A3 Problem Solving: The Left Side Is the Hard Part

Most teams fill in the right-hand side of an A3 in about twenty minutes. The left takes three weeks, two site visits, and at least one argument about what the data is actually showing. That is not a failure. That is the point.

The A3 format was designed to slow you down deliberately, specifically on the left. Background, current condition, problem statement, root cause. Four boxes that most people underestimate, then wonder why their countermeasures do not hold.

What the Left Side Is Actually Asking You to Do

The left-hand side of an A3 is not a formality. It is asking you to prove that you understand the problem before you touch it.

Background: why does this matter? Current condition: what is actually happening, not what you think is happening? Problem statement: the gap between where you are and where you need to be, in plain numbers. Root cause analysis: why is that gap there?

Each of those steps requires you to go and look. Not to sit in a meeting room and reconstruct events from memory. Go to the process, watch it run, talk to the people who work it every day. What you observe and what gets reported upward are rarely the same thing.

This is the part that takes time. You might walk the line three or four times before the pattern becomes clear. You might run a 5 Whys and discover it circles back to a training gap you had already dismissed. You might find that the problem statement you wrote on day one was wrong.

That is fine. That is why you do it.

Why People Rush to Countermeasures

There are a few reasons teams skip to the right-hand side early.

The first is pressure. Someone senior wants to see action, and action looks like a column of countermeasures with owners and due dates. A box full of current-state observations does not look like progress to someone who has never run an A3.

The second is confidence. Experienced people often believe they already know the root cause. They have seen this type of problem before. The machine drifts, the supplier is inconsistent, the shift handover is sloppy. So they jump to fixes they know.

The third is discomfort. Sitting with an unresolved problem, going back to the floor a fourth time, admitting that the 5 Whys is not converging, these things feel like failure. Writing countermeasures feels like forward motion.

None of those reasons is good enough to shortcut the left side.

The Cost of a Poorly-Defined Problem

When the left side is thin, the right side fails quietly. The countermeasure gets implemented, the metric improves briefly, and then within two or three months it drifts back. At which point someone suggests doing another A3.

This is expensive in time, and it is demoralising for the people involved. They did the work, followed the process, and nothing stuck. The problem was not that the countermeasure was wrong. The problem was that the countermeasure was answering the wrong question.

A rushed root cause analysis tends to stop one or two Whys short of the actual cause. It finds the symptom closest to the surface and mistakes it for the source. The fix addresses that symptom, and the underlying condition continues.

How Long Is Long Enough?

There is no fixed answer, but a useful test is this: could you explain the current condition to someone who has never seen the process, using only data and direct observation, with no assumptions? If yes, you are probably ready to move right.

If your current condition box contains phrases like “operator error” or “machine unreliable” without supporting evidence, you are not ready. Those are conclusions, not observations.

Another test: does your problem statement contain a number? “Yield is below target” is not a problem statement. “First-pass yield on line 4 is running at 87%, against a target of 94%, over the last six weeks” is a problem statement. It tells you exactly how large the gap is and gives you something to measure a countermeasure against.

The time spent getting those details right is not time lost. It is time that prevents you from solving the wrong problem.

If your team is still deciding which problem deserves an A3 in the first place, losses.cadencestandard.com is a useful starting point. It helps you see where time and output are actually going, so you are not guessing at the biggest gap before you begin.

Once you are into the A3 itself, the six-step guide maps well to the sequence, particularly around root cause analysis and setting a measurable target condition. Worked templates for the full format are in the toolkit if your team needs a consistent structure to run across different areas of the site.