Abstract
If you change a lot at once and it goes wrong, you can't tell which part caused it. PDCA keeps each change small and treats it as an experiment. You plan a change and predict what it'll do, try it on a small scale, check the result against the prediction, then decide whether to adopt it, adapt it or drop it. Then you start the next round.
Walter Shewhart described the cycle and W. Edwards Deming spread it, especially in Japan, where it became part of kaizen. The check step is what makes it work. Because each change has a target, you learn something even when it fails.
Method
- 01
Plan
Describe the problem, what you think will improve it, and the number that would show it worked.
- 02
Do
Try the change on a small scale for a set period, and write down what you actually did.
- 03
Check
Measure the result and compare it with your target. Note anything that surprised you.
- 04
Act
Make the change the new standard, adjust it for another round, or drop it and try something else.
- 05
Repeat
Start the next cycle from wherever you ended up. Small gains add up over several rounds.
When to use it
- You want to improve a process gradually instead of redesigning it in one go.
- You're not sure a change will help and want evidence before rolling it out.
- A team needs a regular habit for small experiments.
Common pitfalls
- No measurable target, so the check step turns into a matter of opinion.
- Running the trial too big. Keep it small so failing is cheap.
- Skipping the act step, so changes that worked never become the standard.
- Changing several things at once and not knowing which one did it.
Origin and references
Attributed to Walter A. Shewhart, popularised by W. Edwards Deming (1939).
- Shewhart, W. A. (1939). Statistical Method from the Viewpoint of Quality Control. Graduate School of the Department of Agriculture.
- Deming, W. E. (1986). Out of the Crisis. MIT Press.
- Imai, M. (1986). Kaizen: The Key to Japan's Competitive Success. McGraw-Hill.