본문으로 건너뛰기
cccskills
무료GitHub에 공개

root-cause-analysis

Find the root cause of a problem, not just symptoms. Use when the user says "root cause", "why does this keep happening", "5 whys", "fishbone diagram", "cause analysis", "problem investigation", "defect analysis", "incident analysis", "A3 problem solving", "what's causing this", "ishikawa" - even if they don't explicitly say "root cause analysis".

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

설치 방법 보기

포함된 파일(2)

  • SKILL.md5.6 KB
  • references/fishbone-categories.md4.2 KB

SKILL.md(원문)

설치하기 전에 에이전트에게 전달되는 지시 내용을 확인할 수 있습니다.

Reference Files

  • references/fishbone-categories.md - Six fishbone categories adapted for BA context (People, Process, Technology, Policy, Environment, Data) with example causes for each. Read this in Step 4 when building the Ishikawa diagram.

Overview

Based on Business Process Change by Paul Harmon - integrating Six Sigma DMAIC root cause methodology into business process improvement. Also draws on Lean Business Analysis by Mark Sherrington for the A3 Problem Solving technique - a one-page structured format that forces clarity by constraining the entire analysis to a single sheet. The key insight: most "root causes" identified in practice are actually symptoms. True root cause analysis requires asking "why" until you reach a systemic cause you can change - something in the process, policy, or system design, never an individual person.

Workflow

Step 1: Write the problem statement

Be specific and measurable. A good problem statement answers: what, where, when, and how much.

BAD:  "The system is slow."
GOOD: "Order processing takes 4.2 hours on average, exceeding the 2-hour SLA.
       This has occurred for 67% of orders in the past 30 days, primarily
       affecting [region/segment]. Impact: $[X] in late delivery penalties."

Step 2: Gather data

Before hypothesizing causes, collect evidence:

  • Timeline of when the problem started and any changes that preceded it
  • Frequency and pattern (constant, intermittent, time-based, load-based)
  • Scope of impact (which users, processes, or systems are affected)
  • Prior attempts to fix and their outcomes

Step 3: Run 5 Whys analysis

Start with the problem statement and ask "why" iteratively:

Problem: Order processing exceeds 2-hour SLA
Why 1?  Manual validation step takes 90 minutes
Why 2?  Validator must cross-reference 3 systems manually
Why 3?  Systems don't share a common order ID
Why 4?  Each system was built independently without integration design
Why 5?  No enterprise data architecture was defined when systems were procured
ROOT CAUSE: Absence of enterprise data architecture standards for procurement

Rules for 5 Whys:

  • Branch when multiple causes exist at a level
  • Stop when you reach a cause that is systemic and actionable
  • If a "why" answer is a person's name, go deeper - root causes are systemic

Step 4: Build a Fishbone (Ishikawa) diagram

Use 6 BA-specific categories (see references/fishbone-categories.md):

  • People: skills, capacity, roles, communication
  • Process: steps, handoffs, bottlenecks, documentation
  • Technology: systems, integrations, performance, functionality
  • Policy: rules, approvals, compliance, SLAs
  • Environment: organizational context, budget, competing priorities
  • Data: quality, availability, accuracy, access

Place the problem at the head. Add potential causes along each bone. Circle the causes that appear across multiple categories - these are systemic.

Step 5: Apply Pareto analysis

Rank causes by frequency or impact. The 80/20 rule: typically 20% of causes drive 80% of the problem. Focus countermeasures on the vital few, not the trivial many.

Step 6: Validate the root cause

Before declaring a root cause, test it:

  • Can you reproduce the problem by triggering this cause?
  • Does removing this cause prevent recurrence?
  • Does the causal chain from root to symptom hold under scrutiny?
  • Is this cause actionable (you can change it)?

Step 7: Document using A3 format

| A3 Section | Content |
|------------|---------|
| Background | Why this problem matters |
| Current Condition | Problem statement with data |
| Goal | Target state after fix |
| Root Cause Analysis | 5 Whys + Fishbone summary |
| Countermeasures | Actions to address root cause |
| Implementation Plan | Who, what, when |
| Follow-up | How to verify the fix worked |
| Expected Results | Measurable improvement target |

Anti-Patterns

1. Stopping at symptoms Bad: "Users didn't follow the process." (Why didn't they? That's a symptom.) Good: "The process requires 12 steps across 3 systems with no integrated view, so users created workarounds."

2. Root cause is a person's name Bad: "John made an error." Good: "The process has no validation check at step 5, allowing data entry errors to propagate."

3. 5 Whys that go in circles Bad: Why? Because X. Why X? Because Y. Why Y? Because X again. Good: If you loop back, you've missed a branch. Split and explore both causal paths.

4. Fishbone with only technology causes Bad: All causes listed under "Technology" - ignoring process, people, and policy. Good: Populate at least 4 of the 6 categories before concluding.

5. No validation of the root cause Bad: "We think the root cause is X, so let's fix it." Good: Test the hypothesis - can you reproduce the problem? Does the fix address it?

Quality Checklist

  • Problem statement is specific, measurable, and time-bounded
  • Data gathered before hypothesizing (not after)
  • 5 Whys reaches a systemic cause (not an individual)
  • Fishbone covers at least 4 of 6 categories
  • Pareto analysis identifies the vital few causes
  • Root cause is validated (reproducible, removable, actionable)
  • Countermeasures address root cause, not symptoms
  • A3 document is complete with follow-up plan and expected results

리뷰

아직 리뷰가 없습니다. 사용해 본 소감을 남겨 주세요.

댓글

로그인 없이 쓸 수 있습니다. 사용 후기, 질문, 작동하지 않은 환경 등을 편하게 남겨 주세요.

아직 댓글이 없습니다. 첫 댓글을 남겨 보세요.

보내면 바로 공개됩니다. 이메일 주소나 전화번호 같은 개인 정보는 쓰지 마세요. URL은 2개까지입니다.

0 / 1,000자

이 양식은 reCAPTCHA로 보호되며 Google 개인정보처리방침 및 서비스 약관이 적용됩니다.

같은 저장소의 스킬

Rate any product, feature, or experience on the 11-star scale (Brian Chesky's Airbnb thought experiment). Use when user says "rate this experience", "11-star", "star rating", "experience audit", "how good is this", "experience rating", "product audit", "quality assessment", or wants to evaluate product quality and identify improvement paths. Also trigger when user wants to benchmark a feature, assess where a product stands, or map out what "great" looks like - even if they don't explicitly say "11-star".

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

qa-aman/claude-skills212026년 10월 19일 업데이트

Generate A/B copy variants grounded in Eugene Schwartz's awareness and market-sophistication stages and John Caples' tested-headline discipline. Every variant declares its angle, the awareness stage it targets, and a falsifiable hypothesis, and the skill recommends which pair to test first and what delta counts as signal. Use when the user says "A/B test this", "write variants", "give me options", "test different angles", "copy variants", "ab copy", "test this headline", "multiple versions", "which version is better", or wants to test copy before publishing. For reviewing copy that already exists, see copy-review. For sample size and stopping rules, see growth-experiment. For diagnosing a page rather than its words, see page-cro. For full paid ad sets by platform, see ad-campaign-writer.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

qa-aman/claude-skills212026년 10월 19일 업데이트

Design statistically rigorous experiments to test data science hypotheses. Use when the user says "design an A/B test", "run an experiment", "test this hypothesis", "is this difference significant", "how many samples do I need", "randomized controlled trial", "causal inference", "uplift test", "holdout group", "significance test", "power calculation", "avoid p-hacking", or needs to determine whether an observed effect is real before making a product or model decision. For lean build-measure-learn product experiments without the statistics, use experiment-design instead.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

qa-aman/claude-skills212026년 10월 19일 업데이트

Write measurable acceptance criteria for requirements or user stories. Use when the user says "write acceptance criteria", "definition of done", "how do we test this", "fit criteria", "given when then", "Gherkin scenarios", "what does done look like", "testable conditions", "how will we know this works" - even if they don't explicitly say "acceptance criteria".

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

qa-aman/claude-skills212026년 10월 19일 업데이트

Build a strategic account plan for winning, retaining, or expanding a named account. Use when the user says "build an account plan", "strategic account planning", "how do I grow this account", "account expansion strategy", "whitespace analysis", "QBR prep", "key account review", "map the stakeholders in this account", "land and expand plan", "how do I break into [company]", or wants a structured strategy for a specific account.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

qa-aman/claude-skills212026년 10월 19일 업데이트

Write paid ad copy for LinkedIn, Google, Meta, and YouTube. Produces multiple variants per platform, each targeting a different Eugene Schwartz awareness level and Cialdini persuasion principle. Use when the user asks for ad copy, ad variants, "write LinkedIn ads", Google ads, Meta ads, paid social copy, ad headlines, or wants creative for a paid campaign. Reads brand voice, ICP, and positioning from knowledge/. For organic social posts, see linkedin-post or social-calendar. For the landing page the ad points at, see landing-page-writer. For testing variants, see ab-copy-writer.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

qa-aman/claude-skills212026년 10월 19일 업데이트

qa-aman의 스킬 모두 보기

이 스킬의 문제 신고