Exit Reports

There is a question almost no software review answers, and it is the one that decides whether a bad decision costs you a month or costs you three years.

If you leave, what actually comes with you?

Not "does it export" — every product exports something. The question is what is inside the export, and more importantly what is not: the comments, the attachments, the custom fields, the record of who changed what and when. That is where the value of three years of work usually lives, and it is the part that most often does not travel.

Why this section exists

Vendors are not hiding this information. Most of them publish it, in careful detail, in help articles nobody reads until the week they are trying to leave.

What is missing is anyone reading those documents side by side and writing down what they say. So a business chooses a product on price and features, works in it for four years, and discovers the constraints at the exact moment it has the least leverage and the least time.

These pages are that reading, done in advance, with the receipts attached.

What an exit report contains

Each one is the same shape, so you can compare them without re-learning a format:

• An exit score out of 30, from six criteria scored zero to five, with the reasoning printed beside each score rather than hidden behind a methodology link.
• What comes out and what stays behind, as two plain lists.
• A before-you-cancel checklist specific to that product's quirks.
• What we could not verify — the questions the vendor's documentation does not answer, published rather than papered over.
• The date the documentation was checked, at the top of the page, because a fact about export tooling has a shelf life.

The full rubric, including the things we deliberately refuse to score, is in how we score exit reports.

The reports

Sixteen so far, covering the products already reviewed elsewhere on this site, so the buying advice and the leaving advice sit next to each other. Scores run from 29 out of 30 down to 17.

Project and work management

• Notion — 25/30. Unusually good on comments, with a workspace export that can take up to 30 hours and a download link that expires in seven days.
• Jira Cloud — 22/30. A hard, non-configurable ceiling of 10,000 work items per CSV export; Atlassian publishes the batching workaround rather than a way to raise it.
• monday.com — 21/30. Admin-only, a 24-hour build feeding a 24-hour link, one attempt per day — and archived boards are excluded from the account export entirely.
• Asana — 20/30. Names the formats and the plans, then says nothing at all about what the file contains.
• Teamwork.com — 20/30. The full export is a MySQL database backup: complete, honestly described, and unopenable without a database server.
• Trello — 20/30. The JSON export carries only the 1,000 most recent actions on a board, and CSV requires Premium.
• Wrike — 18/30. The export available on every plan excludes descriptions, attachments and comments; the complete one needs Business and cannot be reimported.
• ClickUp — 17/30. CSV only, with export availability varying by plan and user role in ways the documentation does not spell out.

Password managers

• Bitwarden — 29/30. The most complete exit on this site: four formats, one encrypted, one carrying attachments, with no caps, paywalls or expiring links.
• Proton Pass — 25/30. A PGP-encrypted archive in a format other tools can read — but you cannot export from the mobile apps at all.
• 1Password — 24/30. The CSV export carries only login and password items, which means custom fields, and anything you stored in them, stay behind.
• Dashlane — 24/30. Export on every plan, an encrypted direct transfer on recent devices, and a blunt warning about the CSV route in Dashlane's own documentation.
• NordPass — 19/30. CSV and nothing else: no encrypted archive, no direct transfer, and no statement of which item types travel.

Files and storage

• Dropbox — 25/30. Files leave as files, which is the cleanest possible outcome. Browser folder downloads stop at 10,000 files, and the desktop client is the workaround most people miss.
• Google Takeout — 24/30. Broad coverage, with an archive that expires in about seven days and can only be downloaded five times.
• OneDrive — 23/30. The most precise limits published by anyone here, and the tightest — plus one absolute exclusion: OneNote notebooks cannot be downloaded at all.

How to use these before you buy

The point of reading an exit report is not to avoid products that score badly. Plenty of excellent software has a poor exit, and a low score is a reason to plan rather than to walk away.

Three practical uses:

Price the exit before you commit the data. If the export drops comments and history, you now know that adopting the product means accepting that those live inside it permanently. That may be perfectly acceptable — but it should be a decision rather than a discovery.

Keep the irreplaceable things somewhere portable. If a product cannot export the one field you would be lost without, do not put it there. This is the cheapest possible mitigation and almost nobody does it, because nobody checks until it is too late.

Export once, early, while you are happy. Running an export in your first month tells you what the file actually contains at a point when nothing depends on the answer. It is twenty minutes, and it converts an unknown into a known.

What we do not do

We do not claim to have run every export ourselves. These reports are built by reading what each vendor publishes, quoting it, and linking to it, so that every claim on the page is one you can check in a minute.

We do not score things the documentation does not cover. Where a vendor is silent, the report says so in its own section rather than inferring an answer, and the score is calculated only from what is actually documented.

And we do not treat a good exit score as a recommendation, or a bad one as a warning. It is one input, sitting alongside price, fit and everything else — see the buying guides for the rest of the decision, and the comparisons when you are down to two.