I do agree with SQLBill, but some people not only don't have the expertise, they also do not have the disk space required to do the backup.
Take for example, I had a customer who had 3 SQL databases, all around 200gb each - he definitely had the free space on the server to let the db's grow but just wasn't enough, nor could he justify buying another 600gb of disk space just to do backups of SQL.
One thing with Veritas is that they have a notification service for any Hotfixes or other changes that come out -
One other point, this is with Exchange 5.5.
Customer was using Backup Exec 9.1 to backup the information stores, working without issues. They did did a defrag of the databases as part of normal maintenance they do, then 1 night later he got corrupt store errors in BE. No errors were reported in the App logs, so not sure exactly at this point what it was.
Tried out NTbackup - it worked 100% everytime.
Spoke to Veritas - they were saying that the databases were corrupt in some way. the problem was ISinteg, eseutil were not coming up with any errors on the IS or Public Folder databases.
They proved the point by trying to restore a backup of Exchange made by Ntbackup - the restore would finish - but the Exchange services would not start. It was saying the logs were inconsistent.
So I had to run a ESEUTIL /P on the IS to make it consistent.
We later found out the problem was that Ntbackup was not correctly backing up the Transaction logs (it was backing up any logs). We couldn't get a straight answer out of MS regarding this problem - in some ways they denied it was their problem.