Working Sheet · sent by email, same day
A tracking plan your engineers will actually follow
The exact format we hand to clients in implementation sprints - columns, conventions, and the process rules that keep it alive after launch.
Most tracking plan templates fail the same way: they document events without documenting decisions. This one forces the decisions. Every row must name its owner and the product question it answers - events that can't fill those columns don't get instrumented, which is the entire anti-bloat mechanism.
The columns, and why each exists
| Item | Detail |
|---|---|
| Event name | In your one enforced convention (we default to Object-Action Title Case: 'Report Exported'). The template validates casing. |
| Trigger | Precise firing condition - 'server confirms signup', never 'user signs up'. Ambiguous triggers are where duplicate events are born. |
| Properties (name / type / allowed values) | Typed, with controlled value lists. Free-text property values are future junk. |
| Platforms | Web / iOS / Android / server - so cross-platform gaps are visible before launch, not in a chart. |
| Owner | A person, not a team. Unowned events are unqueried events are wasted volume. |
| Question answered | The analysis this event serves. The column that keeps taxonomies small. |
| Status & version | Proposed / live / deprecated, with the release that changed it - the audit trail that makes drift diagnosable. |
The process rules matter as much as the columns: the plan lives where engineers work, every instrumentation PR links to its row, and a quarterly prune removes events with no queries. Adopt the columns without the process and you've made a prettier graveyard.
In practice: Want it filled in for your product? The implementation sprint's first deliverable is this document, completed.
Frequently asked questions
How many events should the plan start with?+
For one product surface: typically 20-50 well-chosen events. If your draft exceeds a hundred, the 'question answered' column is being skipped - that column is the diet.
Sheets or a data-management tool?+
Start in a spreadsheet - approval workflows and commenting matter most early. Graduate to Amplitude's data management once the taxonomy stabilizes; keep the spreadsheet as the proposal stage.
Get a free, scored audit of your Amplitude instance
Send us read-only access and get a scored findings report within 48 hours: taxonomy health, duplicate events, governance gaps, and the three fixes with the highest data-trust payoff. No commitment.
Request the free audit