Most businesses have some kind of backup. The question is whether it would actually get you working again after a bad day: a failed server, a stolen laptop, a flood in the office or a ransomware attack that encrypts everything it can reach. The 3 2 1 backup rule is a simple test you can apply to your own setup in a few minutes, and it is still one of the most useful ideas in IT.
What the 3 2 1 backup rule means
The rule has three parts:
- 3 copies of your data. Your live data plus at least two backups.
- 2 different types of storage. For example, a backup appliance in the office and cloud storage, so that one fault or one type of failure cannot take out every copy.
- 1 copy offsite. At least one copy in a different physical location, so a fire, flood or burglary at your premises does not destroy your backups along with your computers.
The idea is that no single event, whether a hardware fault, a mistake, a disaster or an attacker, should be able to destroy every copy at once.
The modern version: 3 2 1 1 0
Ransomware changed the picture. Attackers now deliberately look for backups and try to delete or encrypt them before they demand payment. The UK's National Cyber Security Centre (NCSC) warns in its guidance on offline backups that ransomware has encrypted not just original data but also connected USB drives and network storage holding backups. That is why many IT providers now talk about 3 2 1 1 0:
- 1 copy offline, air gapped or immutable. Offline means disconnected from your network. Immutable means stored so that it cannot be changed or deleted for a set period, even by an administrator. Either way, an attacker who gets into your systems cannot reach it.
- 0 errors. Backups are checked and test restores complete without errors. This is the part most often skipped.
The NCSC's advice for small organisations follows the same thinking: make sure at least one backup is not permanently connected to your live systems, and that you can actually restore from your backups.
RPO and RTO in plain English
Before choosing how to back up, it helps to answer two questions. IT people give them acronyms, but the ideas are simple.
Recovery point objective (RPO): how much can you afford to lose?
If your systems failed right now, how far back could you go without serious harm? If you back up once a night and the server fails at 4pm, you lose a day's work. For an accountant in January, that might be painful. For a busy online shop, it could be very costly. Your RPO tells you how often you need to back up.
Recovery time objective (RTO): how long can you be without it?
How long could the business cope without a system before it really hurts? An hour? A day? A week? Restoring a whole server from cloud storage over a typical office connection can take far longer than people expect. Your RTO tells you how quickly you need to be able to restore, and whether you need a local copy or a standby system to hit it.
Answer these for each important system, not just the business as a whole. Email, your accounts package and your shared files may all have different answers.
Applying 3 2 1 to a small business
Here is how a typical business of 10 to 50 staff might apply the rule. Your setup may be different, but the pattern is the same.
| Data | Copy 1 (live) | Copy 2 | Copy 3 (offsite and protected) |
| Microsoft 365 email, OneDrive, SharePoint and Teams | Microsoft 365 | Built in retention and recycle bins (short term only) | Independent Microsoft 365 backup to separate cloud storage |
| On premises server or line of business software | The server | Local backup device or storage for quick restores | Encrypted, immutable cloud backup |
| Laptops and desktops | The device | Files kept in OneDrive or SharePoint rather than only on the device | Covered by the Microsoft 365 backup |
| Cloud software such as accounts or CRM | The supplier's platform | Regular exports where the software allows it | Exports stored in your own backed up storage |
Two common traps: first, a USB drive that stays permanently plugged into the server is not offsite and not offline. Second, file sync is not backup. If a file is deleted or encrypted, sync faithfully copies that change everywhere. Our article on whether Microsoft 365 backs up your data explains this in more detail.
Testing restores: the step everyone skips
A backup job that reports "success" every night only tells you that the software ran. It does not tell you that the right data was included, that the files are readable or that you know how to get them back under pressure. Testing is what turns a backup into a recovery plan.
A sensible routine for a small business:
- Restore a few individual files and emails every month and check they open.
- Restore a full mailbox or a whole folder every quarter.
- At least once a year, restore a complete server or system to a test environment and time how long it takes. Compare that with your RTO.
- Write down what you did, what worked and what did not.
Your 3 2 1 backup checklist
- We have listed every system and where its data lives, including cloud services.
- We have agreed an RPO and RTO for each important system.
- Every important system has at least three copies of its data.
- The copies use at least two different types of storage.
- At least one copy is offsite.
- At least one copy is offline or immutable.
- Backups are encrypted, and the encryption keys are stored safely somewhere else.
- Access to the backup system is protected with multi factor authentication and limited to named people.
- Someone checks backup reports every day and follows up failures.
- Test restores happen on a schedule and the results are recorded.
- The backup plan is part of our business continuity plan.
How Cyber Fits can help
We design, monitor and test backups for businesses in Sunderland and across the North East, covering Microsoft 365, servers and cloud systems. If you would like a second opinion on whether your backups meet the 3 2 1 rule, see our backup and recovery service or book a free IT review.