Picking a Lean Six Sigma Project That Actually Pays Off
A Lean Six Sigma program lives or dies on the projects you choose to run through it. Pick well and the method does its job: a stubborn problem gets measured, understood, and fixed in a way that holds. Pick badly and even a flawless DMAIC effort produces a tidy report that changes nothing. After more than a year of disrupted supply chains and teams scattered across kitchens and spare bedrooms, the temptation in 2021 is to launch improvement work everywhere at once. That is exactly when disciplined selection matters most.
The good news is that the failures are predictable. Teams tend to make the same handful of selection mistakes, and each one is avoidable once you can name it.
The mistakes that sink a project before it starts
Choosing a problem with no measurable defect. Six Sigma reduces variation in something you can count. "Improve morale" or "make the office more collaborative" are worthy goals, but if you cannot define a defect and measure its rate, you have no baseline and no way to prove improvement. Before committing, ask: what exactly is the unit, and what makes one defective?
Picking a solution disguised as a problem. When the charter says "implement the new scheduling tool," the answer has already been decided. DMAIC is wasted on confirming a foregone conclusion. State the problem as a gap — late deliveries, rework hours, claim errors — and let the analysis point to the cause.
Boiling the ocean. A scope that spans three departments and six months will exhaust the team and dilute accountability. Strong projects are narrow enough to finish in three to four months and broad enough to matter. If you cannot draw a box around the process, the project is too big.
Ignoring data availability. A project that needs six months of clean historical data you do not have is not ready. During the pandemic many baselines were thrown off by demand spikes and shutdowns; confirm the data exists and is trustworthy before you charter the work.
No engaged process owner. If the manager who owns the process is not asking for the fix, the improvement will not survive handoff. Sponsorship is not a signature on a charter — it is a person who will defend the change after the team moves on.
A simple test before you commit
Run every candidate through a short screen. The projects that pass tend to deliver; the ones that fail are better redirected before resources are spent.
Is the pain real and felt by someone who can act? A problem nobody owns is a problem nobody fixes.
Can you state the defect and measure it today, with data you trust?
Is the cause genuinely unknown? If you already know the fix, just implement it — DMAIC is for problems where the root cause is unclear.
Will it finish in roughly one quarter, by one focused team?
Does the benefit clear the cost of the effort, in dollars, hours, risk, or customer impact?
Match the project to the moment
In a recovery year, the highest-value projects often sit where disruption left scar tissue: a procurement process patched together during shortages, an approval workflow that broke when everyone went remote, a quality check that quietly slipped. These problems have clear defects, real cost, and owners who feel the pain daily — which is exactly the profile of a project worth chartering. Resist the urge to chase the most visible fire and instead pick the one where measurement will most reliably lead to a durable fix. A short, well-scoped win also builds the credibility a young improvement program needs to earn its next, bigger project.
If you are deciding where to focus your improvement effort and want a clear-eyed view of which projects will actually pay off, XNM's strategic advisory can help you scope and prioritize the work.