5-Why Analysis

Overview

The 5-Why method (also "5 Whys") is a simple but powerful technique for root-cause analysis. Starting from a concrete problem, you repeatedly ask "why?" — typically five times — until you reach a fundamental cause whose removal solves the problem sustainably.

Problem statement: The starting point: a concrete, observed problem or deviation. The more precisely it is phrased, the better the following "why?" questions work.

"Why?" chain: Every answer is followed by another "why?". Each level moves one step closer to the root. The number 5 is a rule of thumb — some problems resolve after 3, others need 7 or more.

Root cause: The last answer beyond which "why?" no longer yields a technically or organizationally actionable answer. This is where the improvement action attaches.

Branches: Sometimes a question has several plausible answers. The chain then branches into parallel strands — each may lead to its own root cause.

5-Why is cheap, fast, and team-friendly. It suits clearly bounded individual incidents. For complex, multi-factor problems, Ishikawa and FMEA are better tools.

Approach

• Describe the problem concretely — what happened, where, when, how often?
• Ask "why?" and enter the most plausible answer.
• Apply "why?" to the answer — and so on.
• If several answers fit, branch — each strand is followed individually.
• Stop as soon as a technically or organizationally actionable root is reached.
• Derive the action at the root — not at a symptom on a higher level.
• Verify the effectiveness of the action with data after implementation.

Pitfalls

Blame instead of cause: Answers like "the operator was not paying attention" are not root causes but blame. Keep asking: why could inattention turn into a problem? Is it training, standard work, visual inspection?

Stopping too early: The first plausible answer is rarely the root. Stopping after "why 2" treats a symptom. Go at least until the answer describes something in the process or system.

Asking too long: Eventually you arrive at "because it costs so much" or "because that's how the industry is" — answers the team cannot change. Stop here and act on the previous level.

Only one strand: For complex problems, one chain rarely produces the full solution. When several answers are plausible, branch — otherwise important causes are missed.

Not verified with data: Answers in a why chain are hypotheses. Before — or at the latest after — the action, check with data whether the identified root was actually the driver.

Done alone: A why chain from one person reproduces that person's assumptions. In a team with the people who know the process, the result is more robust.

Examples

This module ships with the following example datasets — load any of them in the app with a single click.

Available in the following cycles

• DMAIC: Define
• DMADV: Define
• 8D: D4 — Root Cause Analysis