Featured Video
This Week in Quality Digest Live
Innovation Features
Scott Shackelford
An old idea, updated for the 21st century
Dean Solberg
If you can think about it and model it, you can 3D-print it
Jordan Kraemer
Stop doing companies’ digital busywork for free
Jeremy Straub
AI scientists should be free from concern that some AIs might be banned
Knowledge at Wharton
In hindsight we tend to attribute only talent and hard work to successful careers

More Features

Innovation News
MIT awards more than $1 million to organizations creating greater economic opportunity for workers
Berkeley Lab and Magic Leap Inc. scientists create widely controllable ultra thin optical components
$79 device delivers dedicated neural network processing to a range of host devices
Drip irrigation targets the plant and not the soil
New approach uses light instead of electricity
If you want to understand a system, try and change it
Components will be designed from the onset to inhabit multiple configurations during service
More than seven billion lives may depend on it
New interface allows more efficient, faster technique to remotely operate robots

More News

Kevin Meyer


The Simple Leader: Plan, Do, Study, Adjust

Even Deming preferred PDSA to PDCA

Published: Wednesday, November 30, 2016 - 16:35

“Excellent firms don’t believe in excellence—only in constant improvement and constant change.”
—Tom Peters

The plan-do-study-act (PDSA) cycle is the core component of continuous improvement programs. You may have heard it called the plan-do-check-act (PDCA) cycle—and they are very similar—but I have come to prefer PDSA, with the A standing for “adjust,” for reasons I’ll explain shortly. Understanding the cycle and its application to continuous improvement is critical for leadership. But first, a history lesson.

In November 2010, Ronald Moen and Clifford Norman wrote a well-researched article, “Circling Back” in Quality Progress (November, 2010) that detailed the history behind PDCA and PDSA. The cycles have their origins in 1939, when Walter Shewhart created the specification-production-inspection (SPI) cycle. The SPI cycle was geared toward mass production operations, but Shewhart soon realized the potential application of the scientific method to problem solving, writing that “it may be helpful to think of the three steps in the mass production process as steps in the scientific method. In this sense, specification, production, and inspection correspond respectively to hypothesizing, carrying out an experiment, and testing the hypothesis. The three steps constitute a dynamic scientific process of acquiring knowledge.”

At the time, W. Edwards Deming was working with Shewhart to edit a series of Shewhart’s lectures into what would become Shewhart’s Statistical Method from the Viewpoint of Quality Control, published in 1939. Deming eventually modified the cycle and presented his design-production-sales-research cycle in 1950, which is now referred to as the Deming cycle or Deming wheel. According to Masaaki Imai, Toyota then modified the Deming wheel into the PDCA cycle and began applying it to problem solving.

In 1986, Deming again revised the Shewhart cycle, with another modification added in 1993 to make it the PDSA cycle, or what Deming called the Shewhart cycle for learning and improvement. (Deming never did like the PDCA cycle. In 1990, he wrote Ronald Moen, saying: “Be sure to call it PDSA, not the corruption PDCA.” A year later he wrote: “I don’t know the source of the cycle that you propose. How the PDCA ever came into existence I know not.”)

The PDCA cycle has not really evolved in the past 40 years and is still used today at Toyota. The PDSA cycle continues to evolve, primarily in the questions asked at each stage. Although both embody the scientific method, I personally prefer the PDSA cycle, because “study” is more intuitive than “check.” Deming himself had a problem with the term “check,” as he believed it could be misconstrued as “hold back.” I also prefer “adjust” to “act,” as it conveys a better sense of ongoing, incremental improvement. Just be aware that some very knowledgeable and experienced people prefer the pure PDCA.

Let’s take a look at each component of PDSA:
Plan: Ask objective questions about the process and create a plan to carry out the experiment: who, what, when, where, and a prediction.
Do: Execute the plan, make observations, and document problems and unexpected issues.
Study: Analyze the data, compare it to expectations, and summarize what was learned.
Adjust: Adopt and standardize the new method if successful; otherwise, identify changes to be made in preparation for starting the whole cycle over again.

It’s important to realize that the PDSA cycle is valuable at both process and organizational levels. For example, you start the plan stage of the PDSA cycle while evaluating your current state and creating a hoshin plan. As you execute the annual and breakthrough objectives of the hoshin plan, you move into the “do” quadrant. On a regular basis, you evaluate the hoshin plan and the results of the goals (study), then modify it as necessary for the next revision of the hoshin plan.

This article is an excerpt from The Simple Leader: Personal and Professional Leadership at the Nexus of Lean and Zen (Gemba Academy LLC, 2016).


About The Author

Kevin Meyer’s picture

Kevin Meyer

Kevin Meyer has more than 25 years of executive leadership experience, primarily in the medical device industry, and has been active in lean manufacturing for more than 20 years serving as director and manager in operations and advanced engineering, and as CEO of a medical device manufacturing company. He consults and speaks at lean events; operates the online knowledgebase, Lean CEO, and the lean training portal, Lean Presentations; and is a partner in GembaAcademy.com, which provides lean training to more than 5,000 companies. Meyer is co-author of Evolving Excellence–Thoughts on Lean Enterprise Leadership (iUniverse Inc., 2007) and writes weekly on a blog of the same name.



Hi Kevin,

I prefer to use the cycle starting from the A point where A = Act = Review, so APDS

A = Act = Review the evidence to understand the problem and identify the causal factors and risks exposed  (AD/ID, process performance info, Swiss Cheese Model as binary..BTW, my experience is that Ishikawa is useless..)

Plan = Design what to do to eliminate the causes and/or manage the risks

Do = Carry out the action plan, apply monitors,

S = In due course, study (analyse) the results of the monitoring vs the scope of the problem, then back to

A = Review the problem  again...Is the problem under control or do we have to go around again?

I find it's a universal failing, part of the human condition, even, to come up with solutions to undefined problems..and it never helps, Dr Deming called it tampering. 



Great topic!

Too few people know this, these days. I hear lots of "PDCA," and when I bring up PDSA, I get a lot of weird looks. I can remember, though, someone asking Deming a question about "PDCA." He turned to someone next to him and asked "What was that he said?" a couple of times (Always a bad sign). Finally, he rose out of his chair and thundered, "It is P-D-S-A! Call it NOTHING else!" 

People who think he came up with it tell me to check my references, but it's PDSA in both Out of the Crisis and The New Economics. Maybe it was PDCA in earlier writing...