Service
Backup for businesses
A backup nobody has ever restored is not a backup, it is a large folder. The question that matters is not whether backups run, it is how long it would take you to start invoicing again if a server were encrypted tomorrow. That can be measured, and it can be reduced.
The problem
What we find when we look
- Backups that have been failing for months, warning an inbox nobody reads
- Backups on a disk attached to the same server: if the attack reaches the server, it reaches the backup
- No backup of Microsoft 365 email or files, because the cloud is assumed to handle it
- The data backed up but not the configuration: the files are there and the system has to be rebuilt from scratch
- Nobody knows how long a restore takes, because it has never been tried
- Not a single test restore in years
Scope
What is in, and what is not.
What we cover
- Physical and virtual servers, data and system
- Workstations that hold local information
- Microsoft 365 and Google Workspace: email, files and sites
- An off-site copy, isolated from the production network
- An agreed retention period, so you can go back to a point before the problem and not just to yesterday
- Periodic restore tests, with the result documented
- A recovery plan: what gets brought back in which order, and who does what
Decisions to make first
- How much information you can afford to lose: an hour's work, a day? That sets how often it copies
- How long you can be stopped: that drives the recovery design, and it is what drives the cost
- What is genuinely critical: backing everything up at the same priority is expensive and slows down recovering what matters
- Recovering data from a failed disk with no prior backup is a different, laboratory service, with a different cost and no guarantee of result
How it works
How it gets built
What needs saving
An inventory of data and systems, and where each thing lives. There is always some important piece of data somewhere nobody was backing up.
Objectives in writing
How much loss is acceptable and how much downtime. Without those two numbers nothing can be designed, only bought.
An isolated copy
At least one copy out of reach of the production network. That is the difference between an incident and a closure.
A real restore test
We restore for real and time it. That number — how long it took — is the only reliable evidence that the plan works.
Service area
Scope
Design, deployment and monitoring are remote, so the service does not depend on location.
On-site work appears when there is physical equipment to install, or a recovery to run on hardware in the office, within the Baix Llobregat and Barcelona area.
Questions
Frequently asked questions.
- Microsoft 365 already keeps my email. Why another copy?
- The recycle bin and the service's retention protect against accidental deletion for a limited time. They do not protect against a mass deletion carried out with valid credentials, an account closure, or a mistake discovered months later. They are different things.
- How often is the restore tested?
- The frequency is agreed according to how critical the system is. What is not negotiable is that it gets tested: a backup with no verified restore is an assumption.
- Where is the data stored?
- In the location you agree, and it gets documented: that is information you need to hold under the GDPR, not just for peace of mind.
How long until you could invoice again?
If you don't have the number, it is worth finding out. Reviewing your current backups is free.
Talk to a technicianYou may also need
Related services.
- CybersecurityEmail protection, endpoint, firewall, audits and compliance for smaller companies.
- Cloud & AzureManaged servers and cloud migration when the numbers add up, not because it is fashionable.
- IT maintenanceA fixed monthly fee: support included, preventive reviews and a grip on what you have.
Would you rather talk it through? Call 622 317 788. You get the same person who will do the work.