How to import last seen dates ?
Jira does not expose "last seen" dates to apps, so Advanced User Manager cannot read them on its own. You supply them by importing the user export that Atlassian Admin produces. After the import the dates are stored on each account and power the Activity & last seen screen and the last-seen columns on Jira users.
Before you start: you need Jira administrator permission to use Advanced User Manager, and access to admin.atlassian.com for your organization to produce the export. No API token or organization API key is needed for this import.
Step 1 — Export the user data from Atlassian Admin
Go to admin.atlassian.com and open your organization.
Open Directory → User management (older sites label this section Users).
Click Export users.
Select the option that includes product access / last-seen details, then click Download. Atlassian either downloads the CSV straight away or emails you a link, depending on the size of your directory.
Keep the file as it is. Do not open and re-save it with edited headers — the importer identifies the last-seen columns by their original Atlassian names, and expects the account ID in the first column.
The next three screenshots show Atlassian Admin, not this app. Atlassian updates its admin screens from time to time, so labels may differ slightly on your organization.
Step 2 — Import the file into Advanced User Manager
There are three ways into the same dialog. Use whichever you are closest to:
In the console sidebar, open Jira users, click the Add users button and choose Import last-seen data.
Or from Overview, use the Import last-seen CSV quick action (or Import → in Needs attention).
Or from Activity & last seen, click Import last-seen data at the top right.
In the Import last-seen data dialog, click Select CSV file and choose the CSV you downloaded in Step 1. The import starts as soon as the file is read — there is no separate confirmation step. A progress bar runs while rows are written, and a message tells you how many users were updated and how many rows were skipped, for example Last-seen dates imported for 58 users.
Re-importing is safe. Existing last-seen values are updated in place, not duplicated, so you can refresh the data on a schedule by exporting and importing again.
Which columns are matched
The importer looks at the header row and picks out the Last seen in … columns by product name. Atlassian appends the site name to each of these headers, for example Last seen in Jira - acme; the site suffix is ignored. Matching is case-insensitive, the columns may appear in any order, and any of them may be missing. The first column must hold the Atlassian account ID, as it does in the original export.
Header in the Atlassian export | Stored as |
|---|---|
| Jira |
| Service Mgmt |
| Confluence |
| Compass |
Other Last seen in … columns your site may contain — Jira Product Discovery, Jira Customer Service, Studio and so on — are ignored; the app tracks the four products above. If a cell lists several dates separated by commas, the last one is used.
"Never accessed" is ignored, not stored
Where a user has never opened a product, the export writes Never accessed (older exports write -). The importer deliberately does not treat that as a date: the cell is left empty rather than being recorded as an activity date. That keeps "never used Confluence" distinct from "used Confluence a long time ago" in the inactivity buckets.
A consequence: rows where every product says Never accessed carry nothing to write and are skipped. If that is true of the whole file, the import stops and tells you that every row says "Never accessed" — nothing to import.
If the file is refused
If the header row contains no recognisable Last seen in … column, the file is rejected before anything is written and the app says so: No "Last seen in …" column found. Export from Atlassian Admin → User management with last-seen data included.
Common causes:
The export was produced without the product access / last-seen option selected. Re-run Step 1 and include it.
The file was opened in a spreadsheet and saved with renamed or reordered headers. Use the original download.
A different CSV was picked — for example a bulk-create list of email addresses, or a jira-users.csv exported from this app. Only the Atlassian Admin export works here.
If the file is accepted but nothing lands, the message reads No rows imported - N rows were rejected. Check the account ids match this site. That normally means the account IDs in the export belong to a different site than the one you are importing into — export from the organization that owns this site.
What last-seen data unlocks
Once the import has run, two things become useful:
Activity & last seen (sidebar, under MANAGE) — coverage of how many users have last-seen data, counts for inactive 60+ days and active this week, a breakdown by how long users have been away (active this week, 8–30, 31–60, 61–90, 90+ days or never), per-product coverage for Jira, Confluence, Service Mgmt and Compass, and the least recently seen users. See Track User Activity.
Last-seen columns on Jira users — open Jira users, click Columns, and switch on any of Jira, Confluence, Service Mgmt or Compass under Last-seen columns. Jira is on by default; the other three are off until you switch them on. The Seen · Jira column is sortable, so you can order the table by longest absence.
The status control on Jira users defaults to Active, so already-deactivated accounts are hidden. Switch it to Inactive or All if a user you expected to see is missing.
To take the numbers out of the app, select the rows you are interested in and use Export in the floating selection bar. It exports the selected rows only.
Deactivating an account frees its licensed seat, but it only works for managed accounts — accounts in a domain your organization has verified and claimed — and needs an organization API key created without scopes. See Deactivate/Activate Managed-Accounts before you act on what you find here.