Skip to main content

Insights

Insights are saved queries over your events: pick an event, choose an aggregation, filter and group it, and save the result as a chart on your dashboard.

What an insight is​

An insight is defined by:

  • Event — a specific event name, or all events
  • Aggregation — event count or unique users
  • Filters — property conditions (e.g. plan = "pro")
  • Group by — a time interval (day, week, month) or a property value
  • Visualization — line, bar, pie, table, or a single number

Example: "Weekly signups on the pro plan" is user_signed_up, filtered to plan = "pro", grouped by week, drawn as a line chart.

Saved insights run on demand and can be pinned to your project dashboard as widgets.

How to instrument for insights​

Insights are only as good as the events underneath them:

  1. One name per action. Use a single, consistent event name for each action (purchase_completed, not sometimes purchase and sometimes buy_completed). Insights query by exact event name.
  2. Put dimensions in properties. If you'll want to break a metric down by something — plan, screen, product category — send it as a property rather than encoding it in the event name. feature_used with { feature: "dark_mode" } can be grouped by feature; dark_mode_used cannot.
  3. Use consistent property values. "pro" and "Pro" are different values in a breakdown.
// Good: one event, groupable by property
track('feature_used', { feature: 'export', plan: 'pro' });

// Hard to analyze: name explosion
track('export_feature_used_by_pro_user');

Ask AI​

On paid plans you can also ask questions in plain English ("how many users exported data last week?") and get a chart back — useful for one-off questions that don't warrant a saved insight.

Next steps​