Three questions to ask before you buy any software

A feature list is a sales document. It is written by the people who want you to buy the thing, and it is designed to make every product look like the answer to a question you have not asked yet.

Here are three questions that are much harder to answer, and much more useful.

"What happens on the worst day?"

Demos show the happy path. Your team does not live on the happy path.

Ask what happens when the data is late, when someone leaves mid-process, when two people edit the same record, when the integration is down. The gap between how a tool behaves on its best day and on its worst day is the gap you will actually be living in.

"Who has to change their behaviour for this to work?"

Every tool has an implicit deal: it gives you an outcome, and in return somebody has to do something differently. Enter data earlier. Close a job properly. Use one field instead of a free-text note.

If nobody has identified who that somebody is, the tool will not land. The most common cause of a failed rollout is not the software. It is that the behaviour change it silently depended on was never named, never agreed, and never resourced.

"What is the exit?"

Ask how you get your data out, in what format, and whether the export includes the things that matter: history, attachments, the relationships between records, not just a flat table of rows.

You are not being cynical. You are asking whether this is a tool or a trap. A vendor confident in their product has a good answer ready, because they do not expect you to need it.

If a system cannot be left, it has not been chosen. It has been submitted to.

And the one to ask yourself

Before any of it: what would happen if we simply stopped doing this step?

It is astonishing how often the honest answer is "nothing much", and how rarely anyone asks, because the step has been there long enough to look like the work itself.