Data Analytics Skills in PM Skills Marketplace: SQL, Cohort Analysis, and A/B Testing

The PM Skills Marketplace includes three core data analytics skills—SQL Query Generation, Cohort & Retention Analysis, and A/B Test Statistical Analysis—accessible via CLI-style commands in the pm-data-analytics module.

The PM Skills Marketplace provides product managers with code-free analytics capabilities through its dedicated pm-data-analytics module. These data analytics skills transform raw data into actionable insights via natural language commands, eliminating the need to write complex SQL or Python from scratch. Each skill is implemented as a standalone capability with specific command signatures defined in the repository's source files.

Three Core Data Analytics Skills Explained

SQL Query Generation

The SQL Query Generation skill, defined in pm-data-analytics/skills/sql-queries/SKILL.md, converts natural language requests into production-ready SQL for BigQuery, PostgreSQL, MySQL, and Snowflake. Invoked via /pm-data-analytics:write-query, it generates optimized queries with performance suggestions and plain-language explanations of the logic.

Cohort & Retention Analysis

Located in pm-data-analytics/skills/cohort-analysis/SKILL.md, this skill analyzes user engagement cohorts, retention curves, and feature adoption trends. The /pm-data-analytics:analyze-cohorts command ingests CSV, Excel, or JSON files to compute retention rates, build heat maps, and export Python snippets for reproducible analysis.

A/B Test Statistical Analysis

The A/B testing skill, specified in pm-data-analytics/skills/ab-test-analysis/SKILL.md, performs rigorous statistical validation including significance testing, sample-size checks, and confidence-interval calculations. Using /pm-data-analytics:analyze-test, it accepts experiment exports and produces a decision matrix recommending whether to ship, extend, or stop a test.

Shared Workflow Architecture

All three data analytics skills follow a standardized four-step workflow implemented across the pm-data-analytics module:

1. Data ingestion – Upload a schema, CSV, Excel, or JSON file.
2. Contextual clarification – The tool queries for dialect specifics, time ranges, or metric definitions.
3. Automated generation – The skill produces SQL, Python, or visualization code based on the inputs.
4. Explanation & iteration – Output includes plain-language explanations and refinement prompts.

This workflow is documented in pm-data-analytics/README.md, which serves as the central reference for the module's command signatures and capabilities.

Practical Usage Examples

Generate BigQuery SQL for Daily Active Users

To generate a production-ready query for daily active users broken down by plan tier:

/pm-data-analytics:write-query Show me daily active users for the last 30 days, broken down by plan tier

According to the source code in pm-data-analytics/commands/write-query.md, this returns a BigQuery-compatible SELECT statement with COUNT(DISTINCT user_id), GROUP BY date, plan_tier, and annotated comments explaining each clause.

Analyze Cohort Retention from CSV

Upload a cohort engagement file and run retention analysis:

# Upload `cohort_engagement.csv`

/pm-data-analytics:analyze-cohorts

As implemented in pm-data-analytics/commands/analyze-cohorts.md, the output includes a retention heat map (cohort versus weeks), a Python script using pandas to compute retention percentages, and an insight summary highlighting early-churn cohorts.

Evaluate A/B Test Statistical Significance

Process experiment results and receive statistical recommendations:

# Upload `experiment_results.csv`

/pm-data-analytics:analyze-test

The command defined in pm-data-analytics/commands/analyze-test.md executes sample-size validation, Z-tests or chi-square tests, and returns p-values, lift metrics, 95% confidence intervals, and a decision table (Ship/Extend/Stop) based on significance and guardrail metrics.

Key Source Files and Implementation

The data analytics capabilities of PM Skills Marketplace are defined in the following repository paths:

• pm-data-analytics/README.md – Module overview and command reference
• pm-data-analytics/skills/sql-queries/SKILL.md – SQL generation specifications
• pm-data-analytics/skills/cohort-analysis/SKILL.md – Cohort analysis specifications
• pm-data-analytics/skills/ab-test-analysis/SKILL.md – A/B test specifications
• pm-data-analytics/commands/write-query.md – Command definition for SQL generation
• pm-data-analytics/commands/analyze-cohorts.md – Command definition for cohort analysis
• pm-data-analytics/commands/analyze-test.md – Command definition for A/B testing

Summary

• The PM Skills Marketplace ships three primary data analytics skills in the pm-data-analytics module: SQL Query Generation, Cohort & Retention Analysis, and A/B Test Statistical Analysis.
• Each skill is invoked via CLI-style commands (/pm-data-analytics:write-query, /pm-data-analytics:analyze-cohorts, /pm-data-analytics:analyze-test) and defined in dedicated SKILL.md files.
• Skills support BigQuery, PostgreSQL, MySQL, and Snowflake dialects for SQL; CSV, Excel, and JSON for cohort and A/B test analysis.
• All capabilities follow a four-step workflow: data ingestion, contextual clarification, automated generation, and explanation with iteration.
• Source code locations include pm-data-analytics/skills/ for skill definitions and pm-data-analytics/commands/ for command implementations.

Frequently Asked Questions

What specific databases does the SQL Query Generation skill support?

The SQL Query Generation skill supports BigQuery, PostgreSQL, MySQL, and Snowflake, as specified in pm-data-analytics/skills/sql-queries/SKILL.md. The tool automatically generates dialect-specific syntax and optimization suggestions based on the target database you specify during the contextual clarification step.

Can I export Python code from the cohort and A/B test analysis skills?

Yes. Both the /pm-data-analytics:analyze-cohorts and /pm-data-analytics:analyze-test commands generate ready-to-run Python scripts. According to the command definitions in pm-data-analytics/commands/analyze-cohorts.md and pm-data-analytics/commands/analyze-test.md, these scripts use pandas and statistical libraries to ensure reproducible analysis of your uploaded data.

How does the A/B Test Statistical Analysis skill determine whether to ship or stop an experiment?

The skill runs Z-tests or chi-square tests on the uploaded experiment data, calculates p-values, lift, and 95% confidence intervals, then produces a decision matrix. As implemented in pm-data-analytics/skills/ab-test-analysis/SKILL.md, the recommendation (Ship, Extend, or Stop) is based on statistical significance thresholds and guardrail metrics validation.

Where are the skill specifications documented in the repository?

Each data analytics skill has a dedicated specification file in the pm-data-analytics/skills/ directory: sql-queries/SKILL.md for SQL generation, cohort-analysis/SKILL.md for retention analysis, and ab-test-analysis/SKILL.md for A/B testing. Command implementations are stored in pm-data-analytics/commands/.

{
  "mcpServers": {
    "instagit": {
      "command": "npx",
      "args": ["-y", "instagit@latest"]
    }
  }
}