PDCA Cycle
A four-step loop for continuous improvement: plan, do, check, act.
What it is
PDCA is a simple, repeatable method for testing a change and learning from the result. Instead of rolling out a solution everywhere, a team tries it on a small scale, measures what happened, and then decides whether to adopt, adapt or drop it.
Each turn of the cycle builds on the last.
When to use it
For any process improvement, problem-solving effort or pilot of a new idea.
The four steps
- Plan Define the problem, set a target and decide what change to test.Ask: What do we expect to happen, and why?
- Do Carry out the change on a small scale.Ask: Can we try it on a small scale?
- Check Compare the results with what was expected and learn from the difference.Ask: What did the results show?
- Act Standardize the change if it worked, or adjust and run the cycle again.Ask: Adopt, adjust or abandon?
Example
A clinic wants to cut missed appointments. It plans a text reminder two days ahead, tries it with one doctor's patients for a month, finds no-shows fall by a third, and extends it to the whole clinic.
How to run it
- Choose one problem and state a prediction.
- Run a small test.
- Compare results with the prediction.
- Standardise what worked, then start the next cycle.
Limits
It is slow for urgent problems, and teams often skip the check step and treat a plan as proven. Its simplicity hides the discipline needed to measure honestly.