PDCA Cycle
What is the PDCA Cycle?
The PDCA cycle (Plan-Do-Check-Act) is an iterative four-step method for continuous improvement in which teams plan a change, test it, check the results against expectations and act on what they learn.
Updated November 2026 · 3 min read
The PDCA cycle, short for Plan-Do-Check-Act, is a four-step method for improving processes, products and services through repeated small experiments. A team plans a change, tries it, checks the results against what it expected and then acts on what it learned, either adopting the change, adjusting it or abandoning it. Then the cycle starts again. PDCA is one of the foundations of quality management and continuous improvement, and its logic, essentially the scientific method applied to everyday work, appears in many later approaches.
The cycle is associated with two American statisticians. Walter Shewhart, of Bell Telephone Laboratories, described a cycle of specification, production and inspection in his 1939 book Statistical Method from the Viewpoint of Quality Control. W. Edwards Deming, who called it the Shewhart cycle, taught a version of it to Japanese executives in 1950, and Japanese managers adapted it into Plan-Do-Check-Act. Deming later preferred Plan-Do-Study-Act (PDSA), arguing that "check" suggested mere inspection, when the point was to study results and learn. The two names are used largely interchangeably.
The Four Steps
| Step | What happens | Key questions |
|---|---|---|
| Plan | Define the problem, analyze its causes, decide on a change and predict what it will achieve | What are we trying to improve? What do we think will happen, and how will we measure it? |
| Do | Carry out the change, ideally on a small scale, and collect data | Did we do what we planned? What did we observe? |
| Check (or Study) | Compare the results with the prediction and analyze the difference | Did it work? Why or why not? What did we learn? |
| Act | Standardize the change if it worked, adjust it, or abandon it, and decide what to try next | Do we adopt, adapt or abandon? What is the next cycle? |
The prediction made in the Plan step is easy to skip and important to keep. Without it, there is nothing to check results against, and teams tend to declare every change a success.
Where PDCA Is Used
- Quality management systems. ISO 9001 and other ISO management system standards, including ISO 14001 for environmental management and ISO 45001 for occupational health and safety, are explicitly structured around PDCA.
- Kaizen and Lean. Toyota and other Lean organizations use PDCA as the basic rhythm of improvement, often documented on a single A3-size sheet.
- Healthcare. PDSA cycles are a standard method in clinical quality improvement, where small tests of change are run before rolling changes out across a hospital.
- Agile and DevOps. Iterations, retrospectives and experiments with feature flags follow the same plan, try, measure and adjust loop.
PDCA and DMAIC
DMAIC (Define, Measure, Analyze, Improve, Control), the Six Sigma method, can be seen as a more structured, data-heavy version of the same cycle, suited to complex problems with significant variation. PDCA is lighter and faster, suited to frequent small improvements that frontline teams can run themselves.
An Example
Consider a hypothetical outpatient clinic where 22% of patients miss their appointments.
- Plan: staff believe many patients simply forget. They decide to test a text reminder two days before each appointment, and predict that missed appointments will fall to about 15% for patients who get the reminder.
- Do: for one month, they send reminders to patients of two of the clinic's six doctors.
- Check: missed appointments for those patients fall to 16%, while the other doctors' patients stay at 22%. Follow-up calls reveal that many remaining no-shows are patients who wanted to reschedule but could not get through by phone.
- Act: the clinic rolls the reminders out to all doctors and starts a second cycle, adding a reply option to reschedule by text. In that cycle missed appointments fall to 11%.
Common Pitfalls
- Skipping Check, so changes are made but never evaluated.
- Running the cycle once, treating PDCA as a project rather than a continuous habit.
- Changes too big to learn from, where so much changes at once that no one can tell what worked.
- No baseline data, making improvement impossible to demonstrate.
PDCA is the engine of continuous improvement and total quality management, and a building block of operational excellence.
Related terms
Kaizen
Continuous, small improvements made by everyone, every day.
Lean Management
Maximizing customer value and minimizing waste through better flow.
DMAIC
Six Sigma's five phases: Define, Measure, Analyze, Improve, Control.
Continuous Improvement
An ongoing effort to make processes better through small, regular changes.