Audit Feature Adoption Tracking Across User Workflows

Get a complete analytics audit mapping exposure, first-run completion, and habitual milestones to spot blind spots in your feature tracking.

You getFunnel Health Diagnostic, instrumentation Audit Table and 3 Diagnostic Questions

Prompt

Prompt
Act as a principal product analytics architect. I want to audit telemetry for an existing product workflow to detect tracking gaps across user adoption stages.

First, ask me 3 brief questions and wait for my reply before creating the audit:
1. What is the feature, its core value event, and who uses it?
2. What user actions are currently tracked as analytics events?
3. What defines regular, habitual use (for example: twice weekly, monthly billing run)?

Once I reply, generate an instrumentation audit following these rules:
- Break the funnel into three strict tiers: Exposure (discovery/entry points), First Complete Cycle (initial time-to-value), and Habitual Retention (repeated usage).
- Compare currently tracked events against critical missing events required to diagnose user drop-off.
- Name every suggested event with snake_case naming (object_action format) and include 2-3 required payload properties.

Output format:
- Funnel Health Diagnostic: 2-sentence summary of tracking maturity and main blind spot.
- Instrumentation Audit Table: Columns for Stage, Milestone Goal, Currently Tracked, Missing Recommended Events, and Essential Event Properties.
- 3 Diagnostic Questions: Questions product managers can answer only after implementing the missing events.
- Implementation Risks: One short note on telemetry load or event volume risks.

Example output

Example for: an automated invoice matching feature in a B2B accounting platform.

Funnel Health Diagnostic

Your current instrumentation tracks overall page views and batch approvals but completely obscures whether users manually override machine suggestions. This creates a blind spot where you cannot determine whether drop-off stems from low discoverability or distrust in automated match accuracy.

Instrumentation Audit Table
StageMilestone GoalCurrently TrackedMissing Recommended EventsEssential Event Properties
ExposureUser discovers matching interfaceinvoice_page_viewedmatch_banner_impression, match_entry_clickedsurface_location, pending_invoice_count, user_role
First Complete CycleUser runs and confirms first automated matchbatch_confirmedmatch_rule_adjusted, match_exception_flagged, first_batch_reconciledmatch_confidence_score, manual_override_count, elapsed_seconds
Habitual RetentionUser reaches automated weekly reconciliationbatch_confirmedweekly_match_summary_viewed, automated_sync_toggledrecurrence_cadence, invoices_auto_cleared, match_accuracy_rate
3 Diagnostic Questions
  1. What percentage of users who see match_banner_impression click through to execute match_entry_clicked within the first 48 hours?
  2. Are users abandoning between match_exception_flagged and first_batch_reconciled due to high manual override friction?
  3. Does enabling automated_sync_toggled correlate with a 30% reduction in weekly reconciliation time?
Implementation Risks

Tracking match_banner_impression on high-traffic inbox views could multiply client-side event volumes tenfold without client-side debouncing or session deduplication.

Written by AI from this prompt. Your results will vary.

Tips and common edits

Why this works

  • Forces a strict 3-tier structure to uncover hidden gaps between first discovery and sustained weekly usage.
  • Mandates snake_case event names and payload properties so results can be pasted straight into engineering specs.
  • Includes diagnostic questions to verify that the recommended data actually answers core retention problems.

Paste this prompt into ChatGPT, Claude, or Gemini when preparing analytics specs or diagnosing why an existing feature shows low retention numbers.

Common edits

  • Add a requirement to map existing events to specific platforms like Mixpanel, Amplitude, or PostHog.
  • Request server-side versus client-side tagging recommendations for each missing event row.

Make it repeatable

Keep using this prompt without pasting it in every time.

Keep exploring

More in Product Management