The Phase Everyone Rushes: Why Control Is Where Improvements Die
By the time a DMAIC project reaches Control, the exciting part is over. The team has Defined the problem, Measured the baseline, Analyzed the root causes, and Improved the process — and the chart finally moves in the right direction. The natural instinct is to declare victory and move on. That instinct is exactly why so many improvements quietly evaporate. Control is the phase that decides whether a gain is permanent or just a good month, and it is the one teams shortchange most.
The pattern is familiar to anyone who has audited a process a year after a project closed: the new method is gone, the old workaround is back, and nobody can say when it slipped. Through the disruptions of the past two years, this has been especially common — a process tuned for stable demand quietly reverts the moment volumes swing or staff turn over. Control exists to stop precisely that drift.
Where Control Goes Wrong
Treating it as paperwork. A control plan filed and forgotten controls nothing. The plan only works if someone owns each response and actually acts on it when a signal fires.
Charting everything, watching nothing. Teams stand up control charts on a dozen metrics, then stop looking. A few well-chosen charts that people genuinely review beat a dashboard no one reads.
No reaction plan. A chart that shows a process going out of control is useless if no one has decided in advance what to do about it. Each key signal needs a named owner and a defined response before you hand the process back.
Skipping the handoff to the process owner. The improvement team eventually leaves. If the day-to-day owner was never trained, never bought in, and never given the updated standard work, the gain leaves with the team.
Forgetting to confirm the savings are real. Control is also where you verify that the improvement held long enough to count — and where Finance signs off on benefits that are actually there, not projected.
Making the Gains Stick
Sustaining a result is less about heroics and more about a few durable habits built into how the work runs every day. The goal is to make the improved way the path of least resistance, so the process holds even when attention moves elsewhere.
Update standard work and any procedures so the new method is the documented way, not a memory.
Where it helps, mistake-proof the step (poka-yoke) so the wrong action is hard or impossible to take.
Keep a small set of control charts on the vital few measures, with clear out-of-control rules.
Write a reaction plan: for each signal, who responds and what they do.
Hand the process to a named owner with training, the standard work, and a scheduled check-in.
Confirm and book the benefit with Finance, then schedule an audit a few months out.
Done well, Control is quiet by design. There are no dramatic wins, just a process that keeps performing after the project team has moved on and the original problem stays solved. That durability is the whole point of Lean Six Sigma — anyone can improve a number for a quarter, but holding it is what separates a real improvement from a temporary one. In an environment where conditions keep shifting, a process that can absorb change without backsliding is worth far more than a one-time spike on a chart.
If your improvement projects post strong results that fade once the team disbands, XNM's strategic advisory can help you build the controls that make those gains last.