Agent skill · data analytics · borghei
product-analytics
Product analytics for instrumenting products, defining metrics, and building retention funnels. Use when designing a metric tree, instrumenting a feature, auditing instrumentation, defining a North Star, or building an analytics roadmap.
Why this skill is useful
Adds 3 executable scripts for auditing event taxonomy, designing metric trees, and analyzing retention cohorts that are not commonly found in public documentation.
What it needs
Requires python3 installed locally. About 5k tokens when loaded. Last updated 2026-07-21. 462 stars on the source repository.
What this skill does
Product Analytics A product analytics skill focused on decisions from data, not dashboards. Covers the metric tree, instrumentation patterns, funnel + retention + cohort analysis, and the operational rituals that turn measurement into product changes. When to use this skill Designing the North Star metric and its tree of input metrics Auditing product instrumentation (events, properties, gaps) Building or refreshing an activation funnel for a new product or feature Designing or analyzing retention cohorts (D1/D7/D30/W1/W4/M1/M3) Building or refining the PM analytics dashboard Translating product data into decisions and roadmap inputs Auditing dashboards for actionability (kill the vanity) Inputs the advisor expects Product type (B2B SaaS, consumer, marketplace, etc.) Current analytics stack (Amplitude / Mixpanel / GA4 / Segment / Snowflake + dbt + Looker) Existing North Star + input metrics Current event taxonomy + instrumentation gaps Top product questions you can't answer today Org expectations: who consumes analytics, at what cadence Clarify First Before designing the metric tree or audit, confirm these inputs. If any is unknown or vague, ASK — do not assume: [ ] Product type — B2B SaaS, consumer, marketplace, etc. (drives the North Star pattern and input metrics) [ ] The value moment — what "delivered value" looks like for a user (defines the North Star and activation event) [ ] Current analytics stack and event taxonomy — Amplitude/Mixpanel/GA4/Segment plus existing events (drives the instrumentation audit and gap list) Stop rule: ask only the 2-3 that most change the output. If the user says "just draft it," proceed and list your assumptions at the top of the artifact. Workflows Workflow 1 — Design the metric tree 1. Define the North Star (one number that summarizes value delivered). 2. Decompose into inputs (drivers of the NS). 3. Add guardrails / counter-metrics that catch unintended consequences. 4. …
How to use it
Reference it in AdaL, Claude Code, Cursor or any coding agent — nothing to install:
@skills borghei/product-analytics