The everyday restore
Most restore requests are small. Someone deletes a folder by mistake, overwrites a spreadsheet, or needs last month's version of a proposal. These are solved by the recycle bin, version history in OneDrive or SharePoint, or a single-item restore from the backup console. Done well, the person has their file back before their coffee goes cold.
To make these restores easy, keep enough history and make the process known. Show staff how to use version history themselves for recent changes, and make sure whoever looks after IT can search the backup by file name and date. If restoring one file means logging a request and waiting two days, people will start keeping their own copies on USB sticks, which creates new problems.
The whole-office scenario
Now imagine the office is unusable: a fire overnight, a flood from the floor above, a burglary that took every computer, or ransomware that encrypted every device on the network. The question is no longer which version of a file you want. It is how to get fifteen people working again, with what equipment, from where, and in what order.
This kind of recovery is a project. It may involve buying or hiring laptops, setting up internet at a temporary location, restoring servers to cloud infrastructure or new hardware, re-installing line-of-business software, reconnecting printers and phones, and telling clients what is happening. The backups are essential, but they are only one part of the work.
What makes the difference on the big day
Offsite copies are the foundation. If every backup was in the same building, there may be nothing to restore. Beyond that, the businesses that recover fastest have documentation: a list of systems, where each is backed up, licence keys and installation media, supplier contacts, network settings, and admin passwords stored securely somewhere reachable without the office.
They also have a priority order agreed in advance. Email and phones usually come first so the business can communicate, followed by the system that makes money, such as job management or accounting, then file shares, then everything else. Deciding this in the middle of a crisis wastes hours and causes arguments.
Cloud services change the picture
When email, files and key applications live in Microsoft 365, Google Workspace and cloud software, a whole-office disaster becomes far more manageable. Staff can sign in from home or a borrowed laptop and keep working while the premises are sorted out. The recovery effort shifts from rebuilding servers to replacing devices and restoring anything that only existed locally.
That said, a ransomware event or compromised admin account can still affect cloud data. Independent cloud backups and strong sign-in security matter just as much for cloud-first offices, because the cloud account itself can become the thing that needs recovering. A cloud-first office should still rehearse what happens if the main admin account is locked or hijacked, because that is the cloud equivalent of the building burning down.
Testing both kinds of recovery
Test the small restore monthly, picking random files from different systems, to make sure everyday recovery works and the retention is long enough. Test the big restore at least yearly by walking through a full scenario with the actual backups and plan. Time how long each critical system takes to come back, and note anything that depends on access to the building.
The two tests reveal different problems. The small one shows gaps in coverage, such as a folder or user not included. The big one shows gaps in planning, such as missing licence keys, no spare hardware or nobody knowing the router settings. Both kinds of finding are valuable, and both are much cheaper to fix before a real event than during one.
Preparing without overspending
You do not need an enterprise disaster recovery budget to be ready. Start by moving what you can to cloud services with independent backup, keep an immutable offsite copy of anything still on site, write a short recovery plan with priorities and contacts, and store it where you can reach it from a phone. Then test both kinds of restore on a schedule.
Those steps cost relatively little and turn a potential business-ending event into a stressful but survivable week. The deleted-file restore keeps everyday frustrations small; the whole-office plan keeps the business alive. If you only have time for one improvement this year, make it the offsite, immutable copy, because without it the rest of the plan has nothing to restore.
In short
- Single-file restores rely on version history and enough backup retention.
- Whole-office recovery needs offsite copies, documentation and an agreed priority order.
- Cloud services make site disasters easier but still need independent backup.
- Test small restores monthly and full scenarios yearly.