product-analytics

What This Skill Does

The product-analytics skill empowers the OpenClaw agent to serve as a high-level product strategist and data analyst. It enables the agent to translate raw product data into actionable insights, helping you define meaningful KPIs tailored to your current product lifecycle stage—whether you are searching for Product-Market Fit (PMF), scaling in the growth phase, or optimizing a mature product. The agent utilizes proven methodologies like AARRR (Pirate Metrics), the North Star framework, and Google’s HEART framework to structure your analytics approach. It excels at parsing complex cohort data, identifying retention inflection points, and designing metric hierarchies that bridge the gap between executive-level oversight and daily feature-level performance tracking.

Installation

To integrate this skill into your environment, run the following command in your terminal: clawhub install openclaw/skills/skills/alirezarezvani/product-analytics

Use Cases

• Defining a North Star metric that aligns your engineering, design, and marketing teams.
• Performing a comparative analysis of retention curves for different user signup cohorts to identify where onboarding friction exists.
• Structuring an executive-level dashboard that tracks key health metrics without drowning stakeholders in noise.
• Interpreting shifts in feature adoption rates following a major product release.
• Developing a measurement plan for a product moving from a growth phase to a mature, efficiency-focused phase.

Example Prompts

1. "Analyze our retention cohort data from the last three months; why is there a sharp drop-off in week 2, and what product changes should we prioritize to fix it?"
2. "We are a B2B SaaS in the growth stage. Help me define a KPI hierarchy that focuses on conversion velocity and expansion revenue."
3. "Can you set up a dashboard logic structure using the HEART framework to measure user experience for our new mobile dashboard feature?"

Tips & Limitations

• Tip: Always provide the agent with specific time windows (e.g., 'weekly vs. weekly') to ensure accurate trend comparisons.
• Tip: When interpreting data, ask the agent to distinguish between 'signal' and 'noise' to avoid reacting to transient data spikes.
• Limitation: The skill provides structural and interpretative analysis; it requires you to connect it to your specific data source APIs or provide raw data dumps for it to perform quantitative calculations.