A3 problem solving
A3 Problem Solving
A3 is a structured problem-solving process documented on a single sheet: from problem understanding through root cause to verified countermeasures. The format forces clarity — if the problem can’t be explained on one page, it isn’t understood.
Use the method when
- →The problem is recurring or complex enough that quick fixes have failed
- →Several departments or viewpoints are involved, and one shared picture is needed
- →As a training format: the A3 is the best known way to learn systematic thinking
Don’t use it when
- →The problem is trivial and the solution known — then just solve it (just-do-it)
- →What’s really needed is statistical analysis of variation — then it is DMAIC territory
Before you start
One named problem owner who writes the A3 (not a committee)
A coach/leader who reviews with questions, not with solutions
Access to the gemba and to data: the A3 is written with facts, not from the meeting room
How to do it — step by step
Background and problem statement
1-2 hoursWhy is this problem important now — linked to a goal or a customer? The problem is framed as a measurable gap: “Complaint rate 3.1% vs. target 1.5%” — never as a hidden solution (“we need a new machine” is not a problem, it’s a proposed solution).
Understand the current state — at the gemba
3-6 hoursGo to the process, talk with those who work in it, and gather data: when does the problem arise, where, on which variants, how often? Stratify — the problem is rarely evenly distributed, and the skew points to the cause. Draw and visualise rather than describe.
Set a target
15 minConcrete and time-bound: from X to Y by a date. The target must follow from the gap analysis, not from what sounds ambitious.
Root cause analysis
4-8 hoursFishbone to open the breadth, 5 Whys to drill the depth on the most likely causes, and verification: a suspected root cause must be able to turn the problem on and off, otherwise it is a hypothesis. Causes that end in “human error” are drilled further: what made the error possible?
Countermeasures — tested as experiments
1-3 weeksFor each verified root cause, choose a countermeasure that attacks the cause (not the symptom). Test on a small scale with PDCA: predict the effect, carry it out, measure against the prediction. Prefer mistake-proofing and standard changes over “heightened attention”.
Verify the effect and anchor
2-4 weeks of follow-upDoes it work measured over time, not just the first week? Then the standard is updated, those affected are trained, and the measurement point is tracked on the board for a period. The A3 is closed only when data shows the gap is closed and stays closed.
The classic mistakes
How it goes wrong in practice, and this is where most implementations part ways.
The solution was chosen in advance
The A3 is written backwards to justify a decision already made. It always shows, and it teaches the organisation that the analysis is cosmetic.
The coach delivers answers instead of questions
Then the A3 becomes the leader’s analysis with the employee’s name on it, and the training effect, which is half the purpose, doesn’t materialise.
Closed without verification
“Countermeasure implemented” is not the same as “problem solved”. Without measurement over time, the A3 is a claim with nice graphics.
Related standards
The standard is free. The anchoring is the craft.
The method only works when it becomes daily routines and leadership behaviour. The operations check shows in 4 minutes where your operations stand, and what should come first.