Steve Blank was one of the first people to form the Lean Startup approach in his book, "The Four Steps to the Epiphany." He recently had a discussion that was recorded with . The details can be found in an article on Inspirationfeed.
What I wanted to examine was the Lean Startup principle of launching the Minimum Viable Product (MVP).
Rather than putting products through year-long testing phases, Blank suggests that startups develop a minimum viable product (MVP) and get it to market as soon as possible. This approach will accelerate the testing phases because startups can continually make changes to the MVP as they produce it. This also allows startups to make the necessary changes to the MVP before it becomes too costly.
-- David Schmidt , Inspirationfeed
Just the other day I saw a developer comment complaining about the "go fast and break things" culture in many engineering departments these days. Technology choices have a tendency to stick around longer than anyone expects, so the short-term decisions end up becoming the foundation of a project. If the choice isn't the optimal one for a long term foundation, it's nearly impossible to go back and rebuild the core of your work. That's why some engineers fiercely in favor of "over-engineering" and the year-long testing phases that Steve Blank criticizes.
Of course, the assumption and hope is that your startup will be lean and agile enough to quickly rebuild things from scratch if needed, even if that thing is now bringing in revenue with customers relying on it.
What are your thoughts? How do you balance the MVP approach with over-engineering?