DMAIC Explained: The Five Phases of Six Sigma (With Examples)

Define: what problem are you actually solving?

The Define phase produces a project charter: a clear problem statement, the metric (Y) that defines success, and the process boundaries. Skipping this step is the single most common reason DMAIC projects fail, teams start analyzing before agreeing what 'success' even means.

Example: 'Reduce invoice processing errors from 8% to under 2% within one quarter' is a Define-phase problem statement; 'improve invoicing' is not.

Measure: what does the data actually say?

Before analyzing causes, you need a reliable baseline. Measure establishes how the process currently performs, validates that your measurement system itself is trustworthy (Measurement System Analysis / Gage R&R), and calculates process capability.

Analyze: why is it happening?

This phase separates Six Sigma from guesswork. Tools like fishbone diagrams, 5 Whys, hypothesis testing and regression are used to identify and statistically validate root causes, not just the most obvious or politically convenient one.

Improve: test, then implement

Solutions are piloted against the validated root causes and measured before full rollout. This is where Design of Experiments (at Black Belt level) or simpler before/after comparisons (at Green Belt level) prove a fix actually works.

Control: make it stick

The most-skipped phase, and the reason many improvements fade within months. Control puts monitoring (SPC charts), documentation (control plans) and ownership in place so the gain survives after the project team moves on.