What Is the PDCA Cycle?

The PDCA cycle is a four-step method for improving a process by testing a change on a small scale before committing to it. Plan a change, Do it as a trial, Check the results against what you expected, then Act on what you learned. It is also known as the Deming cycle, and the loop is built to repeat rather than run once.

The four steps are straightforward:

• Plan: define the problem, work out its root cause, and decide on a change to test along with what success should look like.
• Do: make the change on a small scale, in a controlled way, and record what happens.
• Check: compare the result against the prediction from the plan, using data rather than impressions.
• Act: if it worked, standardise it and roll it out; if it didn’t, use what you learned to adjust and run the cycle again.

The reason it is drawn as a circle is that the Act step feeds straight back into the next Plan. A finished cycle simply rolls into the next one, which is what makes PDCA a method for continuous improvement rather than a one-off fix. The approach traces back to Walter Shewhart and was popularised by W. Edwards Deming, whose name it still carries.

Its value is that it slows the rush from problem to solution. Testing small and checking against evidence keeps risk low and stops teams scaling a fix that was never going to work, which is why PDCA sits at the heart of lean and kaizen.

Where PDCA breaks down is predictable. Teams plan endlessly and never Do, or they Do and never Check, so they never learn whether the change helped. The most common failure is skipping Act: a real improvement is found, then never written into the standard, so the gain leaks away and the same problem returns. For each stage in more depth, the PDCA guide works through a fuller example.

EviView gives teams a place to run PDCA properly: document the plan, assign the trial to an owner, review the result against the data, and roll the winning change into the standard so it holds. Book a demo.