Linear CSV export vs. automated backup
A Linear CSV export is useful when you need a spreadsheet or a portable record of issue data. A backup becomes useful when you need an earlier copy, the context around the issues, and a way to recover it. The deciding question is: if something goes wrong next month, what will you need back, and how will you put it back?
We build BackupLatte, a backup service for Linear. This guide compares Linear's built-in CSV exports with our scheduled backups and explains when a simpler export is enough.
What you get from a Linear export
Linear's export documentation describes a workspace issue CSV under Settings → Administration → Import Export. Admins can run it; on Enterprise, owners can. You can choose to include private teams. The emailed download link expires after 12 hours.
The CSV includes issue titles, descriptions, statuses, assignees, labels, dates and references to projects, parents and cycles. Attachment files are excluded, though descriptions may retain links. Comments are not among the listed CSV fields.
There are separate project and initiative CSV exports. Linear also lets you copy issues as Markdown, including comments, or save individual issues as PDFs. Its API and integrations provide other ways to move data. Those options have different coverage; “Linear export” does not mean just one format.
CSV and BackupLatte compared
This comparison is specifically about the built-in issue CSV and BackupLatte, rather than a promise about every backup product.
| Question | Built-in issue CSV | BackupLatte |
|---|---|---|
| What starts a copy? | A user runs the export | Daily schedule, plus one manual run per workspace every 24 hours |
| Does it include comments? | Not in the documented CSV fields | Yes, for accessible issues |
| Does it copy uploaded files? | No; links may remain in descriptions | Copies supported files linked from backed-up content; gaps are reported |
| Who keeps older copies? | Your team stores and manages exports | 7 days by default; configurable from 1 to 365 days |
| How do you recover data? | Read the file or use an import workflow | Download a snapshot or restore supported data into Linear |
| Does it undo edits to existing records? | No automatic rollback | No automatic rollback in the source workspace |
For BackupLatte's exact scope, see backup contents, scheduling and retention.
An export file is only part of a recovery plan
Being able to open a file and being able to resume work from it are different requirements. A spreadsheet may be enough to find an old title or compare estimates. Rebuilding an issue with its discussion, files and related records takes a different workflow.
Linear's CLI importer accepts its CSV format, but does not import comments or projects. Linear distinguishes it from dedicated import assistants, which can support more data. A CSV import should therefore not be treated as a complete workspace restore.
Before relying on any export, try recovering a small sample into a separate workspace. Check the result against the work you actually do: can someone follow the discussion, open the files and understand how the recovered issue relates to its project? Record what you had to repair manually.
Three situations that lead to different choices
You need a report for a planning meeting
Suppose you want to compare estimates, priorities and completion dates in a spreadsheet. A CSV is a reasonable starting point. You are asking for selected facts, and you can inspect the file before using it. Paying for a recovery service solely to produce that report may add little value.
Store the export somewhere appropriate for the information it contains. Decide who can access it and when to delete it; a downloaded file becomes another copy your team needs to manage.
A project disappears and you notice immediately
Start with Linear's own recovery tools. Its recently deleted list keeps deleted issues and other supported items for up to 30 days. If the missing item is available there, restoring it directly may be the quickest option.
A backup is useful when you need an older copy that those tools no longer hold, provided a suitable snapshot was captured and retained. Neither a new export nor a newly connected backup service can recreate data that was already gone before it had a chance to save it.
A bulk edit overwrites important descriptions
Consider a hypothetical team that exports on Friday, adds work over the weekend, and discovers a bad edit on Monday. Friday's file might contain the old descriptions, but it cannot contain the weekend changes. The team must decide which earlier values to recover and which newer work to keep.
A more frequent schedule can give that team a more recent recovery option. It still needs a copy from before the mistake. Daily backups also leave an interval between runs: changes made after the last usable snapshot may be absent.
With BackupLatte, the recovery route for overwritten content is to restore an earlier snapshot into a separate, preferably empty workspace, compare it with current work, and bring back what you need. Restoring into the original workspace leaves existing records unchanged. It brings back missing records; it does not roll existing descriptions or statuses back to an earlier version. The restore guide explains both targets.
What to check before choosing automated backup
A schedule saves someone from remembering to export. It does not, by itself, prove that you can recover what matters. Ask these questions of a service or a backup system you build yourself:
- Coverage: which records and files can it read, and which can it write back? A downloadable record is not necessarily restorable.
- Access: can the connected account see every team you need protected? Check private teams explicitly instead of assuming that “workspace” means everything.
- Retention: how long might it take your team to notice a problem? Keep enough history to cover that delay, within your data retention requirements.
- Failures: how will you learn that a run failed or a file was skipped? Check the last usable copy, not just whether a schedule is enabled.
- Recovery: who can start it, where will the data go, and what happens to existing work? Practise before an incident makes those questions urgent.
- Portability: can you download and inspect the saved data if you stop using the service?
A custom API export can be a valid choice when your team wants to own this work. Budget for maintaining collection, storage, monitoring and recovery together. Automating the download answers only the first part of the problem.
Where BackupLatte fits
BackupLatte captures supported Linear data daily, including issues, comments, projects, documents, archived records and linked uploaded files. Coverage depends on the connected user's access and Linear's API. Snapshots report gaps. With a confirmed account email and problem notifications enabled for the workspace, email alerts notify you about failed or incomplete backups. You can download the data and copied files as a ZIP.
It is not an exact reconstruction of every aspect of a workspace. Collection happens while work continues. Some saved data cannot be restored, and recovered records can have different issue numbers or authorship. Read the Linear limitations and restore behavior before making it part of your recovery plan.
Use a CSV when you need a manageable record for analysis or a narrow transfer. Use scheduled backups when recovering earlier work is important enough to require retained copies, monitoring and a practised restore process. You can use both: export for today's report, and keep backups for the day you need yesterday's work.
For the wider recovery workflow, continue with How to back up and restore your Linear workspace.