Measure Phase of Six Sigma / DMAIC

Six Sigma is based on the DMAIC cycle. DMAIC stands for Define, Measure, Analyze, Improve, Control, and is a step-by-step approach to solving large problems.

Six Sigma refers to the standard deviations and is geared primarily towards reducing variation, so the output of a process fits within six standard deviations of the mean. With such a numbers-oriented approach to improvement, it is no surprise that one of the major phases is Measure.

Now, most companies regularly measure, and measuring will continue throughout the process. So why is there a specific Measure phase?

There are two main reasons. The first is to gather data to confirm the Define phase, specifically the charter. When a project is set up, information is incomplete. The Measure phase must fill in the blanks about the project goals and assumptions.

The second is to get ready to start analyzing the problem. Data collection can take a long time. If you wait until you are analyzing the process to start looking at numbers, your team is going to be working in fits and starts. They won’t be able to create any momentum.

The Measure phase combines collecting existing data with gathering new data. And more importantly, it involves determining what data you will need to proceed with the project.

The purpose of having a separate Measure phase is very simple. It is intended to streamline the time you have your team gathered. Imagine the first time the team meets if there was not already a pile of relevant data to look at. They might go look at a process for a few minutes and start asking a few basic questions. The answer to nearly every question would be “we don’t know” or “we’ll have to get that pulled off the database.” Both answers mean the team’s progress comes to a screeching halt.

If the data is time-based, such as in the number of defects per day, you might even have to adjourn for a few weeks until the right data is available.

So that creates a dilemma for team leaders. The purpose of a team is to get them to come up with solutions as a group. You don’t want to lead them down a pre-ordained path to a specific solution. There will be generic data you know you will need no matter what you decide. But as project leaders make their data collection plans, they will probably be well served to think a little bit about potential solution. The data they settle on should support those options.

There are two issues with this. The first is that the data that is available frequently biases the team. They see that as an easier path which leads to ideas using that data getting chosen over those that don’t. They make solutions fit the available data rather than the other way around as it should be. It can make alternative ideas dry up.

The second issue is that collecting data takes time and effort. That means you have to get enough to keep the project moving without getting so much that you waste a lot of money.

Measure is not just about collecting numbers, though. It also would involve scrutiny over your processes related to the project. Gather up all work instructions, quality standards, KPI information, etc. Whatever you can find about the process, make it accessible to the team.

You can also consider making a video of the process. There’s a potential minefield here with people…

Extended Content for this Section is available at academy.Velaction.com

Don’t take having a specific Measure phase to mean that there will not be measuring in other phases. Whenever you come up with a shortage of data, you measure. The Measure phase is intended simply to streamline the other phases.