Solution · By need

Downtime prevention

Outages are not a question of if. A drive gives up, a road crew cuts the fibre, a colleague clicks a booby-trapped invoice. The figure that counts is how quickly invoices go out and orders come in again. Our first conversation therefore covers three points: the longest each system may stay offline, the amount of recent work you could bear to re-enter, and who takes charge during the opening half hour. Backups, alerting and spare capacity are designed around those answers rather than bought first and justified later.

3-2-1
copies on separate media, one held off-site
100 %
remote delivery, no site visits
24/7
alert handling with Premium
15 min
Premium response time

Building blocks

Six components. The first one is a spreadsheet of targets; the remaining five are sized by it. That keeps spending tied to real business impact instead of to whatever a vendor happens to sell.

Book a call with an engineer

Recovery targets per system

ERP, mail, online shop, file share: each gets a recovery time objective (RTO) and a recovery point objective (RPO). Those two numbers drive every technical choice that follows.

3-2-1 backup regime

Scheduled jobs with retained versions, an off-site copy and a locked, write-once copy that neither users nor malware can erase. At intervals agreed in the contract we pull a sample back into a sandbox to prove it works.

Monitoring

We watch servers, the Microsoft 365 tenant, connectivity, storage headroom, expiring certificates and critical services. Premium customers get alerts acted on day and night, so the fault is often fixed before the first complaint.

Spare capacity

A backup connection with another carrier that takes over automatically, and where the business case exists, a standby copy of key servers hosted in a data centre that steps in if the primary goes silent.

Electricity and failing parts

Your UPS is set up to power servers off gracefully when mains supply drops, and its battery condition is tracked. If a disk or battery must be swapped, a colleague of yours or a service technician does it while we coach them over a video call.

Business continuity plan

A short written playbook: restart sequence, named owners, who to contact, paper workarounds for the meantime and a holding message for customers while systems are unavailable.

What we guard against

Five common scenarios. Each one needs a dedicated safeguard, and covering one does nothing for the rest.

Dead disk or server

Safeguard: backup and a standby copy. It happens most often and is the easiest to prepare for. The manufacturer's service or your own person replaces the hardware; getting data and software running again is our job.

Ransomware

Safeguard: a copy out of the intruder's reach. Modern attacks deliberately seek out backup repositories and scramble them too. A write-once copy, guarded by its own credentials and MFA, survives that.

Something deleted by mistake

Safeguard: generous retention. Nobody notices the vanished contracts folder in SharePoint until weeks have passed, and a backup that only goes back a few days cannot help.

No connectivity

Safeguard: a different provider. Two circuits bought from one operator usually run through the same duct and fail at the same moment. An LTE or 5G router is plenty as a spare for most offices.

Mains failure

Safeguard: UPS with a clean power-off. Pulling the plug on a running accounting database can do more lasting harm than the interruption itself.

Until someone has restored from it, a backup is only a hope. Typical surprises include jobs that have silently failed for months, a schedule pointed at an empty directory, or an encryption password that left with a former employee. A test recovery replaces guesswork with facts.

Questions and answers

Running is not the same as recoverable. Ask instead when a restore was last attempted and how many hours it took. We offer a simple exercise: your most recent copy is brought back in a walled-off sandbox with a stopwatch running. Either you gain confidence, or you learn where the chain breaks, and each result is useful.

It follows from your targets. Being operational again by the next working day costs little. Being back inside sixty minutes calls for a warm standby in another facility, which is a bigger commitment. We therefore price each system's acceptable outage first, and often find that some systems are well served by the modest tier. A single recovery test is available at PLN 190/hour excl. VAT.

An Apply engineer connecting remotely, inside the response window of your plan: Premium 15 minutes, Business 1 hour, Start 4 hours. Adding the 24/7 IT support service keeps someone on call overnight and at weekends. Where physical work is unavoidable, a replacement power supply for instance, we direct one of your people through it. The plan is phrased for non-specialists and a copy is kept by you.

That is what the regular drills do: on a schedule set in the contract, selected data comes back in a sealed-off sandbox and the duration goes into a log. Annually it is worth gathering directors and key staff around a table to walk through something like “Monday 7:00, the server shows a ransom note” and see where the written plan and actual behaviour diverge.

It may. NIS2 and the amended Polish act on the national cybersecurity system cover, among other things, incident handling, backups and business continuity. Whether your company is in scope is worth checking with a lawyer. If it is, the continuity plan and the drill log become useful parts of your documentation.

Is your company ready for an outage?

We examine how your backups are configured, run a real restore of one copy and time how long each system actually needs to return.

Hours
Mon-Fri 8:00-18:00 CET, reply within one working day
Meetings
Online via Teams or Google Meet

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.