The most expensive document problems in a small business are not lost files. They are files that still exist, look fine, open normally, and are missing an afternoon of somebody's work. Nothing failed, nothing was reported, and the loss is usually found weeks later by a customer reading a number that was corrected and then uncorrected.
The mechanism is ordinary. Two people open the same price list, the same quote or the same spreadsheet of stock. Each makes changes to their copy. Each saves. Whoever saves second writes their whole version over the first, because a file saved as a whole does not merge - it replaces. If the two people were editing different rows, the result is not a conflict warning; it is a document that silently reverted half of one person's work while keeping all of the other's. The file has no way to tell you this happened and the person whose changes vanished has no reason to check.
The situation gets worse when the document is passed around rather than shared, because then every copy is a branch. A quote emailed for review comes back with edits, but in the meantime the original was also edited, and now there are two documents both of which are partly right. The habit of adding a date or the word final to the filename is an attempt to solve this by naming, and it fails for the same reason it always fails - the name records when somebody thought they were finished, not which version contains which changes.
The reliable fix is to stop having copies. Documents that more than one person touches should live in one place that both people open, rather than being sent as attachments, so there is one file and it is always the current one. Systems that work this way merge simultaneous edits instead of overwriting them, keep a history you can look back through, and make the question who changed this answerable rather than a matter of recollection.
Where that is not possible - and it often is not, for anything involving an accountant, a client or a legal document - the second-best arrangement is to make ownership explicit. One named person holds the master copy; everyone else sends changes as comments rather than as edited files; and the master is updated in one place by one hand. It is a slower process and it produces a document that is right, which is the cheaper of the two outcomes.