Data Quality Incident Root Cause Analysis Builder

You are a data reliability lead. Help investigate a data quality incident and produce a root cause analysis with prevention steps.

**Incident summary:** [what looked wrong]
**Metric/table/dashboard affected:** [name and link/description]
**Business impact:** [wrong report, decision risk, customer-facing error, revenue impact]
**Detection source:** [analyst, stakeholder, alert, data test, customer]
**Time window affected:** [start/end or unknown]
**Systems involved:** [source app, warehouse, ETL/ELT, BI tool, reverse ETL, spreadsheet]
**Recent changes:** [deploys, schema changes, tracking changes, backfills, vendor changes]
**Evidence available:** [queries, logs, screenshots, row counts, test failures]
**Known constraints:** [limited lineage, no owner, missing logs, urgent board report]
**Stakeholders:** [data, product, finance, marketing, executives]

Produce:
1. **Incident summary** — plain-English explanation of what happened and why it matters.
2. **Impact assessment** — affected metrics, audiences, reports, date range, confidence level, and decisions at risk.
3. **Investigation plan** — step-by-step checks from source event to final dashboard.
4. **Root cause hypotheses** — ranked causes with evidence needed to confirm or reject each.
5. **SQL/query checks** — sample validation queries or logic to compare counts, freshness, nulls, duplicates, joins, and filters.
6. **RCA narrative** — timeline, root cause, contributing factors, detection gap, and resolution.
7. **Remediation plan** — data fix, stakeholder communication, dashboard annotations, and backfill validation.
8. **Prevention plan** — tests, ownership, lineage, alert thresholds, release checks, and runbook updates.
9. **Stakeholder update draft** — concise message for non-technical stakeholders.

Separate confirmed facts from assumptions. Do not invent data that was not provided.