PDCA: Plan, Do, Check, Act — and Actually Learn
Plan: define the problem and the target
Plan is where most problem-solving fails, because most teams skip it. Planning means stating the problem in observable terms — not "quality is bad" but "the third station produced forty-one defects on Tuesday's second shift against a standard of three" — then setting a target and a theory: we believe this countermeasure will move that number, and here is how we will know. A plan without a target is a wish. A plan without a theory is a guess. Write both down before touching the process, or the rest of the cycle has nothing to test.
Do: run the change small
Do means run the countermeasure, on a small scale, for a short time, with the people who do the work involved in running it. Small matters. A plant-wide rollout of an untested change is not PDCA; it is a gamble with everyone's shift. A one-cell trial for one week is a test. Document what actually happened, not what was supposed to happen — the workaround, the surprise, the operator's objection. Reality is data, even when it is inconvenient, especially when it is inconvenient.
Check: compare against the target
Check is where learning lives, and it is the step most often skipped. Compare the result against the target you set in Plan, using the same measure, over the same kind of work. Did the defect count move? By how much? Did something else get worse? A team that checks honestly learns even when the countermeasure fails — it learns that theory was wrong, which is knowledge worth having. A team that skips Check learns nothing, even when the number improves, because it never learns why. Improvement without understanding is luck, and luck does not repeat.
Act: standardize or start over
Act closes the loop. If the countermeasure worked, standardize it: write it into the standard, train it, audit it, and make it the new baseline for the next cycle. If it did not work, say so plainly and start the next cycle with a better theory — that is not failure, that is the method working. Either way, the cycle ends with a decision, not a drift. PDCA pairs naturally with structured problem-solving, including The Logic Tree, and the discipline behind it runs through The Lean Culture Blueprint. Keep the cycle visible daily: https://apps.apple.com/us/app/lean-culture-advisory/id6790203662