Product Analytics

What this skill helps you do

A product analytics skill for defining KPIs, building metric dashboards, and running cohort or retention analysis across product stages. Covers metric framework selection (AARRR, North Star, HEART), stage-appropriate KPI definition from pre-PMF through mature, dashboard hierarchy design, and cohort retention curve interpretation. Includes a CLI utility for retention, cohort matrix, and funnel conversion analysis from CSV data. Designed for product managers who need deterministic metrics, not vanity numbers.

Install and get started

Copy the full instructions into your AI tool. Test one low-risk example before connecting real store data.

Original Skill instructions

You are a product analytics expert. Select framework: AARRR for growth loops, North Star for alignment, HEART for UX quality. Define stage KPIs — pre-PMF: activation and early retention; growth: acquisition efficiency and conversion velocity; mature: retention depth and revenue quality. Design three dashboard layers: Executive (5-7 metrics), Product Health (acquisition, activation, retention, engagement), Feature (adoption, depth, repeat usage). Run cohort analysis comparing retention curves across signup cohorts — identify inflection points at onboarding. Use metrics_calculator.py for retention, cohort matrix, funnel analysis. Every KPI needs target, threshold, owner, and decision rule. Never report single-point retention or averaged-across-segments metrics.

Useful tasks

• Metric framework selection and KPI definition for pre-PMF, growth, or mature products
• Cohort retention analysis comparing curve shapes across signup or feature-exposure cohorts
• Funnel conversion analysis identifying drop-off stages and journey friction points
• Dashboard hierarchy design with executive, health, and feature layers
• Feature adoption interpretation connecting metric movement to product changes

How to use it

• Always compare retention curves across cohorts, not isolated single-point snapshots
• Every KPI needs: target, threshold, owner, and 'if below X, then Y' decision rule
• Executive dashboards should show 5-7 metrics max — 30+ metrics is dashboard overload
• Segment by cohort, plan tier, channel, or geography — blended metrics hide segment differences
• Use period-over-period with same-period-last-year context to control for seasonality

More skills for this workflow

• Weekly Ecommerce Performance Analyst

  AI-powered weekly business review in 5 minutes

• Analytics Tracking

  Set up, audit, and debug GA4/GTM tracking with event taxonomy and conversion tracking.

• xlsx

  Create, edit, and analyze spreadsheets with formulas, formatting, and recalculation.