Why Good Quality Improvement Needs Change Management

Every quality director has seen it. A Green Belt project with a clean charter, solid baseline data, and a fix that worked in the pilot. Leadership signs off. The control plan gets filed. Six months later, a walk through the unit shows the old process back in place, a few workarounds nobody wrote down, and a control chart that quietly stopped being updated in month three.

The usual explanation is that the team lost focus. The more accurate one is that the project was designed for a process and delivered to people, and nobody planned for the second part.

The fix was fine. The adoption was not.

Lean Six Sigma and PDSA are good at what they were built for. They give a team a disciplined way to define a problem, measure it, find the cause, test a fix, and hold the gain. Every step has tools, and most belt programs teach those tools well.

What they teach less well is the people side. Who sponsors the change and what that sponsor must visibly do. What the staff who run the process need to hear, and from whom. What skills they need before go-live, not after. Where the pushback will come from and what it is really about. And what keeps the new way in place after the project team moves on.

Those five things have a name. Change management is the discipline of sponsorship, communication, training, resistance management, and reinforcement. It is not a separate project. It is a layer that sits on top of the improvement work you are already doing.

Where it fits in DMAIC

Each phase of DMAIC has a natural place for the people side.

Define. Write the case for change in plain language staff can repeat. Name the sponsor and what they will do. Map who is affected, who has influence, and who could block. Decide what “adopted” will look like in countable terms, so Control has something to measure against.

Measure. Baseline the people, not just the process. How consistently is the current process actually followed? How ready, willing, and able is each group? What do people really do on a shift, versus what the process map says?

Analyze. Find out why current behavior persists. A force field analysis shows which restraining forces are movable. A root cause analysis of adherence separates awareness problems from skill problems from workload problems, because each calls for a different fix.

Improve. Build the change plan alongside the process fix, not after it. Communication plan, sponsor actions with dates, training with practice time, a resistance plan, and a pilot that tests whether people will use the change, not only whether it works.

Control. Track adoption alongside the outcome measure. Check proficiency, not attendance. Give managers a reinforcement routine that fits in a huddle. Name a person, not a committee, to own the new way after the project closes. Close the loop with the staff who gave input.

The full map, with 29 tools sorted by phase, is at /dmaic.

Where it fits in PDSA

PDSA teams run smaller, faster cycles, and the people side moves with them.

Plan. Predict adoption the same way you predict the outcome. Who does this test touch, what will they need, and what does “used it” look like?

Do. Run the test with the people who do the work. Watch what actually happens. Treat surprises as data.

Study. Compare adoption with your prediction. Who used the change, who did not, and why? That answer shapes the next cycle as much as the outcome data does.

Act. Adopt, adapt, or abandon. If you adopt, reinforcement and a named owner are what keep the gain. If you spread, sequence the units that are ready first.

The PDSA version of the map is at /pdsa.

Three things to do on your next project

First, add one line to your charter or aim statement: what adoption will look like, in numbers. If you cannot write it, you cannot measure it in Control.

Second, ask your sponsor for one visible action per phase. A message from them, a walk through the unit, a question at a huddle. Sponsors who only sign the charter are not sponsoring.

Third, in your pilot or test cycle, count use as carefully as you count results. A fix that works only with the volunteers who designed it is not proven yet.

None of this replaces your belt curriculum or your PDSA worksheet. It sits beside them. That is the whole point: the improvement method you already trust, with the people side planned instead of hoped for.

If you run a belt program or PDSA teams in a hospital, clinic, or behavioral health program, the Quality Improvement modules package all of this by phase. See how they work at /quality-improvement.