Are you ever tempted to skip the "M" in DMAIC?

In my experience working with the DMAIC model, the most common stall point in the process is the Measure phase. This is the step that requires us to assess baseline levels of performance. Here, we attempt to quantify a performance problem by gathering actual process measurements, customer metrics, and/or financial indicators.

A common problem that Six Sigma professionals encounter is the lack of measurable data. Business needs are often expressed to us based solely on anecdotal evidence or manager intuition. During project definition, I’ll often hear statements like:

“We need to increase revenue”
“Profits are too low”
“Customers aren’t happy”

A good follow-up to these is to simply ask: “how do you know this is a problem; what evidence are you basing this on”? If you’re then presented with concrete data, control charts, financial trends, etc., then you know you have a project where you can make a measurable difference. If, on the other hand, the claims are not quantifiable, you need to stop!

It can be tempting to overlook the lack of data and move forward to root causes and solutions. But this creates a situation where results can’t be validated, or even worse, we spend time and money working on the wrong thing. Before moving on in the cycle, we'll need to either identify or create the key indicators.

Here are some questions to ask yourself before moving past the measure phase:

• Have relevant metrics (KPIs) been selected or created?
• Do these metrics represent true customer interests?
• Has historical data been gathered to illustrate trends?
• Is the data statistically significant / valid?
• Have the data measurement systems been validated using MSA?
• If improved, will these metrics indicate clear project success?

If we plan to fully invest our time and energy into a process improvement project, it only makes sense that we first figure out the best way to keep score.