Insights

Build ad-hoc analytics queries without saving to a dashboard. Insights is the most powerful page in Pug — use it to answer one-off questions and prototype charts before pinning them to a dashboard.

Path: /p/:projectId/insights

Query builder walkthrough

1. Select events

Choose one or more events from your project’s catalog. Multi-event queries compare series on the same chart or build funnels.


Single event:   page_view          → daily active users line chart
Multi event:    page_view, click   → compare two series
Funnel:         signup_started → signup_completed → purchase

Events only appear in the picker after they’ve been received at least once. Send test events via the SDK if your catalog is empty.

2. Apply filters

Filters narrow which events are included in the query.

Filter scope	Applies to
Global filters	All selected events
Per-event filters	One specific event in a multi-event query

Filter on any property — custom traits, auto-properties, or event properties:


Global:   $geo.country = "US"
Per-event (purchase):  revenue > 100
Per-event (page_view): $url contains "/pricing"

Property types are inferred from your event schema. See Events to inspect available keys.

3. Choose visualization

Type	Best for
Line	Trends over time
Area	Volume trends with filled area
Bar	Discrete period comparisons
Funnel	Step-by-step conversion
Retention	Cohort return rates
Table	Raw breakdown rows

4. Set aggregation

Aggregation	Computes
Count	Total event occurrences
Unique users	Distinct profiles per bucket
Sum	Total of a numeric property (e.g. `revenue`)
Average	Mean of a numeric property

For revenue analysis, select Sum on the revenue property of purchase events.

5. Add breakdowns

Breakdowns split results by a property dimension:


Breakdown: $geo.country     → one series per country
Breakdown: plan              → one series per plan trait
Breakdown: $browser          → browser distribution

Combine breakdowns with any visualization. Funnel breakdowns show conversion rates per segment side-by-side.

Recipe: Daily active users


Events:       page_view
Range:        Last 30 days
Granularity:  Day
Aggregation:  Unique users
Visualization: Line


Events:       signup_started → signup_completed (in order)
Range:        Last 7 days
Visualization: Funnel
Conversion window: 7 days
Breakdown:    $utm_source

See the full Funnel analysis guide.

Recipe: Revenue by country


Events:       purchase
Range:        Last 30 days
Granularity:  Day
Aggregation:  Sum (property: revenue)
Breakdown:    $geo.country
Visualization: Bar

See E-commerce tracking.

Recipe: Day-1 retention


Visualization: Retention
Starting event: signup_completed
Return event:   page_view
Range:          Last 90 days
Granularity:    Week

See Retention analysis guide.

Time range and timezone

Insights uses the project timezone from Settings. All bucket boundaries align to that timezone — a “day” is midnight-to-midnight in project time, not UTC.

Override with a custom range for campaign-specific windows.

Export and reuse

Action	Result
Copy query spec	JSON insight spec for the Insights API
Add to dashboard	Creates an insight tile on a Dashboard
Share link	URL with query state (team members with project access)

The same query spec powers Overview KPIs, dashboard tiles, and API responses — build once, use everywhere.

Common issues

Symptom	Fix
Event not in picker	Send at least one instance via SDK; check Events catalog
Empty chart	Widen time range; remove filters; confirm events exist in range
Funnel shows 0% conversion	Increase conversion window; verify event order matches actual user flow
Breakdown has too many rows	Add a global filter to narrow the segment first