Why Do Six Sigma Teams Fail?

I worry when I hear about companies shutting down their Six Sigma effort. What's the problem? Six Sigma should be helping, not hurting. 

When I first got into quality improvement, we had the best training, started lots of teams, but most of the teams failed to deliver. So, I started applying Lean Six Sigma to itself. I may be the only trainer/consultant to do so.

Root Cause Analysis
I took a step back and started asking Why? Why? Why? Why? Why? Here's what I found:

  • Teams are formed before the data is analyzed. Why is this a problem? Unfortunately, when you bring a group of people together, but don't have a clear focus driven by data, you end up with what I call "the 100 yard dash for the directionally impaired."

    Solution: Analyze the data first to narrow your focus using line graphs and pareto diagrams;then pick a team that has expertise in that area to do the root cause analysis.

  • Teams choose their own problem to work on. Why is this a problem? Unfortunately, most of the time, teams want to fix their suppliers or their customers. Union employees want to fix management; management employees want to fix workers. 

    Or they choose something trivial to "get experience." Remember the dark side of the 80/20 rule: 80% of the effort only produces 20% of the benefit. This is why Six Sigma fails.

    Solution: Let the data lead you to a problem that you can work and that you own. You can't fix someone else's problem. You can give them the data and the analysis, but you cannot solve it for them. They won't implement your solution.

  • DMAIC. Why is the Six Sigma improvement process a problem? Unfortunately, DMAIC begins with Define and Measure, so most teams get lost in defining the process and implementing new measures.

    Solution: Skip Define and Measure; go straight to Analyze, Improve and Control. You already have enough data. Somebody somewhere is keeping a count of the number and type of mistakes, errors, defects, repair, rework or waste. Find some data you can analyze to narrow your focus. Then improve and control. 

  • Chartered teams meet one hour a week foreverWhy is this a problem? Because there's 167 hours of delay between meetings. It violates the lean principles of eliminating delay and one-piece flow.

    Solution: SWAT Teams. When the data has been analyzed and the problem solving effort laser-focused, a team of subject matter experts (SMEs) only needs to meet for two-to-four hours to identify the root causes, countermeasures and implementation plan.

  • Scope creep. Teams invariably want to solve world hunger, boil the ocean and fix everything all at once. When teams scatter their focus, they solve nothing.

    Solution: Use pareto diagrams to narrow the focus. Then analyze one "big bar" of the pareto chart at a time.

  • Whale Bone Diagramming. If a team starts covering the conference room walls with fishbone after fishbone diagram, the focus is more like a flashlight than a laser.

    Solution: Go back to the data and narrow the focus.

  • "Just in case" training. Many teams and team leaders (e.g., green or black belts) get lots of training long before they apply it. Why is this a problem? Because humans lose 90% of what they learn in 72 hours if they don't apply it immediately.

    Solution: Just in time training. Give the SWAT team an hour of training and then throw them right into root cause analysis. They'll learn more working on a real problem than they'll ever learn in training.

Here's My Point
If Lean Six Sigma isn't producing immediate, measurable, ongoing benefits, management will kill it. Teams can't afford to waste time. Mistakes in team formation, team meetings and data analysis can doom a team's chances of success. And team failures can kill Six Sigma.

Rights to reprint this article in company periodicals is freely given with the inclusion of the following tag line: "© 2008 Jay Arthur, the KnowWare® Man, (888) 468-1537, support@qimacros.com."

Free Lean Six Sigma Yellow Belt TrainingTake our FREE Lean Six Sigma Yellow Belt training online.