# How to Make a Cause and Effect Diagram, Step by Step

Canonical: https://sakura5.app/blog/how-to-make-a-cause-and-effect-diagram/  
Published: 2026-10-31  
Author: the Sakura5 team · 7 min read  
Updated: 2026-10-31  
Publisher: Sakura5 (https://sakura5.app) · support@sakura5.app

> **TL;DR:** How to make a cause and effect diagram that finds the real problem: a measurable effect, the right categories, checkable causes, and a worked example.

![How to Make a Cause and Effect Diagram, Step by Step](https://sakura5.app/img/blog/inline/how-to-make-a-cause-and-effect-diagram-1.jpg)

To make a cause and effect diagram, write the problem as a specific effect in a box on the right, draw a horizontal spine leading to it, add four to six angled “bones” for cause categories such as people, methods, materials, machines, measurement and environment, then list possible causes along each bone and keep asking “why” until every cause is something you can check. That shape is why it’s also called a fishbone or Ishikawa diagram.

The diagram matters because most teams argue about fixes before they agree on causes. A cause and effect diagram puts every theory on one page, grouped so nothing obvious gets skipped, and it makes the gap between “someone thinks this” and “we checked this” visible.

Below: why most fishbone diagrams end up as decoration, a seven-step method for building one, a full worked example from a small bakery, what to do when the categories don’t fit, and short answers to common questions.

## Why most cause and effect diagrams end up as wall art

The usual version gets drawn in a meeting, photographed, and never looked at again. Three things cause that.

**The effect is too vague.** “Customer complaints” is a topic, not an effect. Every cause in the room fits under it, so the diagram fills up with forty sticky notes and no direction. A good effect has a number and a time frame: “Late deliveries rose from 2 to 9 per week since March.”

**Causes and solutions get mixed.** Someone writes “need more staff” on the people bone. That’s a fix wearing a cause’s clothes. The cause might be “two drivers on leave in April,” and the fix might be something else entirely.

**Nobody checks anything.** The diagram is a list of hypotheses. If it ends the moment the last sticky note goes up, you’ve brainstormed, not diagnosed. The value comes from the step afterward, where you mark which causes are confirmed and which were just loud.

## How to make a cause and effect diagram, step by step

Plan on 30 to 45 minutes for the first draft, alone or with three to six people who know the process. Paper, a whiteboard or a digital canvas all work; you’ll move causes around a lot, so something editable helps.

![How to make a cause and effect diagram, step by step — how to make a cause and effect diagram](https://sakura5.app/img/blog/inline/how-to-make-a-cause-and-effect-diagram-2.jpg)

*Plan on 30 to 45 minutes for the first draft, alone or with three to six people who know the process.*

### 1. Write the effect as a measurable gap

Put the effect in the “head” of the fish, on the right. State what should happen, what is happening, and since when. If the group can’t agree on the sentence, stop and settle it first — a disputed effect produces a disputed diagram.

### 2. Draw the spine

One horizontal line from the left edge to the effect box. It’s just structure, but it sets the reading direction: causes on the left flow toward the effect on the right.

### 3. Choose four to six cause categories

The classic manufacturing set is the 6 Ms: manpower (people), methods, machines, materials, measurement and mother nature (environment). Service teams often use the 4 Ps instead: policies, procedures, people and plant. You don’t have to use either. Pick the categories where causes could plausibly live for *this* effect and draw each as an angled bone off the spine.

### 4. Brainstorm causes on each bone

Go bone by bone, not all at once. Working one category at a time stops the group from piling every idea under “people” and leaving “measurement” empty. Write causes as short factual phrases, and push back on anything phrased as a fix.

### 5. Ask “why” to add sub-causes

For each cause, ask why it would happen and attach the answer as a smaller branch. Two or three levels is usually enough. Stop when a cause is checkable — something you could confirm with a log, a count, a conversation or a quick test.

### 6. Mark each cause: confirmed, ruled out, or unknown

This is the step that turns the drawing into a diagnosis. Gather the evidence, then tag every leaf. Often two or three confirmed causes cluster on the same bone, and the real problem stops being a matter of opinion.

### 7. Act on the confirmed causes and set a recheck date

Write one action per confirmed cause, an owner, and a date when you’ll look at the effect’s number again. Keep the diagram; if the number doesn’t move, the unknown leaves are where you go next.

If the blank bones are where you stall, [Mindmap AI](https://sakura5.app/mindmapai/) can generate a starting map from a one-line problem statement. Put the effect in the center, treat the first ring of branches as your categories, and with MindMap Pro tap any cause to expand it into sub-causes with AI. What it suggests is a list of candidates to check, not a verdict. The radial layout isn’t the classic fish shape, but the logic — effect, categories, causes, sub-causes — is identical.

## A worked example: late Saturday deliveries at a neighborhood bakery

Here’s a scenario to make the steps concrete. Hollis Street Bakery delivers pre-ordered cakes and bread boxes on Saturday mornings. Over the last six Saturdays, 14 of 62 orders arrived after the promised window, up from one or two late orders a week before. The owner, Dana, and two staff spend one Monday afternoon on it.

![A worked example: late Saturday deliveries at a neighborhood bakery — how to make a cause and effect diagram](https://sakura5.app/img/blog/inline/how-to-make-a-cause-and-effect-diagram-3.jpg)

*Here’s a scenario to make the steps concrete.*

**Effect (head of the fish):** “Saturday deliveries late on 14 of 62 orders over six weeks; previously 1–2 per week.”

**Categories (five bones):** people, methods, materials, equipment, environment.

- **People:** a new driver started six weeks ago. Why would that matter? He doesn’t know the shortcut around the Saturday market. Checkable: compare late orders by driver.
- **Methods:** orders are loaded in the order they were baked, not by route. Why? Nobody wrote a loading sequence down. Checkable: watch one loading.
- **Materials:** the sourdough proofs longer in cold weather, so the first batch comes out later. Checkable: oven log start times.
- **Equipment:** the second oven had a repair in September. Checkable: maintenance receipt and bake times.
- **Environment:** the farmers’ market moved to a new street, closing a road on the main route. Checkable: city notice and the delivery map.

**Checking (about two hours):** late orders are split evenly between both drivers — the new driver is ruled out. The oven log shows bake start times unchanged — materials and equipment ruled out. The market moved six weeks ago, and 11 of the 14 late orders were on addresses east of it — **confirmed**. Loading by bake order means east-side boxes often go out last — **confirmed** as a contributing cause.

**Actions:** plan a new route around the closed street (Dana, this week); load the van by route, east-side orders first (staff, starting Saturday); recheck the late count after four Saturdays. The first instinct in the room had been to blame the new driver. The diagram shows that would have fixed nothing.

## When the categories don’t fit, or the diagram sprawls

**The 6 Ms feel forced.** For a software bug, a study habit or a marketing drop, “machines” and “materials” may be empty. Replace them. A website conversion problem might use traffic, page, offer, checkout, tracking and timing. The categories are a checklist against tunnel vision, not a rule.

**There are fifty causes.** That nearly always means the effect is too broad. Split it: “the shop is struggling” is several effects, each worth its own diagram. Around five bones and three levels deep is a sensible limit before you start checking.

**The cause is a sequence, not a list.** Some problems are a chain of events — A triggered B, which caused C. A fishbone shows parallel possibilities well but chains poorly. Once you’ve confirmed the cause, a decision tree or a [troubleshooting flowchart](https://sakura5.app/blog/flowchart-for-troubleshooting-a-problem/) is the better tool for making the fix repeatable.

**You want solutions on the same page.** Classic fishbones stop at causes. If you’d rather hang fixes directly off confirmed causes, a free-form map works better; our guide to [using mind mapping for problem solving](https://sakura5.app/blog/how-to-use-mind-mapping-for-problem-solving/) walks through that version. In [Mindmap AI](https://sakura5.app/mindmapai/), maps sync through your own iCloud account across iPhone, iPad and Mac, and the outline view turns the confirmed branch into a short write-up for whoever needs the answer.

For background on where the method comes from and how quality teams use it, the [American Society for Quality’s fishbone diagram overview](https://asq.org/quality-resources/fishbone) is a solid reference.

## FAQ

**Is a cause and effect diagram the same as a fishbone diagram?**

Yes. Cause and effect diagram, fishbone diagram and Ishikawa diagram all name the same tool: a problem on the right, a spine leading to it, and angled bones for categories of causes. The name “Ishikawa” comes from Kaoru Ishikawa, who popularized it in Japanese quality management. “Fishbone” simply describes the shape.

**What are the main categories in a cause and effect diagram?**

Manufacturing teams often use the 6 Ms: manpower, methods, machines, materials, measurement and mother nature. Service teams commonly use policies, procedures, people and plant. Neither list is mandatory. Choose four to six categories where causes for your specific problem could plausibly live, and replace any that stay empty.

**How many causes should a fishbone diagram have?**

There’s no fixed number, but a workable diagram usually has three to six causes per bone and two or three levels of sub-causes. If you pass fifty items, the effect is probably too broad and should be split. What matters more than the count is that every leaf is specific enough to check.

**What’s the difference between a cause and effect diagram and the 5 Whys?**

The 5 Whys follows one chain of causes deeply by asking why repeatedly. A cause and effect diagram spreads wide first, listing many possible causes across categories. They work well together: build the fishbone to see every candidate, then use repeated whys on each bone to reach a cause you can actually confirm.

About the app: [MindMap AI: Mind Map Maker](https://sakura5.app/mindmapai/) · App Store: https://sakura5.app/go/mindmapai?from=/blog/how-to-make-a-cause-and-effect-diagram/