A complete workflow for setting up product analytics: define metrics, implement tracking, validate data quality, create dashboards, and establish monitoring rituals that drive decisions.
Identify what to measure before implementing anything
What are you trying to achieve? Growth, retention, monetization?
The one metric that best represents value delivered to customers
Awareness → Activation → Engagement → Retention → Revenue
What user actions indicate progress? Start with 10-20 critical events.
Naming convention: Object_Action (e.g., Report_Created, Dashboard_Viewed)
Start with activation metrics first. Most teams over-index on acquisition and under-index on "aha moments." Track what makes users successful, not just what gets them in the door.
I'm setting up product analytics for [PRODUCT DESCRIPTION]. Our North Star Metric is: [METRIC] Our business goal is: [GOAL] Help me define the critical events to track across this user journey: 1. Awareness (how they find us) 2. Activation (first value delivered) 3. Engagement (ongoing usage) 4. Retention (coming back) 5. Revenue (monetization) For each stage, suggest: - 3-5 key events to track - Event naming using Object_Action format - Properties to capture with each event - Why this event matters Focus on events that indicate user success, not just activity.
How to use: Replace bracketed items. Review AI suggestions and validate with team. Prioritize events that indicate value delivered.
Choose and configure your analytics platform and data pipeline
Set up separate projects for dev, staging, production
Add JavaScript SDK, mobile SDK, or server-side library to your codebase
Set up user ID, device ID, and anonymous tracking
Separate production data from test data with API keys or projects
Add tracking code for your defined events
Don't try to track everything at once. Roll out in phases:
// Example: Amplitude event tracking
amplitude.track('Report_Created', {
report_type: 'sales_dashboard',
template_used: true,
data_source: 'salesforce',
created_by_role: 'manager'
});
// Example: Mixpanel event tracking
mixpanel.track('Dashboard_Viewed', {
dashboard_id: '12345',
dashboard_name: 'Executive Summary',
viewed_by_role: 'admin',
load_time_ms: 234
});Object_Action format. Dashboard_Viewed, not "user viewed dashboard"
Who, what, where, when. User role, item ID, source page, timestamp.
Plan type, signup date, company size—helps with segmentation
Update your tracking plan with actual implementation details
Ensure tracking is working correctly before making decisions
Bad data is worse than no data. Teams lose trust in analytics when numbers don't make sense. Invest time here to prevent months of wasted effort.
Build views that answer your key questions
How many users reach their "aha moment"?
Are users getting value from the product?
Do users come back?
Your one metric that matters
I need to design an analytics dashboard for [DASHBOARD PURPOSE]. Our key questions are: 1. [QUESTION 1] 2. [QUESTION 2] 3. [QUESTION 3] Available events: [LIST YOUR EVENTS] Help me design this dashboard: - What charts/visualizations should I include? - What metrics should each chart show? - What segmentations are most valuable? - What time periods to display? Focus on actionable insights, not vanity metrics.
Make analytics a habit, not a project
Use analytics to apply Value Stream Mapping and Theory of Constraints.
Track time between stages to measure flow:
Find your constraint with data:
You've gone from zero to a complete analytics setup.
North Star Metric and key events defined and tracked
Activation, engagement, retention reports ready
Team reviews data regularly and makes informed decisions
1. Use analytics to validate features from Research to Backlog
2. Apply VSM to find workflow bottlenecks with data
3. Run experiments and measure impact