google_analytics.property_traffic
Property-wide daily traffic for the CanadaLogin GA4 property, with no dimension beyond the day.
Each row represents one day of aggregate site-wide traffic.
Keywords: Google Analytics, GA4, traffic, sessions, active users, page views
Provenance
Pulled daily from the GA4 Data API (runReport) for the CanadaLogin property, via a Lambda authenticating through Google Cloud Workload Identity Federation (no stored service account key). Written as JSONL, then cast to typed Parquet by a Glue ETL job.
- Updated: Daily, at 07:00 ET
- Retention: 7 years
- Steward: EDCP Data and Research Team
- Contact: Colin Douglas
- Location:
s3://<transformed-bucket>/google_analytics/property_traffic/
Fields
One row per date.
| Field Name | Type | Description |
|---|---|---|
| date | string | The reporting day, as a partition key (for example, “2026-07-23”) |
| sessions | bigint | Number of sessions started on this day, site-wide |
| activeusers | bigint | Number of distinct active users on this day, site-wide |
| newusers | bigint | Number of users on this day who had not been seen before |
| screenpageviews | bigint | Number of page views on this day, site-wide |
Notes
activeusers is de-duplicated by GA4 within the day; it is not the sum of per-page unique users. This table is the collection canary for the GA4 pipeline: because it carries no dimension beyond date, a property-wide active-user collapse here distinguishes “GA stopped collecting” from “one page or one funnel’s tag broke,” which page_traffic rows alone cannot do.
screenpageviews here should equal SUM(screenpageviews) from page_traffic for the same date - the two tables’ only currently-summable metric, and a useful integrity check between them. sessions and activeusers do not reconcile the same way across the two tables (GA4 de-duplicates each within its own dimension combination), so do not expect them to match.