Keeping your team records on paper is dangerous. Damage to buildings from fire or flood can cause permanent loss to years of your collected data. Entering your records on a computer allows you to make exact copies of the data, and be able to replicate them if the original is lost.
If you are using a computer to store your records, make regular backups of your data - old backups are worthless. Do not keep your backup on the same computer, transfer them to a memory key and store it another building. Keep regular backups, and multiple versions incase your last backup is corrupt. Most importantly, test your backup regularly by restoring the data back to where it came from.
If you're going to do your own off-site backups we recommend putplace.com and getdropbox.com as great tools.
If you store your records in Decisions For Heroes, all these backups are done automatically. Every 24 hours we automatically take a versioned backup of all records and send them encrypted over the internet to a second data centre, giving us geographic redundancy. As we version all of this data, we are able to roll-back to any date since we started. Regularly we take a random version of our data backup and load it into a staging server to check it for any inconsistancies.
To double-up on this, every night each of our servers create an exact snapshot replica of themselves and stores these hard-disk images to a physically seperate RAID-10 disk storage. We also take weekly snapshots to give us a clean replica if a corruption extends over 24 hours.
Our triple redundancy comes from our software application backup. Every edit made to our software code is versioned and transported by encrypted link to an off-site data centre. Once ready for release, our versioned code is deployed to a staging server where it is tested on a snapshot of live data, before being released to the production environment.
You worry about the rescues - we'll take care of your data!