Every new system asks you to trust it with your books. Here is what happens when something breaks or data goes wrong.
The database is backed up continuously and can be put back to any moment in the last seven days.
Four weekly and six monthly backups are kept on top of that, so a problem found late can still be undone.
Attached documents are backed up every night, and the backups are stored in a second region.
A backup nobody has restored is only a hope. In our most recent drill the database and every document came back in under five minutes, and we checked that the two matched.
A restore comes up as a separate copy. Nothing replaces your live data until someone has confirmed the copy is right.
Each client runs on its own application and database, so a problem at one company stays at that company.
We do not release updates between the 28th and the 5th, while rent is coming in and the month is closing.
Each release runs on our own test system, on the same cloud infrastructure, before it reaches yours.
Before a release we mark a restore point. If the release causes trouble, the application goes back to the version that worked and the database to the moment before.
Cash receipts post one at a time and remember what finished, so re-running after an interruption picks up where it stopped instead of posting twice.
Every instance is checked every five minutes, and a failure reaches us by text. If something affects your company, we contact you. You should never have to be the one who discovers it.
Current status and ninety days of uptime are on our status page, updated every five minutes, with a note on anything we are working on.
What if the system goes down on the first of the month? What if the books come back wrong? A vendor who has thought about those questions should have the answers ready. These are ours.