carlosverasteguii/analytics-tracking

Analytics Tracking & Measurement Strategy

Core Principles

1. Track for Decisions, Not Curiosity: Every event should help answer a question (e.g., "Do users abandon the receipt upload form?").
2. Naming Conventions (Segment/Rudderstack style):
   • Object Action (e.g., Receipt Uploaded, Budget Created).
   • Properties: snake_case (e.g., total_amount, merchant_name).
3. Data Quality Beats Volume: It's better to track 5 key events perfectly than 100 broken ones.

Recommended Tracking Plan for Daticket

Key Events

| Event Name | Trigger | Key Properties | | :--- | :--- | :--- | | User Signed Up | Successful registration | method (email/google), role | | Receipt Scanned | User completes scan | store_name, total_amount, item_count, ocr_confidence | | Receipt Verified | User confirms/edits OCR | edits_made (boolean) | | Budget Created | New budget set | category, amount, period | | Alert Triggered | Spending > Budget | category, percentage_over |

Measurement Readiness Checklist

• [ ] Event Definition: Are all events clearly defined in a Tracking Plan?
• [ ] Identity Management: Are we correctly identifying users across sessions?
• [ ] Validation: Have we verified that the tracked data matches the DB data?
• [ ] Privacy: Are we PII-compliant? (Avoid tracking passwords, full data in URLs).

Implementation Tips

• Use a wrapper function for tracking (e.g., trackEvent(name, props)) to allow swapping providers (Supabase Analytics, PostHog, GA4) easily.
• Log tracking events to console in Development mode for debugging.