Waste Not

One of the last projects in my dad's long career was in biomedical manufacturing.  It took five years and never went to market.  In that time, his team learned a number of things, particularly about the production of ceramic seals having certain tolerances.  The difficulty in doing this with sufficient quality ultimately stymied the project.  I remember asking him what would happen to all of the knowledge the team had acquired.  Would they share it with the manufacturing community?  How could it be ensured that others could learn from their experience?

Learning is at the heart of the Lean Startup movement, which takes its inspiration from the short cycle times of small batch Lean manufacturing processes.  Eric Ries wrote down the principles of the organizational processes that many successful startups have used.  Many startups find that their original idea goes through many iterations before becoming a successful product.  The worst thing a startup can do is spend a year 'perfecting' a product no one wants.  So, startups should get an (imperfect) product out there, see how people use it via actionable metrics, and modify it and their assumptions accordingly.

The Lean philosophy is essentially about embedding the scientific method into product development processes.  Since you can't rely on customer feedback to learn how people use or could use a product, you should build reporting into your software from the start, not as an afterthought.  Your software should also be able to run A/B tests where some users are given a new interface while a control group sticks with the old one.  The ways in which customers will use a product should be tested by its use.

In a certain way, this is all very straightforward.  No one wants to think that they are not thinking scientifically when evaluating a product, but being Lean is hard.  The scientific method has to be at the heart of everything you do.  It touches what is built as much as how it is built.  The product development and product planning teams have to work very closely together.  Developers can't just build whatever product people think up.  I know of many organizations that could not work this way.

Though Ries's advice is sound for software startups, his main target is waste at a societal level.  Think of all the billions of dollars and hours of time wasted pursuing products that were not or could not be rigorously hypothesized, tested, or proven due to imperfect processes.  The problems of product development are people problems, not technological problems.  Ries shows how to cut waste at an organizational level, with benefits that will be felt throughout society as people start to adopt the principles.

Still, Lean doesn't really help with the problem facing my dad's organization.  What techniques or technologies can be used to share information between organizations?  This may not be a general problem that can be solved.  And, after all, if companies share everything about what they've learned, what advantage do they have in the marketplace?  By writing down Lean principles, Ries is helping to share best practices about management, but there are plenty of domain-specific problems and best practices out there to be tackled.