mikepk.com about current projects contact

Inherit the wind?

November 22, 2005

Fred Wilson has an interesting post framing the formation of startups in the light of the intelligent design versus evolution debate. I think this is a very interesting approach, but I think it is limited in a lot of ways. I don't think company formation is so tidy.

The basic argument goes that there are two ways to start a company. One is more organic, taking some side project or hobby and slowly evolving into a company. The second is to decide you want to create a company and product and 'intelligently design' it from the beginning.

I think strictly adhering to the latter approach is a flawed strategy. If the intent is to design the company and product in a vacuum and then hoist both into the world fully formed, the likelihood of success is extremely small.

Having gone through the exercise of creating a business plan, it's almost amusing how many input variables you have to fudge. It's easy to take a financial model, tweak a few of the input variables, and suddenly the model predicts amazing rocket-powered growth. Take those same variables and nudge them slightly in the other direction and the graphs all become depressing. That's not to say this kind of analysis isn't interesting, revealing what your success factors are likely to be and where your focus should be. This focus should be where your model is most sensitive, the variables that with the smallest movement cause the biggest swings in your success metrics.

In an earlier post I discussed my experience with complex systems. I think people are not generally capable of understanding all the complexity that's implicit in the creation of a startup. All companies to some extent must follow an evolutionary strategy. Building a complex model is great, and can help you frame your strategy, but it's just that, a strategy. Colin Powell said it best, "No battle plan survives contact with the enemy."

I think 'intelligently designed' companies are the ones where the company formation 'hypothesis' has been worked out to great detail, but the successful ones must have a plan to evolve the strategy or even toss it out if grossly wrong. If the hypothesis is wrong, there must be a strategy in place to 'tweak' things for success. The likelihood that the 'intelligent designer', being human, has accounted for all the variables of startup success is very slim.

I think in all cases startup creation is a mix of the two approaches. The hobby or side project that evolves into a company will have some 'intelligent designer' who recognizes the system feedback, the venture potential and formulates a strategy to go to "the next level". That could entail creating revenue streams, addressing a bigger market, spinning out the project, etc... Conversely, the intelligent designer must be willing to evolve his design with feedback from the market. Imagining oneself as being able to fully understand the impact to, and reaction of: customers, activities of competitors, general market vagaries, and the myriad of other variables affecting a startup severely limits potential success.

Taking the discussion even further, I think mentioning the differences between the ID versus evolution debate and startups is useful. I've argued that the creation of a startup is neither solely intelligent design nor evolution, and I think the majority of Americans think the theory of "intelligent design" doesn't preclude the idea of evolution. I think this is due to the clever choice of the name "intelligent design" for the theory.

I have no particular beef with the overall notion of a 'first creator'. In fact the majority of mainstream Americans believe in some idea of a supreme being, this is why the name "intelligent design" doesn't cause a negative public reaction. The ID theory, however, is really 'creationism' set up in direct and exclusionary opposition to evolution as an "equally valid theory". Trying to create a scientific theory, solely on religious or spiritual precepts is anathema to me (even if veiled in pseudo-scientific language).

I think the majority of Americans (and indeed even the Catholic church, no paragon of liberal ideology) would have no problem saying that perhaps evolution is the *mechanism* by which human beings were 'designed'. I think that the majority of Americans would say something similar to (although not exactly) what I argue about ID versus evolution in startup formation, perhaps it's a mix of both.