Quick reference
Memory aid
Plan it, Do it (small), Check it (measure), Act on it (standardize). Repeat forever.
TL;DR
The simplest improvement cycle: Plan a change hypothesis, Do a small-scale test, Check the results against your hypothesis, Act to standardize or adjust. Never stop cycling.
What Is PDCA Cycle?
PDCA is the simplest improvement cycle: Plan what you want to change, Do it on a small scale, Check whether it worked, and Act — either standardize it or try something different. Then repeat.
PDCA embodies the scientific method applied to management. Plan develops a hypothesis about what change will improve outcomes. Do tests that hypothesis in a controlled way. Check compares results against expectations. Act decides whether to adopt, adapt, or abandon the change. The cycle never truly ends — each Act sets up the next Plan. PDCA is the foundation for virtually every modern improvement methodology, from Lean to Six Sigma to Agile retrospectives.
PDCA Cycle (Deming Wheel)
A continuous improvement cycle: Plan a change, Do a small test, Check the results, Act to standardize or adjust.
Plan
Identify & hypothesize
Do
Test on small scale
Check
Measure results
Act
Standardize or adjust
Origin & context
Shewhart described a specification-production-inspection cycle in his 1939 book Statistical Method from the Viewpoint of Quality Control. Deming adapted it in his 1950 lectures in Japan, and Japanese executives recast that version as Plan-Do-Check-Act, which became foundational to the Japanese quality movement and the Toyota Production System.
Core Components
Plan
Identify the problem, analyze root causes, and develop a hypothesis for improvement.
Example
Plan: Customer complaints about delivery speed increased 20%. Root cause analysis suggests the order review step adds 2 days. Hypothesis: Automating review will reduce delivery time by 2 days.
Do
Implement the change on a small scale to test the hypothesis.
Example
Do: Automate order review for one product line as a pilot for 2 weeks.
Check
Measure results and compare to the hypothesis.
Example
Check: Delivery time for the pilot dropped by 1.5 days. Customer complaints for that line fell 35%.
Act
If successful, standardize and scale. If not, analyze why and plan the next cycle.
Example
Act: Results positive but not fully at target. Standardize automation for this line and plan a second cycle to address the remaining 0.5-day gap.
Deming actually preferred 'PDSA' (Plan-Do-Study-Act) because 'Study' implies analyzing and understanding the data, while 'Check' can be reduced to a simple pass/fail. Despite Deming's preference, PDCA became the more popular acronym. The cycle grew out of the one Walter Shewhart described in 1939, which Deming adapted in Japan in 1950 before Japanese executives recast it as PDCA.
When to Use PDCA Cycle
Daily operational improvement
Problem it solves: Provides the simplest framework for testing and implementing process improvements.
Real-world application
A warehouse team uses PDCA weekly: Plan (identify the biggest source of picking errors), Do (test a new shelf labeling system in one aisle), Check (measure error rate), Act (roll out to all aisles or try a different approach).
Foundation for larger improvement methodologies
Problem it solves: PDCA is the engine inside Lean, Six Sigma, and other frameworks.
Real-world application
Every kaizen event follows PDCA. Every Six Sigma DMAIC project embeds PDCA. Every Agile retrospective is a PDCA cycle for team process.
The most neglected step is Check. Teams often implement changes (Do) and assume they worked without measuring. Always compare results to your Plan hypothesis.
How to Apply PDCA Cycle: Step by Step
Before you start
- →A problem or improvement opportunity
- →Basic measurement capability
- →Willingness to experiment
Plan
Define the problem, gather data, analyze root causes, and develop a change hypothesis.
Tips
- ✓Use data, not assumptions, to understand the problem
Common mistakes
- ✗Skipping root cause analysis and jumping to solutions
Do
Implement the change on a small scale.
Tips
- ✓Keep the test small and time-limited to reduce risk
Common mistakes
- ✗Rolling out the change everywhere before testing
Check
Measure results and compare to your hypothesis.
Tips
- ✓Use the same metrics you measured in Plan for comparison
Common mistakes
- ✗Not measuring, or changing metrics mid-cycle
Act
Standardize if successful; adjust and re-plan if not.
Tips
- ✓Document the new standard so improvements stick
Common mistakes
- ✗Moving on without standardizing, allowing regression
Value & Outcomes
Primary benefit
The simplest, most universal framework for testing and implementing improvements systematically.
Additional benefits
- ✓Reduces risk by testing on small scale first
- ✓Embeds the scientific method into daily management
What you'll learn
- →How to apply scientific thinking to operational improvement
- →How to test changes before full-scale implementation
Typical outcomes
Best Practices
📋 Preparation
- •Start with a clear problem statement and data
- •Define success criteria before starting
🚀 Execution
- •Keep cycles short — 1-4 weeks
- •Always measure in Check, don't skip it
🔄 Follow-Up
- •Standardize successful changes
- •Start the next PDCA cycle immediately
💎 Pro Tips
- •PDCA is fractal — you can run multiple cycles at different levels simultaneously
At Toyota, front-line workers are expected to run small PDCA experiments as part of daily work, a habit at the heart of the Toyota Production System.
Limitations & Pitfalls
Too simple for complex, multi-variable problems
Mitigation: Use DMAIC (Six Sigma) for complex problems requiring statistical analysis
Requires discipline to complete all four steps — most people skip Check
Mitigation: Make Check a formal step with defined metrics and a review meeting
Where this comes fromWalter Shewhart / W. Edwards Deming introduced this (1939).
More Management Frameworks
Other frameworks in the Stratrix library.
Design Thinking
A human-centered innovation methodology with five phases — Empathize, Define, Ideate, Prototype, and Test — that uses deep user understanding and rapid experimentation to solve complex problems creatively.
Project & Product ManagementLean Startup
An entrepreneurial methodology that uses a Build-Measure-Learn feedback loop, minimum viable products (MVPs), and validated learning to rapidly test business hypotheses and reduce the risk of building products nobody wants.
Project & Product ManagementRACI Matrix
A responsibility assignment tool that clarifies roles for every task or decision by defining who is Responsible (does the work), Accountable (owns the outcome), Consulted (provides input), and Informed (kept in the loop).
Project & Product ManagementPRINCE2
A structured project management methodology that divides projects into manageable stages with defined roles, processes, and controls, providing clear governance and accountability for project delivery.
Project & Product ManagementKanban
A visual workflow management method that uses boards, cards, and work-in-progress limits to optimize the flow of work through a system, enabling teams to deliver continuously without the structure of fixed-length sprints.
Project & Product ManagementJobs-to-be-Done
A product innovation framework that focuses on the underlying 'job' customers are trying to accomplish rather than the product itself, revealing unmet needs and disruptive innovation opportunities.
More from Stratrix
Test a decision, free
Set a live decision against the companies that made it before you.
AcquisitionsThe Distance TestDivestituresThe Divestiture TestDiversificationThe Diversification TestMake or buyThe Make-or-Buy TestAll Access
Every Casebook, three Strategy Packs a seat each year and the Analyst, for teams that face these decisions all year.
$1,999 a year · $499 a seatWhat is included →New analyses, by email
The defining moves in business, checked against the record, as they publish.
See PDCA Cycle in the wild.
Follow PDCA Cycle across the companies and lenses where it actually shaped the strategy — the decision, the rationale, and how it played out.