Experiment Your Way to Success

How the mother of all learning methods can help your organization improve.

The heart of most effective continuous improvement is experimentation. Experimentation is the mother of all learning methods. It drives learning throughout an organization based on what is real, not based on theory or opinion. Whether you use PDCA, DMAIC, 8D, A3s or any other method in the alphabet soup of continuous improvement, there is a backbone of experimentation whose spirit you can follow, or fail to.

What makes experimentation successful?

1. Demystify the hypothesis. Hypothesis can sound like a scary word, but it is at the heart of what makes improvement an experiment. But it doesn't have to be scary. All you need is the word "expect." If I make this change, I expect this result. If I move this rack, I expect to save 3 seconds in my cycle time. If I send out pre-reading, I expect to shorten my staff meetings by 30 minutes. Anyone can do that. All you need is to pause long enough to think it through.

2. Follow the spirit of experimentation. The hypothesis is what turns "trying stuff" into true experimentation. The reason for the hypothesis is that you're not just testing your idea; you're testing your understanding of current reality. Experimentation starts with observing and understanding current reality. You then develop your hypothesis. You then test your hypothesis. This gives you feedback on how well you understood current reality.

Here's your alternative. You can assume you know current reality. You then jump to a solution. Then you implement the solution. Regardless of what methodology you use, the methodology doesn't force you to think, and you end up on one path or the other.

3. Make it cheap and fast. One of the barriers for people doing experimentation is the expense and the time. We can help people by creating approaches and resources to make experimentation cheap and fast.

A lean organization learns its way forward.

When I was at Chrysler, I built a simulation room. It was quite large but small compared to the plant. We had installed a two-workstation moving conveyor. We had also installed two-workstation conveyors for all the subassemblies, such as doors and instrument panels. These lines moved at line speed but could also move slower if we wanted them to. The purpose was to be able to try out procedures without interrupting the production line. If people had an idea, whether it changed components or just process, they could go into the room and try out several different iterations very cheaply and quickly. This simulation room, while initially a decent-sized investment, paid off quickly and often.

"Moonshine shops" have supported Boeing's lean journey. These small teams can fabricate almost anything to help teams experiment and improve. They've been able to develop everything from small fixtures to functional equipment. They developed such good capabilities that equipment manufacturers were visiting them to learn how. People didn't need approvals; they just showed up at the moonshine shop and had made what they needed. It sped up both experimentation and implementation.

4. Do one experiment at a time. I've had countless conversations in recent years about poorly executed enterprise software implementations. What makes this so difficult? One reason is that it's bundled improvement. Too many things change at once to tell what really works. While there are controlled ways around this such as Design of Experiments, for the most part, it is more effective to make one change at a time. This is so we can distinguish what worked and what did not and more importantly, why. When we isolate the change, cause and effect is more clear.

Many kaizen events violate this. Kaizen events aren't bad, but changing many things at once isn't as good for learning as one step at a time.

I propose that experimentation is the core of learning for a lean organization. It doesn't matter if it's process improvement or technology adoption, experimentation is the source of learning for cause and effect. A lean organization develops scientists, not in the lab, but of their organization. A lean organization learns its way forward. 9

Contributing Editor Jamie Flinchbaugh is a co-founder and partner of the Lean Learning Center in Novi, Mich., and the co-author of "The Hitchhiker's Guide to Lean: Lessons from the Road."

See Also:

Hide comments


  • Allowed HTML tags: <em> <strong> <blockquote> <br> <p>

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.