This is the single most common misreading of User Activity, so it gets its own page.
| Identified Users | Anonymous | |
|---|---|---|
| A row is | One person | One browser profile |
| Same human, two devices | One row | Two rows |
| Has contact info | Yes | Never |
| Answers | Who is using the product | How much traffic is unidentified |
Because the two lists count different units, identified + anonymous is not your total audience and the ratio between them is not an identification rate. Compare each list against itself over time instead.
Windows, Not Totals
Both lists read a window of recent data rather than your whole history. Each source returns up to 200 identities per request, and the per-person facets — pages, sessions, device, country — are computed from that source's most recent 5,000 events in the selected range.
Over a long range on a busy site, older rows will have no event data attached. Those rows cannot be classified by engagement, pages, device or country. Shorter ranges are complete.
If a list looks oddly sparse, narrow the date range before assuming data is missing. A 7-day window on a busy site is complete where a 90-day window is a sample.