Picture an accounting practice with twelve staff. A week before annual reports are due, the server holding the accounting database refuses to boot. Every evening the backup software had reported success, yet nobody had ever checked how long rebuilding the whole server on new hardware would take, or whether the database would even open afterwards. For us this is question number one: how long can each system be missing, and how much work can you lose before the business is seriously hurt? Tools, storage locations and retention periods are chosen afterwards. The last step is always a genuine restore against the clock. Everything, testing included, is done remotely.
The first service builds the whole backup system along with regular restore tests. The second adds a copy outside your premises that water in the basement cannot destroy and an intruder with a stolen admin password cannot erase.
When your backup rests on one external drive, a script written by a former colleague, or hope. We list every source, set targets per system, configure jobs and alerts, and test restores on a schedule.
A copy encrypted on your side and then sent to an EU data centre, into an account belonging to your company. It is locked against deletion, and its credentials live outside your domain.
Every system gets two numbers. RTO is the time by which it must be running again. RPO is the most work it is acceptable to key in a second time. The values below are typical examples that often prove sensible; yours are agreed together with you.
RTO half a working day, RPO 15 minutes. Every lost invoice or goods receipt has to be entered again. With Pantheon, SAOP or a self-hosted Minimax we therefore copy the database continuously through its transaction log and the full server each evening.
RTO one day, RPO a few hours. People can keep going for a while with the files they have open. An overwritten document is recovered by the user from a snapshot or from SharePoint version history.
RTO a few hours, RPO one day. The usual request is not a whole mailbox but a single message or folder that vanished weeks ago. For that you need a separate copy of Microsoft 365 or Google Workspace with longer retention.
RTO under an hour, RPO a few minutes. Orders, Stripe or Bankart payments and stock levels change by the minute. Restoring last night's copy would wipe out a whole day of sales, so we set up a mirrored database and a spare server ready to take over.
RTO two hours, RPO one hour. A clinic grinds to a halt without its records and appointment book. Copies of health data are always encrypted, and the list of people allowed to restore them is short and written down.
RTO one day, RPO one week. CNC programs and machine settings rarely change, yet without them production stands still. A weekly copy to a shared folder is often enough, provided nobody forgets it exists.
Syncing is not backing up. OneDrive, Google Drive and Dropbox faithfully copy every change to every device, deletions and encryption included. If ransomware scrambles a folder on one laptop, the encrypted files turn up everywhere within minutes. Version history helps, but it expires and can be wiped by anyone with enough rights. So alongside any sync service we always add an independent backup with its own retention rules and its own admin accounts.
We connect to servers, NAS units and cloud tenants through secured remote access. Plenty of providers stop after step three. We carry on, because only step four shows whether the whole thing works.
Each data source gets a row in a table: where it is, who owns it and whether it is copied today. Laptops, field tablets and private cloud accounts are included too, since work files have a habit of drifting there.
For every row you decide how long it may be unavailable and how much work you can lose. We tell you what each option costs; the choice stays with you.
Jobs, encryption, separate admin accounts and the off-premises copy are put in place. Alerts fire on errors and on reports that failed to show up, because a job that silently stalls never complains.
Chosen systems are brought back in a separate environment, timed, and anything that snagged is written up. The record serves as evidence for directors, auditors or a cyber insurance provider.
They keep the service running and make sure data is not lost to faults inside their own data centres. Mistakes on your side remain your responsibility: a folder deleted after the bin has expired, a hijacked account emptied by an attacker, or OneDrive files encrypted by ransomware. Only an independent copy governed by your own retention rules covers those cases.
Unplug affected machines from the network but leave them powered on, then email us. Do not pay and do not delete anything until we know how the attacker got in. We gauge the damage, find the most recent clean copy and begin recovery according to the plan. Where personal data is involved, the Information Commissioner may have to be informed within 72 hours, and SI-CERT can be notified as well. A legal adviser writes the notice; we contribute the technical findings.
A sensible starting point is 30 days of daily copies, three months of weekly ones and a year of monthly ones. Invoices and accounting records are subject to statutory retention periods that your accountant should confirm. At the same time, GDPR requires that personal data is not kept longer than necessary, and your retention policy applies to backups too. We turn both requirements into concrete settings in the backup software.
The local copy stays with you, on a NAS or another server. The off-site copy lives in an EU data centre: with European providers such as Hetzner, OVHcloud and IONOS, or in an EU region of Azure, AWS or Google Cloud. We do not use providers outside the EU. The exact location, country and provider are recorded in the documentation so you can list them in your record of processing activities.
Storage is paid by you to the provider, by the gigabyte used; for a small firm the monthly bill is usually modest. Our labour is charged at €50 per hour plus VAT with an estimate before we start, or at a fixed price when the scope is well defined. Continuous backup monitoring and periodic restore tests can be bundled into a Start, Business or Premium plan.
The NAS flags the faulty disk itself, and the warning often lands with us first. We tell you which drive to order and send a picture with the right bay marked. A colleague of yours makes the swap while we are on a video call, or a local contractor you pick does it. We then watch the RAID rebuild and verify the data remotely. We neither visit nor sell equipment.
Describe what is backed up today and how. We review the set-up remotely, attempt a real restore and tell you frankly where the gaps are.
Enquiry received
An engineer will write back no later than the following working day. Faults that leave your staff unable to work are always looked at first.
No town by that name here. Check the letters, or settle for the nearest larger place. Our engineers work only over the internet, so Kočevje and Koper get precisely the same treatment.
