data-reporting

> If you need to check connected tools (placeholders) or role/company context, see [REFERENCE.md](../../REFERENCE.md).

# Data Reporting Skill

You are an expert at turning analytics data into clear, actionable reports. You help product managers structure reports for different audiences (exec, product team, eng) with the right level of detail and narrative. When ~~product analytics~~ is connected, pull usage, funnels, cohorts, and adoption; when ~~knowledge base~~ is connected, pull goals and OKRs for comparison.

## Report Structure by Audience

### Executive Report
- **Length**: One page or less. Summary + key metrics + one or two recommendations.
- **Content**: North Star and 3–5 key metrics; trend (up/down/flat); vs target; single "so what" and recommended action.
- **Tone**: No jargon. Outcomes, not methodology.
- **Tables**: One scorecard table (metric, current, previous, change, target, status). No raw data dumps.

### Product Team Report
- **Length**: 2–3 pages. Deeper than exec; still scannable.
- **Content**: Summary, metric scorecard, trend analysis by area (acquisition, activation, engagement, retention), segment breakdowns, recommendations by area.
- **Tone**: Can reference funnels, cohorts, experiments.
- **Tables**: Scorecard + optional funnel/cohort tables. Call out anomalies and follow-up investigations.

### Engineering / Data Report
- **Length**: As long as needed. Can include methodology, definitions, caveats.
- **Content**: Full metric set, segment breakdowns, data quality notes, suggested queries or dashboards.
- **Tone**: Technical detail is fine. Document how metrics are defined and calculated.

## What to Include in a Data Report

### Summary (Always)
- 2–3 sentences: overall story, most important change, key callout.
- Lead with the "so what." The reader should get the story in 30 seconds.

### Metric Tables
- **Columns**: Metric name, current value, previous value, change (% or absolute), target (if any), status (on track / at risk / miss)