Plan-Do-Study-Act
🧭HOW IT WORKS
The improvement cycle associated with Walter Shewhart and 🏛️ W. Edwards Deming. A small, deliberate experiment on a process, repeated.
- Plan. State the change, and write down what you predict will happen. Decide what you will measure and for how long.
- Do. Make the change on a small scale. Note what went differently from the plan.
- Study. Compare the result with the prediction. The gap between the two is where the learning is.
- Act. Adopt the change, adjust it or drop it. Then go round again.
Deming called it the Shewhart cycle and preferred Study to the Check of the better-known PDCA, on the grounds that checking suggests inspection and studying suggests learning.
Use when: the process is stable enough that a change can be told apart from its ordinary wobble. That is the link to 🧩 Common and special causes: without knowing the common-cause spread, step 3 cannot tell an effect from noise.
Not for: emergencies, or a process nobody has measured yet. Measure first.
Where I am using it: 🧪 Report every metric with a range for a month, from November. The plan, the prediction and the month's results live in that note. Here I keep only the cycle.
The step I skip is the written prediction. Without it, Study turns into deciding afterwards that I expected this all along.