The exit test: getting your data out before you put it in
The most informative twenty minutes of any trial is the one the vendor has no incentive to make easy.
Every evaluation examines how well data goes into a product. Almost none examine how it comes out, which is the question that determines whether you can ever leave.
Running the export during the trial — for real, and opening the files — is cheap, quick, and disproportionately informative. A trial involving task-switching cost should test the awkward cases rather than the happy path; this guide provides a concrete vendor example to design those checks around.
What to actually do
- Export everything, yourself, from the interface, without asking support to do it.
- Open the files. Check that fields are populated and that the structure is comprehensible.
- Check attachments and documents come out as files, not as filenames or links.
- Check the history: comments, audit trail, changes over time. This is what is most often left behind.
- Check relationships survive — which record belongs to which — rather than arriving as disconnected tables.
Core records almost always export. Documents, comments and audit history frequently do not, and they are often the part with the most accumulated value.
Ask the contractual questions now
What is the notice period. What happens to the data at termination. How long is it retained afterwards and in what form. Is there a fee for an export or a migration.
These are easy questions to get answered while a vendor is trying to win your business, and hard ones afterwards. Get the answers in writing and attach them to the file.
Distinguish an export from a migration
An export is a file. A migration is that file in a state another product can ingest. A vendor can satisfy the first entirely while leaving you with something no other system will accept.
The rough check is to look at the export and ask whether a competent person could load it elsewhere without a bespoke script. If the answer is clearly no, that is a lock-in cost and it should be weighed alongside the price.
Lock-in is not a reason to refuse; it is a price
Some degree of switching cost is unavoidable and often worth accepting. The mistake is accepting it without knowing its size.
A product that is excellent and hard to leave may still be the right choice. A product that is adequate and hard to leave is a poor bargain, and you cannot tell the difference without running the test.
Do it again annually
Export capability changes. A product that exported cleanly at purchase may not after two years of new features and a platform migration.
An annual export, kept as a backup, verifies that leaving is still possible and gives you a copy of your own data that does not depend on the vendor continuing to exist. It takes an hour a year.
For privacy risks exposed during a pilot, the NIST Privacy Framework provides a broader independent framework.