Service · Backup and restore

Backup implementation

Software comes last. Two figures come first: the amount of recent work your business could bear to key in a second time, and the length of standstill it could survive. Skip them and every product comparison turns into a coin toss whose result only surfaces on the morning a machine refuses to boot. Our starting point is therefore recovery, and the schedule is derived backwards from it. Everything is assembled across an encrypted remote session, with nobody setting foot on your premises.

RPO and RTO
on paper before any shopping
3-2-1
three copies, one kept elsewhere
Immutable
a copy no credential can wipe
Drill
recovery rehearsed against a clock

Where this service reaches

Copying is half the exercise. The other half is the habit of verifying it. Until something has come back out, what you own is a hypothesis rather than a safety net.

Put your scope to an engineer

Ranking the systems

Each application receives a place in the queue, an acceptable gap in data and a deadline to be usable once more. A practice closing annual accounts answers that question nothing like a workshop running around the clock.

A 3-2-1 layout

Data lives in three places across two unlike storage technologies, with one place beyond your building and beyond your Windows domain. Usually: the server itself, a NAS in the cupboard, rented object storage in Europe.

What the jobs cover

Virtual machines under Hyper-V, VMware or Proxmox. Databases feeding Sage, EBP, Cegid or Odoo. Shared folders, plus mailboxes and document libraries in Microsoft 365. Firewall and switch settings, which almost nobody remembers to include.

Putting a copy beyond reach

The backup console sits outside the domain, its accounts stand apart and demand a second factor, and locked objects decline an early deletion request whoever happens to sign it.

Alerts with a name on them

Failures reach a specific colleague as well as our own team. So does silence, because a job that stops reporting has usually stopped running, and only one of those two facts announces itself.

A drill calendar

Occasional single-file or single-database recoveries, plus a timed rebuild of one entire server, repeating at whatever interval you pick. Each rehearsal leaves a note with a date on it.

The way an engagement runs

Duration follows from the count of systems and the size of the data. Before anything begins we agree a timeline that already contains the opening full copy and the opening rehearsal.

01

Stocktake

Which data is captured now, which is not, where existing copies rest, and whether a single file has ever travelled back out of them.

02

Blueprint

Sizing, retention rules and an overnight slot wide enough for every job to finish before the first colleague signs in.

03

Build

Repositories, jobs, permissions, encryption keys and alerting, assembled over a remote session while your people carry on working.

04

Proof

A recovery, performed and timed. The reading on the clock is the figure you can quote, unlike the one printed on a datasheet.

Reach back counts for as much as frequency. Intruders commonly spend a month inside a network, quietly charting servers and hunting for repositories, before anything at all is encrypted. Hold a week of history against a break-in that began three weeks earlier and every single copy already carries a passenger. Treat thirty dailies plus twelve monthlies as the floor rather than the ceiling.

Questions and answers

Cleanly taken database dumps, entire virtual machines, switch and firewall settings, directory objects or an Entra ID export, licence keys, and whatever your password vault holds for service accounts. Settings are where the gap usually sits. Getting files back is straightforward; reconstructing how somebody wired the network together years ago is not, particularly when that somebody booked a fortnight off.

It is a fair first move for a one-person business and nothing more for a company. Left plugged in, the disk is encrypted alongside its source. Unplugged and carried home, it leans on a human remembering, which they will not, at any rate not every week. A scheduled NAS paired with rented European storage costs little extra and forgets nothing.

No, the environment decides. Veeam and Nakivo suit hypervisors; a Synology box already in the cupboard brings Active Backup along with it; Microsoft 365 taken alone is better served by a purpose-built cloud service; and the smallest offices sometimes need nothing beyond the utility shipped with the NAS. No vendor pays us, so we steer you towards the smallest licence that covers the requirement.

At EUR 95 per hour excluding VAT, or absorbed into a Start, Business or Premium plan. Hours are quoted once the stocktake is finished, never earlier. Storage rental and software licences sit on separate lines, and both figures reach you ahead of any commitment.

A colleague of yours follows our photo guide, or the hardware dealer you already buy from handles it. Choose and purchase the box wherever suits you; our contribution is naming a sensible model and capacity. From the moment it answers on the network, the remainder belongs to us, over remote access. Shipping or fitting equipment is something we simply do not offer.

A backup with a proven way back

Describe what has to survive and how long the business could sit idle. A copy plan built around those two answers comes back to you.

When we are around
Weekdays, 8:00 to 18:00 CET; answers land inside one working day
Talking it through
A call on Teams or Google Meet, whenever writing is not enough

We set strictly necessary cookies only: they keep the site running and remember the city you chose. Nothing here is used for advertising or tracking. More in our privacy policy.