WHY 5-WHY BEATS BRAINSTORMING
When something goes wrong in a plant, the default reaction is to fix the thing that is broken and move on. The press broke, maintenance replaced the part, the line is running again. Three weeks later the same press breaks the same way. Different part this time, same underlying problem.
Brainstorming sessions do not solve this. A room full of people generating ideas for why something happened gives you a list. Some of the list is useful. Most of it is noise. None of it is verified.
5-why analysis manufacturing works because it forces you to keep digging until you hit something you can actually change. Each "why" has to be supported by evidence you can see, touch, or measure. You do not stop at an opinion. You stop when you have found a cause whose removal would have prevented the problem.
The discipline is almost mechanical. Start with the problem statement. Ask why it happened. Get an answer supported by evidence. Ask why that happened. Get another answer. Keep going until you can state a fix that would prevent the problem from occurring again, not just in this specific case, but in the general class of cases. Five is a guideline, not a rule. Some problems resolve in three whys. Some take seven.
WHY MOST 5-WHY SESSIONS FAIL
Four failure modes show up in almost every weak structured root cause manufacturing program.
The session stops at "operator made a mistake." That is a symptom, not a root cause. The next operator on shift will hit the same gap.
The session is run in a conference room. Managers and engineers run the analysis without the people who actually saw the problem on the floor. The conclusions miss the operational detail that would have changed the answer.
The session is run on a problem too vague to solve. "Quality is bad" is not a 5-Why problem. The session produces frustration and no fix.
The session produces a finding but no follow-through. Root cause identified, fix proposed, owner not named, due date not set. Three weeks later the problem recurs.
Each of these compounds. A plant that runs five 5-Whys a month, all of them in conference rooms ending at operator error, produces zero durable improvement. A plant that runs two 5-Whys a month, both on the floor with evidence, ending at process gaps with named owners, sees recurring problems decline.
The discipline is not about the number five. It is about evidence and follow-through. We have written more about how 5-Why fits the broader problem-solving toolkit in our 5-Why glossary entry, which covers the basic structure. This post covers the rigor that separates a real session from a checked box.
A WORKED EXAMPLE, END TO END
Here is a 5 why example manufacturing teams can use as a template.
Problem statement: We scrapped 14 parts on the April 3 run of P/N 10234 due to out-of-tolerance OD on the finishing pass.
Why 1: Why did we scrap 14 parts? Because the OD on the finishing pass was 0.025 inch undersize on all 14. (Evidence: scrap log, all 14 parts measured with dial caliper at the same dimension.)
Why 2: Why was the OD undersize? Because the CNC offset was loaded incorrectly at setup. (Evidence: machine offset log, current offset value matches the previous job, not the current job.)
Why 3: Why was the offset loaded incorrectly? Because the setup operator used the setup sheet from the previous job and did not verify against the current job traveler. (Evidence: operator interview, setup sheet on the machine when the job started was from P/N 10089.)
Why 4: Why did the operator use the previous setup sheet? Because the current traveler was not at the machine at startup. The previous sheet was still on the machine from the last run. (Evidence: traveler tracking system shows the traveler was last scanned at the kitting station the night before.)
Why 5: Why was the traveler not at the machine at startup? Because the scheduling system does not require the traveler to be physically delivered with the material, and there is no check at setup to verify the traveler matches the material. (Evidence: process map, no enforcement step exists.)
The root cause is a missing process step: no verification that the traveler matches the material at setup. The fix is to add that step, either as a paper signoff or a scan at setup.
Notice what the root cause is not. It is not "the operator made a mistake." That is the symptom of the process gap. Blaming the operator leads to retraining, which does not prevent the next operator from making the same mistake. The fix has to change the process so the mistake cannot happen, regardless of who is on shift.
RUNNING A 5-WHY SESSION THAT WORKS
A 5-Why session is short. Thirty to sixty minutes if you are prepared. Longer means you are debating opinions, not pursuing facts.
Four elements make the session work.
Go to the place. Run the session on the floor where the problem occurred, not in a conference room. You will need to look at the machine, check the setup, pull the traveler, examine the part.
Get the right people. The operator who was running the job. The setup person. The supervisor. The quality tech who caught the defect. Maybe the scheduler or materials person depending on the problem. Five people is the right ceiling.
Bring the evidence. The scrap parts. The traveler. The setup sheet. The machine's offset record. If the evidence does not exist, that is your first finding.
Write everything down. The worksheet has space for each "why" and the evidence that supports it. Do not trust memory.
The facilitator's job is to keep the conversation on the current "why" and ask for evidence on every answer. "How do we know that?" is the most important question in the room. If the answer is "I think so" or "someone told me," the answer is not yet supported.
The facilitator also has to manage the blame reflex. When a 5-Why starts heading toward "the operator screwed up," the facilitator stops and asks "why was the operator in a position to screw up?" That redirects toward process and away from people.
THE SEVEN TRAPS TO AVOID
Seven traps in rough order of frequency.
Stopping at the person. "The operator did not follow the procedure" is a symptom. Push further. Why was the operator in a position to deviate without being caught.
Jumping to a solution. Someone proposes a fix after the second "why" and the group latches on. The session ends, the symptom gets addressed, the underlying gap stays. Discipline: do not discuss fixes until the "whys" are exhausted.
Going too wide. "Quality is bad" is not a problem statement. You can run fifty 5-Whys on it and none will produce a fix. Narrow the scope to a single event before you start.
Using opinions instead of evidence. Someone says "we scrap a lot on that machine because it is old." Is it actually old? What is the failure rate? Every "why" answer needs a fact behind it.
Running in circles. Why 3 answers Why 2, but Why 2 is the same as Why 1 phrased differently. You are not progressing. Stop and reframe.
Managers dominating the session. The plant manager holds the strongest opinion in the room and the operators defer. You lose the most important information. Set a rule: the operator speaks first on every "why."
No follow-through. The session produces a root cause and a fix. Nobody assigns the fix with a date. Three weeks later it has not happened and the problem recurs.
WHEN 5-WHY IS THE WRONG TOOL
5-Why is best for problems where there is a clear chain of cause and effect. A specific defect, a specific machine failure, a specific shipping error.
It is the wrong tool for multi-factor problems with no dominant cause. Customer complaints driven by 20 different issues across 5 departments are a scorecard problem, not a 5-Why problem. It is the wrong tool for statistical problems. "Our scrap rate is 4.2 percent and our target is 2 percent" is a Pareto and capability study, not a 5-Why. It is the wrong tool for strategic problems. "Why is our EBITDA below target" is a business review.
When you find yourself running a 5-Why on a problem that does not resolve into a clear causal chain, stop. Switch tools. A fishbone diagram, a control chart, or a process map will serve you better. The Sharpen implementation guide library covers structured problem solving in detail, including A3, fishbone, and PDCA methods.
DOCUMENTING AND FOLLOWING THROUGH
A working 5-Why closes with a one-page worksheet capturing problem statement, date and location, attendees, the "why" chain with evidence at each step, the identified root cause, the proposed fix, the owner, the due date, and the verification method.
The form goes into a central log, not a desk drawer. The log gets reviewed at the monthly quality or operations meeting. Any 5-Why that has not closed its fix in 30 days gets escalated.
Verification is the step most often skipped. After the fix is implemented, measure the problem again. If the scrap rate on that part dropped, or the defect disappeared, the fix worked. If not, the 5-Why was wrong or incomplete. Go back and run another one. A plant that closes 100 percent of its 5-Whys with measured verification develops a culture where root cause work is trusted. A plant that runs 5-Whys without verification develops a culture where root cause work is theater.
A SECOND WORKED EXAMPLE: A NON-QUALITY 5-WHY
5-Why analysis manufacturing is not just a quality tool. Here is a delivery example.
Problem statement: P/N 11023 shipped two days late on March 18, customer escalation to the GM.
Why 1: Why did it ship late? Because the order was completed two days after the promise date. (Evidence: order completion timestamp vs. promise date in the ERP.)
Why 2: Why was it completed late? Because the part started production three days late. (Evidence: production start date in the ERP.)
Why 3: Why did production start three days late? Because the kit was incomplete when the order was scheduled to start. (Evidence: kitting log, two of the seven components were not on the kit cart at scheduled start.)
Why 4: Why was the kit incomplete? Because two purchased parts were not in inventory when the order was released to the floor. (Evidence: inventory transaction log.)
Why 5: Why were the purchased parts not in inventory? Because the buyer used a 21-day supplier lead time in the planning system, but the actual lead time on those two part numbers averages 32 days. (Evidence: PO history for the two part numbers, last 12 orders.)
The root cause is a stale lead time in the planning system. The fix is a quarterly review of supplier lead times against actual PO history, with adjustments for any variance over 15 percent. Notice the root cause is not "the buyer made a mistake." The buyer used the lead time the system gave them. The gap is the absence of a process to keep lead times current.
HOW TO BUILD THE 5-WHY HABIT IN 60 DAYS
Start with one 5-Why per week, sponsored by a leader. Pick the biggest quality, safety, or delivery issue from the previous week and run the session on Friday morning. The first few sessions will be uneven. The discipline takes practice.
After a quarter, expand. Each supervisor runs a 5-Why when they have a problem in their area. Some problems get closed in 20 minutes at a workbench. Others escalate to formal sessions. Track the count and the impact: "This quarter, 22 5-Whys were run. Estimated cost impact from implemented fixes: $84K." Leadership pays attention when the technique produces measurable results.
Watch for the regression. After a year, the sessions will start to feel rote. People will shortcut the evidence-gathering. The facilitator's job is to hold the standard. No "why" gets past without evidence. The discipline is the whole thing.
WHAT TO DO NEXT
A plant that runs 5-Why rigorously will see its recurring problem list shrink every quarter. The problems that return are a sign the method is slipping. Fix the method, the problems go away again.
For the full implementation playbook with the worksheet, the seven traps, and the rollout timeline, the Sharpen 5-Why root cause analysis guide walks through it section by section. The 5-Why pairs naturally with the scrap tracking program (run a 5-Why on the top reason code each week) and with the daily production meeting (every problem surfaced at the meeting routes to a 5-Why or a regroup, never a discussion).
For a structured assessment of where your plant stands across all 10 pillars, the free Sharpen diagnostic at the intake takes about 10 minutes and produces a prioritized roadmap. It will tell you whether problem solving is your binding constraint or whether something upstream needs to be fixed first.