It's been almost two years since venture capitalists moved the cheese on startups, demanding more robust testing of a business hypothesis before writing a large check. Have incubators evolved to meet the new standard of proof?
Some context: In the briskly moving startup ecosystem of the late 2000's, Y Combinator set the gold standard for a new breed of incubator (a class which includes TechStars, DreamIt, et al) based on the express goal of "getting you to the point where you've built something impressive enough to raise money on a larger scale." Presuming that a startup could successfully meet this first major milestone, it could move on to raising seed capital - which funding needed to cover two major milestones:
1. Demonstrate that your product works
2. Show that people want to use it
With a compelling product and early signs of working customer acquisition, an early 2008 startup had a good shot at raising a $5 million Series A round. But then the market crashed and investors became more risk-averse, and late 2008 startups had to instead achieve three major milestones with their seed capital:
1. Demonstrate that your product works
2. Show that people want to use it
3. Prove that you can make lots of money with it
The rapid shift of the third milestone from a Series A goal to a seed round expectation took many startups by surprise. Many founders had to scramble in 2008-2009 to extend the runway of their pre-cheese seed funding in order to build up a baked-in, working business model. This requirement is easing up a bit in 2010 as VCs loosen their purse strings, but it's probably here to stay.
So what should a next-generation incubator do? Like the seed funder that comes after it, it has to go further downstream. A good next-gen incubator will help its companies hit both the first and the second key milestone. In addition to a zealous focus on product, incubator companies will need to develop (and test) customer acquisition models. They'll not only nail down important metrics, they'll also have a strategy for driving them.
Early indications are that YC will continue to evolve its program in this direction - this year's crop of startups is impressively mature - and it's a given that Dave McClure at 500 Startups will be running customer acquisition drills nightly. I strongly suspect that the soon-to-launch AngelPad will also be hitting on both cylinders given a Google experience steeped in metrics.
That leaves some open questions: How will the incubator-team relationship have to change, if the incubator is helping the company further along? Does the ecosystem then need resources that only address milestone #1, or is that simply not feasible any more? In any event, changing the coordinates of one point in the ecosystem means changing navigation throughout it. What else do we need to do?