A backup serves one purpose: getting the business back to work after an incident, whether a dead disk, a human error or ransomware. Veeam is the tool we install most often in Tunisian SMBs, because it backs up physical servers, virtual machines and Microsoft 365 alike, and above all because it lets you test a restore without touching production.

Here is the starting plan, in the order we run it.

1. List what must survive an incident

Before talking tools, take inventory: file server, the management software's database, email, the management team's workstations, the website. For each one, two questions:

  • How many hours of work can we afford to lose? That is the RPO (recovery point). An invoicing database rarely tolerates more than a day; a document archive may tolerate a week.
  • How long can we do without it? That is the RTO (recovery time). An online shop: a few hours. A test server: several days.

These two figures drive everything else. A 20-person SMB usually ends up with three or four criticality levels, no more.

2. Choose frequency and retention

With Veeam, a daily incremental backup is the norm: only the difference since the previous day is copied, which takes a few minutes at night. For critical data, add restore points every 4 hours.

Retention — how many versions you keep — depends on usage: 14 to 30 days online, then a monthly copy kept 12 months for accounting obligations. Ransomware sometimes lies dormant for weeks before acting: a retention that is too short keeps only copies that are already encrypted.

3. Apply the 3-2-1 rule

Three copies of your data, on two different media, one of them off-site. In practice for an SMB:

  1. Production data on the server.
  2. A local backup on a NAS or dedicated server, for fast restores.
  3. An off-site copy: LTO tapes stored in a safe place, or a cloud repository with a host in Tunisia.

The off-site copy must be disconnected or immutable: ransomware encrypts everything it can reach, including a permanently connected backup disk. We detail this on our Infrastructure & Backup page.

4. Test a restore in the first week

This is the step most companies skip, and the one that matters. Veeam can start a machine directly from its backup (Instant Recovery) and automatically check that it boots (SureBackup). We schedule a full, timed test every quarter: we record the real recovery time and compare it with the RTO decided in step 1.

Until that test has been done, you do not have a backup: you have a hope.

5. Indicative budget

For a Tunisian SMB with two or three servers, expect:

  • Veeam licences: per protected machine, billed yearly.
  • Local storage: a professional NAS with redundant disks, between 2,500 and 6,000 TND depending on capacity.
  • Off-site copy: an LTO drive and tapes (one-off investment), or a cloud storage subscription billed per TB per month.
  • Setup and testing: a few days of work, then a quarterly check.

The overall order of magnitude, first year included, is generally between 6,000 and 15,000 TND for an infrastructure of this size. Compare that with the cost of a full week of downtime.

What we offer

Semocode designs, installs and tests Veeam backups for SMBs and clinics in Tunisia, with copies on LTO tapes or in a Tunisian cloud. The first step is a half-day audit of what you have: book a call, and you leave with the list of what must survive and a costed plan.