Sample frameDay 318·Operations · Diagnosis

Complicated failures rarely have one cause. They have two or three that lined up on the same morning.

Operations.

The Cause Map

5 min read·Apply by lunch

The question

When something last went wrong badly enough to need a written explanation — the missed shipment, the wrong invoice, the error a client found before you did — how many candidate causes were on the table before the room settled on one, and how many questions deep did the settled one go? If the honest answers are one and three, you didn't investigate. You confirmed the first idea that fitted.

The idea

Kaoru Ishikawa taught engineering at the University of Tokyo and spent his career inside Japan's postwar quality movement. In 1943, explaining to engineers at Kawasaki Steel how a tangle of factors related to a single outcome, he drew what the world now calls the fishbone: the problem at the head, and a rib for each category of cause — machine, method, material, manpower, later widened to take in measurement and environment. He set it out for a general audience in Guide to Quality Control, which appeared in Japan in 1968 and in English a few years later. The diagram does one job extremely well. It forces breadth before depth, so a room has to populate every rib before it's allowed to fall in love with the first plausible answer. Depth is Taiichi Ohno's contribution: at Toyota he insisted on asking why five times, and his own worked example runs from a machine that stopped, to an overload that blew a fuse, to a bearing that wasn't properly lubricated, to a pump that wasn't pumping, to a worn pump shaft — and lands on a missing strainer that had been letting metal scrap into the pump. Five whys, and the answer is a part you can fit this afternoon. Stop at three and the answer is a pump: it tells you what to replace this week, and nothing about why you'll be replacing it again in March.

The framework

01
Populate every rib before you argue about any of them.
The categories are scaffolding, not truth — their only job is to stop the room converging on the first idea somebody says out loud. If machine, method, material and manpower don't map to your work, use the service version: people, process, policy, system, supplier, information. Give everyone five silent minutes to write candidate causes under each heading, then read them out without debate or ranking. The finding is often an empty rib — a whole category nobody thought about, which tends to be the one that isn't anybody's job to think about.
02
The name at the end of a why-chain.
Ask why enough times and you'll eventually arrive at a person. The retrospective rule from Day 44 holds here too — the review examines the system, never the individual — and in a cause map it has a mechanical use rather than a moral one. A name is not a stopping point; it's a signal you're one question short. If the answer is that someone entered the wrong figure, the next why is why the step accepted a wrong figure at all, and the one after that is why nothing caught it for three days.
03
How far down do you stop?
Five is a rule of thumb rather than a law; some chains close at three, some run to seven. The real test is two-sided: the answer names something you can actually change, and changing it would have prevented this particular failure. Then read the chain backwards, putting the word therefore between each link, and see whether it still holds. If a step doesn't survive that reading, you jumped over something — and the link you skipped is usually where the interesting cause was hiding.
04
Most failures have several causes and no single root.
The phrase root cause promises one villain, and complicated failures rarely oblige. Typically two or three conditions had to line up on the same day — the checklist nobody had updated, the person covering someone's leave, the system that took the entry without a murmur — and removing any one would have been enough. So the map isn't a verdict but a shortlist, and the last step is a choice: of the conditions you found, which is cheapest to change and blocks the most chains? That one gets an owner and a date. Write the others down anyway — a condition you've already named is worth a day of the next investigation.

Members only · 4 principles + template + AI mentor

364 more frameworks are waiting.

The Cause Map is Day 1 of 365. One framework every morning for a year — across strategy, sales, negotiation, leadership, and more.

$1 /day

Billed $365/yr · cancel any time

365 frameworks

12 topics · templates · AI mentor

Reserve your place — $365/yr →

One payment a year · the whole library · no auto-renewal.

More Operations frameworks

All 24 operations frameworks →