Here are some suggestions to improving the documentation from a new user’s perspective. I am pretty sure I managed to figure out backups and snapshots while the S3 sync is not yet clear to me.
Let me point out a few examples:
1. This page: GitHub - gristlabs/grist-core: Grist is the evolution of spreadsheets. mentions:
Enables backups that you can confidently restore in full.
and links to Exports & backups - Grist Help Center - easy to understand and follow. That page then links to Automatic backups - Grist Help Center where terms start to get mixed up.
Grist automatically saves backups of your documents as you work on them. These backups – or snapshots – of the document can be examined at any time.
Grist retains more frequent snapshots of recent changes, and less frequent ones as you go back in time. For example, hourly snapshots are retained for about a day, but monthly snapshots are retained for more than a year.
2. This page Self-managed Grist - Grist Help Center again mixes up both terms:
- Consider enabling snapshot support if you want Grist to handle document backups.
It links to Self-managed Grist - Grist Help Center where things get even more confusing as we are now no longer only talking about backups and snapshots but throwing in syncing too:
Grist’s cloud storage feature allows automatic syncing of Grist documents and document versions to an S3-compatible bucket (available for all Grist versions)
Then goes on to say:
Here is an example of running Grist locally, with snapshots stored in a test MinIO instance:
Which snapshots? The snapshot snapshots or the backup snapshots? 
3. Trying to figure things out, I found some confusing statements on this forum.
one:
If you hook up snapshot storage, then documents (with their attachments embedded in them) can be stored in S3 and similar.
Do you mean stored or synced?
two:
Try restoring from an automatic snapshot, see here: Automatic backups - Grist Help Center again, talkling about autoamtic snapshots and linking to autoamtic backups.