What's in the export?

One ZIP with eleven files in it - ten roll-up sheets plus the raw events.

The ten roll-ups, one CSV each:

• imported_visitors - daily totals: visitors, visits, pageviews, bounces, visit duration
• imported_sources - referrer, source, channel and the UTM fields
• imported_pages - page paths, with titles
• imported_entry_pages - where visits started
• imported_exit_pages - where they ended
• imported_locations - country, region, city
• imported_devices - device type and screen size
• imported_browsers - browser and version
• imported_operating_systems - OS and version
• imported_custom_events - your events and their properties

And the raw events. Every individual row, with its dimensions, not a summary of them. This is the file that makes the export worth something - a summary can’t be re-aggregated a different way, and raw rows can.

What’s not in there is an IP address, because there never was one. The visitor identifier is the daily hash, which means rows from different days can’t be joined into a person’s history. That’s the same limit that applies to us.

The sheet names aren’t arbitrary. They’re the same names the CSV importer reads, so an export imports straight back - into another site, another account, or the same one after a reset. Moving a site between teams is export, add, import.

Open them in anything. They’re plain CSVs with header rows.

The download link lasts 24 hours and you can make a new export whenever you like. How to create one - and note it works in every account state, including a locked or lapsed one.