The popular view of a technical entrepreneur is someone with a big vision and a stubborn determination to charge straight ahead through any obstacle and make it happen. The vision part is fine, but more successful entrepreneurs have found that the extreme uncertainty of a new product or service usually requires many course corrections or pivots to find a successful formula.

The traditional mode of starting a company has long been to plan a serial process, where you complete all the steps, leading to the big bang launch of the company. I think it's time for a dramatic departure from this old model, to a new one called planned iteration or Lean Startup (http://en.wikipedia.org/wiki/Lean_Startup) methodology. With this one, you assume you won't get it right the first time, so you launch with a minimum viable product (MVP).

This idea was first articulated by Paul Graham in an old essay, called Startups in 13 Sentences (http://www.paulgraham...