five-whys

Focused 5-Whys Mode

You are running in focused-five-whys mode. Execute only the procedure below and produce only its canonical output sections — do not run the full 5-phase first-principles analysis. Skip Step 0 technique selection; the user has already chosen this technique by invoking the slash command directly.

When to reach for this

Use 5-Whys when a problem keeps coming back. The symptom is observable, the surface fix has been tried, and you need to find what is actually driving recurrence.

Good fit: the problem has a traceable causal chain; one or a small number of root causes are likely; corrective action is within your control.

Not a good fit: the problem involves multiple interacting subsystems with no single causal chain — that pattern calls for a fishbone (Ishikawa) diagram instead, which maps causes across categories in parallel.

---

Procedure

State the symptom. Write one sentence: the observable problem that keeps occurring. Do not state a suspected cause — state the observable effect.

Ask: Why did this happen? Write every cause you can identify. Do not filter yet. Multiple causes at the first level are expected.

For each cause, ask why again. At each level, ask "What else caused this?" before going deeper into any one branch. Complete the lateral scan at a level before descending. Multiple valid causes each become their own branch.

Stop drilling a branch when BOTH hold:

• You can state a specific corrective action that would prevent recurrence.
• That action is within your practical control.

If a branch reaches a cause with no actionable corrective — a systemic constraint outside your control — record it as a real finding and move to the next branch. A cause you cannot fix is still worth knowing.

Validate each causal link with observable evidence, not inference. If you cannot point to evidence for a link, flag it as assumed before continuing.

---

Example

Symptom: The bread keeps going stale before it is finished.

• Why? → The loaf is too large for one person to finish quickly.
  • Why? → Only standard-size loaves are bought.
    • Why else? — considered: the household eats less bread than it used to; rejected, weekly bread consumption has not changed — only the loaf size on offer has.
    • Why? → Half-loaves were not known to be available.
      • Stop: Corrective action — check the store for half-loaves or switch to a smaller format. Specific, in control, prevents recurrence. (Branch ends here.)
• Why (else at level 1)? → The bread is stored on the counter.
  • Why? → No one sealed the bag after the first slice.
    • Why else? — considered: the bag itself is defective and will not seal; rejected, the bag seals fine when the clip is used — the issue is that the clip is not at hand.
    • Why? → The bag clip is kept in a different drawer.
      • Stop: Corrective action — keep a clip at the bread storage point. Specific, in control, prevents recurrence. (Branch ends here.)

Both branches reach actionable corrective actions. The root causes are a purchasing habit and a storage habit — not the bread itself.

---

Failure modes

Stopping on a count, not a test. Asking "why" exactly five times and declaring done is the most common failure. Five is a typical depth, not a rule. Stop when the test is met, not when the count is reached.

Single-thread drilling. Moving straight down one causal chain without asking "What else caused this?" at each level. This misses parallel causes and produces an incomplete picture.

Inference without evidence. Writing "probably because X" and continuing down that branch. Every causal link needs observable evidence, or must be flagged as assumed.

Confirmation bias. Starting with a suspected cause and steering the chain toward it. The lateral scan at each level ("What else caused this?") is the check against this — it forces consideration of causes that contradict the hypothesis.

---

Handoff

The root cause(s) identified here are inputs to the 5-phase methodology. If you reached for this tool during Phase 2 (Challenge Assumptions), add each root cause as a challenged assumption row in the Classified Assumptions Table. If you reached for it during Phase 3 (Establish Ground Truths), promote each evidence-backed root cause to a ground truth — give it a stable GT-N identifier and a source citation, or the GT-N? suffix if its causal link is still assumed rather than verified. If you reached for it during Phase 4 (Reason Upward), add each cause as a validated step in the relevant Derivation Chain.

---

If a fuller analysis is needed afterward, invoke the main first-principles agent with this output as Known ground truths.