How to write an A3 report: step by step, with an example

To write an A3 report, work through six boxes on one sheet: background, current condition, goal, root cause, countermeasures and follow-up.

An A3 is a single sheet of paper, named after its size, that tells the story of a problem from background to follow-up. The page limit forces you to understand the problem well enough to explain it simply.

John Shook's book Managing to Learn describes the A3 as a way for managers to coach. The steps below follow the usual layout.

What you need before you start

The layout

The left side tells the story so far: background, current condition, goal and root cause. The right side holds the countermeasures, the action plan, the results and the follow-up. Read left to right, the sheet follows a PDCA cycle: the left side is Plan, and the right side covers Do, Check and Act.

A title bar runs across the top with the problem name, the owner, the coach and the date. Keep the date of each revision. An A3 is rewritten several times as the owner learns more, and the old versions show how the thinking changed.

1. Background and problem statement

Explain why this problem matters now, in terms of safety, quality, delivery, cost or customers. Then describe the gap between what should happen and what actually happens.

Do not hint at a cause or a solution yet. Coaching question: would someone from another department understand why it is worth solving?

A vague statement such as 'the team is inefficient' gives nothing to investigate. A useful one names the process, the measured gap, the time frame and the impact. For example: 'Over the past six weeks, 18% of tickets in the billing queue missed the two-business-day response target.'

If the problem feels too large, split it. 'Order fulfillment delays' might become 'late picking confirmations for priority orders on the afternoon shift'. A narrower problem makes every later box easier.

2. Current condition

Go and see at the gemba. Draw the process as it really works, with data on how often the problem occurs, where and how large it is.

Coaching question: how do you know? Was this observed, or taken from a report?

3. Goal

Set a specific, measurable target and a date. The goal should close all or part of the gap in the problem statement.

Coaching question: how will we know, on the date, whether we hit it? If the answer needs a new measure, set that measure up now, before any change is made.

Don't write a solution as the goal. 'Install a new dashboard' is a possible countermeasure. 'Make overdue work visible before the daily cutoff' describes the result you want.

4. Root cause analysis

Use five whys, a fishbone diagram or both to move from symptoms to causes. Verify the likely causes with data or a quick test before acting.

Coaching question: if you removed this cause, would the problem disappear? How could you check that cheaply?

Skipping this step has a measurable cost. In a NIST survey of US manufacturers, plants that relied most on reactive maintenance, fixing equipment after it fails, had 3.3 times more downtime and 16 times more defects than plants that relied on it least.

5. Countermeasures and plan

List countermeasures that address the verified causes, each with an owner and a date. Call them countermeasures, not solutions, because they are hypotheses until the results come in.

Coaching question: which root cause does each countermeasure address? A countermeasure that maps to no verified cause is a guess and should come off the sheet.

Five checks for each countermeasure

6. Follow-up

Check results against the goal. If the target was met, update the standard work so the gain sticks. If not, return to the root cause analysis.

A standard only holds if people can follow it. In the Uptime Institute's 2025 outage analysis, 85% of major data center outages caused by human error stemmed from staff not following procedures, or from flaws in the procedures themselves.

Coaching question: what did we learn that another team could use? Share the A3 with teams that run a similar process, and note any problems the work uncovered that need an A3 of their own.

Check two kinds of measure. A result measure shows whether the outcome improved, such as fewer late orders. A process measure shows whether the new method is being followed, such as orders entered with all required fields. A good result with no process change is often luck.

When the A3 closes, name who will watch the process from now on, how, and what would trigger reopening the problem.

A3 report example

The example below fills in every box for one problem. It is illustrative: the company and numbers are made up to show the level of detail each box needs, not taken from a real case.

Title bar: Late shipments on the Series B pump line. Owner: order desk lead. Coach: operations manager. Version 3, dated at each revision.

Background: the largest customer for Series B pumps requires 95% of orders to ship within two working days. The contract is up for renewal in six months, and the customer has raised late orders twice this quarter.

Current condition: over the last eight weeks, 78% of Series B orders shipped within two days (312 of 400). The owner followed 20 late orders from receipt to dispatch. 14 of them waited more than a day at the credit check, 4 waited for a part, and 2 had address errors. Picking and packing took under three hours for every order observed.

Goal: 95% of Series B orders ship within two working days, measured weekly, by the end of next quarter.

Root cause: five whys on the credit delay. Why did orders wait? The credit check had not been done. Why not? Checks run once a day, at 4 pm. Why once a day? The credit team batches them to clear in one sitting. Why batch? Each check means opening three systems, so the team avoids switching back and forth. Why three systems? Customer limits, open invoices and payment history sit in separate tools. The team tested the cause by running checks every two hours for one week. That week, 2 orders waited more than a day for credit, against 14 in the 20 late orders studied before.

Countermeasures: 1. Orders under the customer's approved limit with no overdue invoices skip the manual check; owner: finance systems analyst; due in four weeks. 2. Manual checks run at 10 am and 2 pm instead of 4 pm; owner: credit team lead; due next week. 3. A one-screen credit view that pulls the three sources together; owner: IT; due in ten weeks. The part shortages go to a separate A3, owned by purchasing.

Results and follow-up: by week six, 93% of orders shipped within two days. The credit view went live in week ten, and over weeks eleven to fourteen 96% of orders shipped on time. The new check rules went into the order desk standard work, and the weekly on-time chart stays on the team board. Purchasing's A3 on part shortages is still open.

A second example: an office queue

A3 template checklist

Draw the A3 by hand on an 11 x 17 inch or A3 sheet, or use a single landscape page in a document. Before you show it to a coach, check each box against these questions.

Title bar: does it name the problem, the owner, the coach and the date of this version?

Background: would someone from another department see why this matters now, in terms the business tracks?

Current condition: is there a picture of the process, and is every number something you saw or measured yourself?

Goal: is there one measure, a target value and a date?

Root cause: does the chain of causes rest on evidence, and was the main cause tested before acting?

Countermeasures: does each one link to a verified cause, with an owner and a due date?

Follow-up: is there a date to check results, and a plan for what happens if the goal is missed?

Common mistakes when writing an A3

Frequently asked questions