Metrics Review
Run a structured weekly, monthly, or quarterly product metrics review that compares periods, checks targets, and ends with recommended actions.
Overview
Metrics Review is an agent skill for the Grow phase that reviews product metrics over a chosen period and surfaces trends with actionable recommendations.
Install
npx skills add https://github.com/anthropics/knowledge-work-plugins --skill metrics-reviewWhat is this skill?
- Slash-style workflow: /metrics-review with time period or metric focus
- Gathers metrics from connected product analytics or user-provided tables
- Asks for comparison baselines, targets, launches, outages, and seasonality
- Supports spike/drop investigation and full-suite reviews
- Outputs organized metrics with actionable insights (trend analysis)
- Multi-step workflow: gather metrics data, then organize metrics (with user prompts for period, focus, and events)
Adoption & trust: 1.5k installs on skills.sh; 19.6k GitHub stars; 3/3 security scanners passed (skills.sh audits).
What problem does it solve?
You have raw analytics or a messy export but no structured review that compares baselines and explains what to do next.
Who is it for?
Solo builders running regular scorecards after launches, campaigns, or when a KPI spikes or drops.
Skip if: Greenfield ideas with no users yet, or deep data-engineering pipeline work without product metrics to interpret.
When should I use this skill?
Running a weekly, monthly, or quarterly metrics review, investigating a spike or drop, or comparing performance to targets.
What do I get? / Deliverables
You leave with an organized metrics picture, context for anomalies, and concrete actions aligned to targets—not just a chart screenshot.
- Organized metrics summary for the selected period
- Trend notes and recommended actions tied to context events
Recommended Skills
Journey fit
How it compares
Structured review workflow—not a BI dashboard replacement or one-off SQL exploration skill.
Common Questions / FAQ
Who is metrics-review for?
Indie product owners who want agent-guided metrics reviews with comparisons and narrative insights, with or without a connected analytics tool.
When should I use metrics-review?
At Grow analytics cadence—weekly health checks, monthly planning, quarterly retros—or immediately when investigating a sudden metric change.
Is metrics-review safe to install?
It may read analytics via configured connectors; review the Security Audits panel on this page and only connect tools you trust with production data.
SKILL.md
READMESKILL.md - Metrics Review
# Metrics Review > If you see unfamiliar placeholders or need to check which tools are connected, see [CONNECTORS.md](../../CONNECTORS.md). Review and analyze product metrics, identify trends, and surface actionable insights. ## Usage ``` /metrics-review $ARGUMENTS ``` ## Workflow ### 1. Gather Metrics Data If **~~product analytics** is connected: - Pull key product metrics for the relevant time period - Get comparison data (previous period, same period last year, targets) - Pull segment breakdowns if available If no analytics tool is connected, ask the user to provide: - The metrics and their values (paste a table, screenshot, or describe) - Comparison data (previous period, targets) - Any context on recent changes (launches, incidents, seasonality) Ask the user: - What time period to review? (last week, last month, last quarter) - What metrics to focus on? Or should we review the full product metrics suite? - Are there specific targets or goals to compare against? - Any known events that might explain changes (launches, outages, marketing campaigns, seasonality)? ### 2. Organize the Metrics Structure the review using a metrics hierarchy: North Star metric at the top, L1 health indicators (acquisition, activation, engagement, retention, revenue, satisfaction), and L2 diagnostic metrics for drill-down. See **Product Metrics Hierarchy** below for full definitions. If the user has not defined their metrics hierarchy, help them identify their North Star and key L1 metrics before proceeding. ### 3. Analyze Trends For each key metric: - **Current value**: What is the metric today? - **Trend**: Up, down, or flat compared to previous period? Over what timeframe? - **vs Target**: How does it compare to the goal or target? - **Rate of change**: Is the trend accelerating or decelerating? - **Anomalies**: Any sudden changes, spikes, or drops? Identify correlations: - Do changes in one metric correlate with changes in another? - Are there leading indicators that predict lagging metric changes? - Do segment breakdowns reveal that an aggregate trend is driven by a specific cohort? ### 4. Generate the Review #### Summary 2-3 sentences: overall product health, most notable changes, key callout. #### Metric Scorecard Table format for quick scanning: | Metric | Current | Previous | Change | Target | Status | |--------|---------|----------|--------|--------|--------| | [Metric] | [Value] | [Value] | [+/- %] | [Target] | [On track / At risk / Miss] | #### Trend Analysis For each metric worth discussing: - What happened and how significant is the change - Why it likely happened (attribution based on known events, correlated metrics, segment analysis) - Whether this is a one-time event or a sustained trend #### Bright Spots What is going well: - Metrics beating targets - Positive trends to sustain - Segments or features showing strong performance #### Areas of Concern What needs attention: - Metrics missing targets or trending negatively - Early warning signals before they become problems - Metrics where we lack visibility or understanding #### Recommended Actions Specific next steps based on the analysis: - Investigations to run (dig deeper into a concerning trend) - Experiments to launch (test hypotheses about what could improve a metric) - Investments to make (double down on what is working) - Alerts to set (monitor a metric more closely) #### Context and Caveats - Known data quality issues - Events that affect comparability (outages, holidays, launches) - Metrics we should be tracking but are not yet ### 5. Follow Up After generating the review: -