Skip to content

Question list · Root cause analysis and problem solving

5W1H questions: 48 to define a problem, and why you ask each one

Short answer

5W1H questions are the questions a team asks under what, where, when, who, why and how to describe a problem in facts before anyone looks for the cause. The answers are written up as one problem statement.

The 5W1H questions are what, where, when, who, why and how. In problem solving you ask them first, so that everyone is working on the same problem. What is the object and what is wrong with it. Where is it found, and where on the part. When did it start, and what is the pattern. Who found it and who is affected, as roles, not names. Why does it matter. How was it detected, and how many. This page lists 48 questions with the reason you ask each one: 41 under those six words and seven for the action plan. Use two or three per word, answer with facts you have checked, and stop when you can write the problem in two sentences that contain no cause.

Last reviewed:

How to use this list

You do not need all 48. Under each of the six words, pick the two or three questions that fit your problem and answer them at the place where it happened. A word is finished when its answer is a fact that someone else could go and check.

The 48 questions in seven groups

  1. Questions 1 to 7

    What

    Name the thing that has the problem and what is wrong with it.

  2. Questions 8 to 14

    Where

    Pin down where it is found, where on the part, and where it is not.

  3. Questions 15 to 21

    When

    Fix the start date and the pattern over time.

  4. Questions 22 to 27

    Who

    Record who found it, who is affected and who owns it.

  5. Questions 28 to 33

    Why

    State why the problem is worth solving now.

  6. Questions 34 to 41

    How

    Count it, measure it and say how you know.

  7. Questions 42 to 48

    Action plan

    Turn each countermeasure into a task someone can check.

The first six groups define the problem. The seventh uses the same six words again to plan the fix.
  • Answer with what you have seen or measured. Mark anything you were told and have not yet checked.
  • Use numbers and names of things, not adjectives. Many, often and bad cannot be compared with next week.
  • Who means a role, a shift or a station. It never means the person to blame.
  • Why means why the problem matters. Keep the cause out until the description is finished.
  • Ask where and when the problem does not happen, as well as where and when it does.
  • Finish by writing one problem statement of two or three sentences from the six answers.

The 5W1H template has a row for each of the six questions, a column for how you know, a line for the problem statement and a 5W1H action plan. The six questions are much older than lean. Rudyard Kipling called them his six honest serving-men in a poem in Just So Stories (1902), which is why the method is sometimes called the Kipling method, and writers on rhetoric were teaching the same set of questions many centuries before him.

Write the answers on the 5W1H template

The six questions on one sheet, the problem statement built from the answers and an action plan. Free, after one short form.

Free to use: print it, copy it and edit it for your team. Enter your name and work email once to download.

See what is on each sheet on the 5W1H template page.

What: questions about the object and the defect

Start with what. Until the team agrees on the object and the defect, every other answer is about a slightly different problem. Answer with a part number and a description someone could photograph.

  1. 1

    What part, product or machine has the problem?

    Why: Name the part number or the asset, not the family. The brackets can mean twelve part numbers, and the problem may live in one of them.

  2. 2

    What exactly is wrong with it?

    Why: Describe the defect as you would see it: a crack, a short shot, a missing label. Words like bad or out of spec send people looking in different places.

  3. 3

    What should it be?

    Why: A problem is a gap from a standard. Write the specification or the target next to the actual value, or nobody can say how big the gap is.

  4. 4

    What did you see with your own eyes, and what were you told?

    Why: Reports get rounded off as they travel. Mark what was observed at the machine and what is secondhand, then go and check the secondhand part.

  5. 5

    What similar part or machine does not have the problem?

    Why: The twin that runs fine is the fastest clue you will get. Whatever differs between the two is a short list of suspects.

  6. 6

    What does a defective part look like next to a good one?

    Why: Put them side by side and photograph them. Differences nobody mentioned show up, and the photo ends arguments about what the defect is.

  7. 7

    Is it one kind of defect or several?

    Why: Two defects with two causes often hide under one name. Split them now, or the countermeasure will fix half the problem.

Where: questions about location

Where has three meanings and you need all of them: where in the plant the problem was found, where in the process it comes from, and where on the part it sits. Keep found and made apart until you have evidence that they are the same place.

  1. 8

    Where was the problem found?

    Why: The place it was detected is a fact. The place it was made is still a guess, so record them separately.

  2. 9

    Where in the process could it first have been seen?

    Why: If the defect was visible three steps earlier, there is a detection problem as well as a cause problem.

  3. 10

    Where on the part is the defect?

    Why: Mark it on a drawing or a photo. Defects that cluster in one spot point to a cavity, a fixture point or a tool.

  4. 11

    Which line, machine, cavity or station does it come from?

    Why: A problem on one of four identical machines is a different problem from one on all four. Check before the team redesigns the process.

  5. 12

    Where does it not happen?

    Why: A line, shift or plant that runs the same part without the problem rules out a long list of causes in one step.

  6. 13

    Where is the suspect stock right now?

    Why: Parts in the warehouse, in transit and at the customer all need a decision today. This question starts containment before the cause is known.

  7. 14

    Where did the material or components come from?

    Why: Lot and supplier narrow the search. If every bad part traces to one coil or one delivery, the investigation moves to receiving.

When: questions about timing and pattern

A date and a pattern do more to narrow the search than any opinion. These questions set the window in which something changed and show whether the problem is constant, drifting or coming and going.

  1. 15

    When was it first noticed?

    Why: Give a date and a time, not a rough period. The team will lay this beside the change log, and last week matches everything.

  2. 16

    When did it last run without the problem?

    Why: The cause arrived between the last good run and the first bad one. This question sets the window to search.

  3. 17

    When in the cycle or the shift does it happen?

    Why: Problems at startup, after breaks and at the end of a tool's life each point somewhere different. Without the timing they all look random.

  4. 18

    Is it constant, getting worse, or does it come and go?

    Why: A sudden step points to an event. A slow drift points to wear or contamination. An on and off pattern points to something that alternates, such as a material lot or a shift.

  5. 19

    What else changed around that time?

    Why: List new lots, tool changes, maintenance work, new people and program edits before memory fades. Do not judge them yet. That belongs to the root cause step.

  6. 20

    When does it never happen?

    Why: A product, a day or a shift with a clean record tells you what the problem depends on.

  7. 21

    How long did it go on before somebody raised it?

    Why: A long gap means people had learned to live with it. That is worth knowing when you later ask why it was not detected.

Who: questions about roles, never blame

Who is the question that goes wrong most often. It asks which roles, shifts and stations are involved, so the team can see a pattern. It does not ask whose fault it is. Write roles in the problem description and keep names for the owner of the problem.

  1. 22

    Who found it?

    Why: Write the role and the step: final inspector, the customer's receiving dock, the operator at the next station. It shows how far the defect travelled before someone caught it.

  2. 23

    Who is affected by it?

    Why: The operator who reworks it, the customer who receives it and the planner who reschedules around it each carry a different cost. Naming them gets the problem the right priority.

  3. 24

    Who runs the process when it happens?

    Why: Record the shift, crew or skill level, not a name. The purpose is to see a pattern, such as new starters or one crew's method, and names turn that into blame.

  4. 25

    Who has seen it before?

    Why: Someone on another shift or in maintenance may have met this problem years ago. Ten minutes with them can save a week.

  5. 26

    Who needs to know today?

    Why: The customer, the quality manager, the next shift. Deciding this while you write the problem down stops the news arriving by rumour.

  6. 27

    Who owns the problem from here?

    Why: One named owner, not a department. A problem owned by quality in general is owned by nobody.

Why: questions about why it matters, not the cause

In 5W1H the why is why this is a problem, not why it happened. The cause comes later, from a 5 Whys or a fishbone built on these facts. If a cause slips into the description, the team stops looking for any other.

  1. 28

    Why is this a problem?

    Why: State what goes wrong because of it: a safety risk, a customer complaint, scrap cost, a missed shipment. If nobody can say, it may not deserve a team.

  2. 29

    Why does it need solving now?

    Why: A problem competing with ten others needs a reason for its place in the queue. A customer deadline or a rising trend is one.

  3. 30

    Which target or standard does it break?

    Why: Tie the problem to a number the plant already tracks, such as scrap rate or on-time delivery. That is how you will know later that it is solved.

  4. 31

    What happens if we do nothing for a month?

    Why: It turns a vague worry into a cost or a risk. The answer also sets how much effort the fix is worth.

  5. 32

    Why was it not caught earlier?

    Why: This asks about detection, not cause. Keep the answer for the investigation. A problem usually needs one fix where it was made and one where it got through.

  6. 33

    Are we describing the problem, or already guessing the cause?

    Why: The die is worn is a cause. Burrs over 0.3 mm on the lower edge is a problem. A cause in the statement closes the investigation before it begins.

How: questions about how it was found and how big it is

How covers two things: how the problem shows itself and how much of it there is. This is where adjectives become numbers. The 5W2H version of the method splits how much into a seventh question, and it is included here.

  1. 34

    How was it detected?

    Why: By a gauge, a visual check, a test rig or a customer return. The method tells you how far to trust the count and whether earlier parts could have slipped through.

  2. 35

    How many parts, or how many times?

    Why: Count them. A lot of scratches cannot be compared with next week. 37 of 1,200 can.

  3. 36

    How big is that as a rate?

    Why: Thirty-seven defects mean one thing in a run of 1,200 and another in a run of 120,000. The rate lets you compare shifts, lines and weeks.

  4. 37

    How far is it from the specification?

    Why: Just outside the limit and far outside are different problems. Measure the size of the miss, not only pass or fail.

  5. 38

    How is the trend moving?

    Why: Three points are enough to see rising, flat or falling. A rising trend changes the urgency.

  6. 39

    How do we know the measurement is right?

    Why: Check the gauge and the method before chasing the process. Now and then the problem is in the measuring.

  7. 40

    How much has it cost so far?

    Why: Put rough numbers on scrap, rework hours, sorting and extra freight. This is the how much that 5W2H adds as a seventh question.

  8. 41

    How does the work normally get done at this step?

    Why: Write the standard method in a line or two. You need it to see whether the problem appears when the standard is followed or when it is not.

5W1H questions for the action plan

The same six words are used a second time, once the cause is known, to write the actions. Here the why does point at the cause. Ask these for every line on the action plan.

  1. 42

    What exactly will be done?

    Why: One action per line, starting with a verb. Look into fixture wear is not an action. Measure the locating pins on fixture 2 is.

  2. 43

    Why will this action fix the problem?

    Why: Each action should point at a cause you have confirmed. If you cannot say which one, the action is a guess.

  3. 44

    Where will it be done?

    Why: Name the line, the machine or the document. A fix on line 2 that should also go on line 4 gets missed when the place is left vague.

  4. 45

    When is it due?

    Why: Give a date, not as soon as possible. A date can be checked at the daily meeting.

  5. 46

    Who owns it?

    Why: One name per action. Two owners each wait for the other.

  6. 47

    How will it be done?

    Why: Write the method and what it needs: a spare part, a shutdown window, a trial lot. Actions stall on the thing nobody arranged.

  7. 48

    How will we know it worked?

    Why: Decide the measure and the check date before the action starts. Otherwise done means installed, not effective.

Weak 5W1H answers, and how to fix them

The questions are easy to ask. The work is in the answers. Each weak answer below is one a team might write in the first five minutes, with the version it should hold out for.

Weak 5W1H answers and better ones

Example numbers

  • Question

    What

    Weak answer: Quality problems on the brackets.

    Better answer

    Burrs over 0.3 mm on the lower edge of bracket 4471. The limit is 0.1 mm.

  • Question

    Where

    Weak answer: In stamping.

    Better answer

    Press 6, station 3 of the die. Presses 4 and 5 run the same part with no burrs.

  • Question

    When

    Weak answer: Recently.

    Better answer

    First seen on Tuesday on second shift, in the first run after the die came back from sharpening.

  • Question

    Who

    Weak answer: Second shift keeps causing it.

    Better answer

    Found by the assembly operator at the next step. It appears on every shift that runs press 6.

  • Question

    Why

    Weak answer: Because the die is worn.

    Better answer

    Burred parts do not seat in the weld fixture. Assembly lost 50 minutes to sorting on Tuesday.

  • Question

    How

    Weak answer: Lots of them.

    Better answer

    37 of 1,200 parts in the run, about 3 percent, found with the go/no-go gauge at assembly.

The better answer is one a stranger could check: a part number, a place, a date, a count.

Worked example: from a complaint to a problem statement

Illustrative example: the plant, the product and the numbers are made up. A food plant packs a 250 g stand-up pouch on two fillers. On Wednesday morning the shift report says leakers on filler 2 again. The team leader and the quality technician take the template to the line and answer one question at a time.

5W1H for leaking pouches on filler 2

Example numbers

  • What

    Answer the team wrote
    The side seal on the 250 g pouch leaks. The seal looks complete but peels open under a light squeeze.
    How they know
    Squeeze test on 20 retained pouches. Three opened.

  • Where

    Answer the team wrote
    Filler 2, on the left seal only. Filler 1 runs the same pouch and the same film without leaks.
    How they know
    Leak position marked on 14 pouches, all on the left seal. Filler 1 check sheets read for the same week.

  • When

    Answer the team wrote
    First found on Monday night shift, two hours after a film roll change. Not seen in the three weeks before.
    How they know
    Hourly seal check sheets and the roll change log.

  • Who

    Answer the team wrote
    Found by the packer at the case station, not at the hourly seal check. Seen on all three shifts since Monday.
    How they know
    Hold tags from Monday to Wednesday.

  • Why

    Answer the team wrote
    A leaking pouch spoils the case it is packed in. One pallet is on hold and the customer's order ships on Friday.
    How they know
    Hold log and the shipping schedule.

  • How

    Answer the team wrote
    14 leakers in 3,600 pouches checked, about 0.4 percent. The target is none found at the case station.
    How they know
    Every pouch on two pallets squeezed by hand and counted.
Each answer names where the fact came from, so nobody has to take it on trust.

The problem statement they wrote: since Monday night, the 250 g pouch from filler 2 has leaked at the left side seal at about 0.4 percent (14 of 3,600 checked) against a target of none at the case station. Filler 1 runs the same pouch and film without leaks. One pallet is on hold and the order ships on Friday.

The statement names no cause. It does hand the team two leads, the roll change and the left seal, and one side question from the who row: the hourly seal check did not catch it. Those go into the 5 Whys. The 5 Whys examples show how that next step is written up.

After 5W1H: which tool comes next

5W1H describes the problem. It does not find the cause. Once the statement is written, choose the tool for the next step by how much you already know.

5W1H and the tools used beside it

  • 5W1H

    The question it answers
    What is the problem, in facts?
    When to use it
    First, before anyone discusses a cause.

  • Is/is not analysis

    The question it answers
    Where does the problem appear, and where could it appear but does not?
    When to use it
    When a similar part, machine or shift runs without the problem.

  • 5 Whys

    The question it answers
    Why did it happen, along one chain of cause?
    When to use it
    When the facts already point to one likely line of cause.

  • Fishbone diagram

    The question it answers
    Which of many possible causes should we check?
    When to use it
    When the cause is not obvious and several areas are suspects.

  • 5W1H action plan

    The question it answers
    Who does what, where, by when and how?
    When to use it
    Last, once a cause is confirmed.
5W1H comes first because every other tool needs a problem that is already defined.
  • The root cause analysis guide explains how the tools fit together and when each one is enough.
  • The fishbone diagram guide takes a problem statement as its head. A statement written with 5W1H gives a much tighter diagram.
  • The 5 Whys template has a line for the problem statement at the top. Copy yours in before the first why.
  • On an A3, the 5W1H answers fill the background and current condition boxes.
  • Practical problem solving starts with the same step under the name clarify the problem.

If the answers are thin, the facts are still on the floor. Go to the station with a few gemba walk questions and come back to the sheet afterwards.

Continuous Improvement Software

Run 5W1H in LeanSuite

Add 5 Whys, 5W1H, 4M1D, fishbone, 5G, MTTR or failure-curve sections to any phase, or build your own in Lean Creator.

See Continuous Improvement SoftwareBest root cause analysis software

Go deeper

This page is the short version. These guides cover each part in full.

All our articles on this subject are in the root cause analysis and problem solving topic.

Free templates and tools

Key terms

Lean glossary

Doing this in LeanSuite

See it on your own lines

Bring the problem you are working on to a 45-minute demo. Or look around with sample data first, or pilot LeanSuite on your own lines for 90 days, with the setup done by our team.

Book a 45-minute demoClick through the interactive demoStart a 90-day pilot

FAQ

Questions about 5W1H questions

  1. What, where, when, who, why and how. In problem solving they are asked before any cause is discussed: what is the object and the defect, where is it found, when did it start, who found it and who is affected, why does it matter, and how was it detected and how many. The answers are combined into one problem statement.

  2. 5W1H describes the problem and 5 Whys finds its cause. 5W1H asks six different questions once each and ends with a problem statement. 5 Whys asks the same question, why, again and again along one chain until it reaches a cause you can act on. Use 5W1H first and take its statement into the 5 Whys.

  3. 5W2H adds a seventh question, how much. It asks for the size of the problem as a count or a cost when it is used to define a problem, and for the cost of the action when it is used for a plan. In 5W1H the same point is covered under how.

  4. No. When 5W1H is used to define a problem, why means why it is a problem: the effect on safety, the customer, cost or delivery. The root cause is found afterwards with a tool such as 5 Whys or a fishbone diagram. Writing a cause into the description stops the team from testing other causes.

  5. Nobody in particular. Rudyard Kipling made the six questions famous in a poem in Just So Stories in 1902, which is why some people call it the Kipling method, but the same questions were taught by writers on rhetoric many centuries earlier. Manufacturing adopted them later for defining problems and writing action plans.

Related guides and tools

More method guides

Or browse the topics, the full library and the lean glossary.

Rather see it on your own problem?

Bring the problem you are stuck on, and we will show you on a call how LeanSuite handles it.

Schedule Demo

Pass it on

We want LeanSuite to be the best place on the internet for lean help. If this was useful, send it to someone on your team or in your network who needs it.