and timing. I thought this was incredibly refreshing; PG seemed disappointed.