Grant,

Nice description; down to earth and understandable.

I would guess (perhaps wrongly) that almost all people in the room have probably backed up a database. A full backup at least. Perhaps never tried to restore before. Perhaps mentioning terms like "point in time" restore and what is needed for that to work (log chain).

Issues that I have seen in the wild (leading to companies calling me) include:

1) Restore over top of the production database rather than making a 'copy' to sit on the same server. I know, restoring to another server is best, but I get calls after folks think this is going to be easy and blow away production.

2) Backup files ending up on the same physical drives as the .mdf/.ndf or .ldf's (backup to same logical drive letter, to different logical drive which is a partition of the same physical disk, to a shared SAN, or on VM drive which maps to the same physical disks).

This may not be in scope for your talk perhaps, and all of this detail should not end up in the description. However, maybe the description can hint at "no, your backups may not really be ok" with a buzzword or two to catch the attention of people who know super basic backups, but that's it.

Jim