root-cause-investigation（根本原因调查）

Root Cause Investigation

When to use

• A key metric dropped (or spiked) unexpectedly and the team needs an explanation
• Stakeholders are asking "why did X happen?" and need an evidence-based answer
• A metric change has been observed but the team is unsure whether it's noise or signal
• Preparing a post-mortem after an incident that affected business metrics
• A trend change happened weeks ago and needs retrospective investigation

Process

1. Validate the change — confirm the metric changed beyond normal variance using a z-score or simple comparison to the rolling average. If the change is within ±1.5 standard deviations, document it as within normal range and close. Use scripts/drilldown_analyzer.py --validate.
2. Establish a timeline — plot the metric over time to pinpoint when the change started. A sudden step change suggests a specific event; a gradual drift suggests a structural shift.
3. Decompose the metric — break the metric into its constituent parts (e.g., revenue = volume × price × mix). Determine which component is driving the change before drilling into dimensions.
4. Drill down systematically — compare the metric before vs. after the change across available dimensions (geography, platform, channel, product category, user segment). Sort by absolute contribution to identify the primary driver. Use scripts/drilldown_analyzer.py --drilldown. See references/rca_framework.md for the structured approach.
5. Test hypotheses — generate explicit hypotheses (volume drop, mix shift, per-unit quality change, data issue) and accept or reject each with evidence. Correlate the timeline with known events from references/hypothesis_testing_guide.md.
6. Write the root cause report — document the primary driver (quantified share of impact), supporting evidence, rejected hypotheses, and tiered recommendations (immediate / short-term / long-term). Use assets/rca_report_template.md.

Inputs the skill needs

• Metric name and historical values (at least 30 days before the change)
• Granular data with dimensional breakdowns (geography, platform, segment, etc.)
• The date or date range when the change was noticed
• A change log or incident log for the same period (product releases, campaigns, outages)
• The business context: what decisions depend on this metric

Output

• scripts/drilldown_analyzer.py — validates the change, computes dimensional drill-downs, and ranks contributors by impact
• references/rca_framework.md — structured five-step RCA method with decision rules
• references/hypothesis_testing_guide.md — checklist of common root causes and how to test each
• assets/rca_report_template.md — report template: what changed, when, primary driver, supporting evidence, timeline, recommendations