
Data backup and recovery
Backups you can actually recover from: a secure 3-2-1 architecture, automated backups, monitoring and regular recovery testing.
Most organisations believe they have backups. Yet at the critical moment there is often no certainty that the data can actually be restored. The difference between having backups and being truly protected only becomes visible during an incident.
Data doesn’t disappear — it is lost the moment it can no longer be restored.
In 2023–2025 a number of European companies were hit by ransomware attacks, such as the LockBit, BlackCat and RansomHub campaigns. A typical scenario:
- the attacker gains access (phishing, RDP, a vulnerability or applications infected with an infostealer);
- explores the infrastructure and moves through it (lateral movement);
- deletes or encrypts the backups;
- triggers the encryption.
The result: on paper, the company had requirements and backup mechanisms in place, but:
- the backup system was joined to the domain;
- the backups could be overwritten;
- they were encrypted together with the rest of the infrastructure;
- the system had been set up long ago, was already legacy and contained publicly known vulnerabilities.
Data backup and recovery
Today’s organisations depend directly on the availability and integrity of their data. Data loss — whatever the cause — can affect not only IT, but service delivery, reputation and the organisation’s ability to keep operating.
The job of a backup system is not just to keep copies. Its purpose is to make it possible to fully restore data and systems after an incident. It is the foundation of organisational resilience.
Backups are not the whole story
Backups help restore data, but on their own they are not enough. A complete approach covers data recovery, system recovery and process recovery. A backup is a critical element, but only one part of overall resilience.
What creates the biggest risk in practice
The biggest problem is usually not the absence of backups, but their quality:
- only data is copied, not the system as a whole;
- configurations and automation components are not included;
- no recovery process is defined;
- recovery is not tested regularly;
- the backup software is not kept up to date;
- unsuitable solutions without additional safeguards are used — backups can be overwritten or corrupted.
A backup that cannot be restored is, in practice, not a backup.
What a “complete backup” really means
Files alone are not enough to restore a system. In practice, backups must include:
- user and business data;
- system configuration data;
- software and its components;
- automation scripts;
- log files and technical information.
Only this approach makes it possible to restore a system in full. For example, if the backups contain no log files, then after a cyberattack the investigation may struggle to establish the real root cause and to confirm that the system is no longer compromised and the attackers have no remaining way back in.
A company made regular file backups. During an incident the data was restored — but the system didn’t work.
Why? Configurations, service dependencies and automation scripts had not been backed up.
The result: recovery took several days and business operations were paralysed.
How to build a secure backup architecture
An effective approach is based on separating data and layering protection. Backups must be technically and logically separated from the primary system, stored in a geographically separate location and protected against both physical and cyber threats. This prevents a single incident from affecting the entire infrastructure.
The 3-2-1 rule
A widely used data protection model — the minimum needed to reduce the risk of losing everything at once:
Types of backup
- full backups — simple to restore;
- incremental — efficient, but more complex;
- differential — a balance between the two;
- snapshots — a quick capture of state, but not a complete backup.
In practice several types are combined.
Where to store backups
- local copies — fast, but vulnerable;
- remote copies — protect against physical incidents;
- cloud copies — flexible, but security and compliance must be assessed;
- isolated (air-gapped) copies — the highest level of protection.
Today a hybrid approach is used.
RPO and RTO: where reality begins
Backup is not just a technical process. It rests on two key parameters that determine how often copies must be made and how quickly systems must be restored:
Security principles
An effective backup system rests on three elements — if any of them is missing, the system cannot be trusted:
- confidentiality — access control, encryption;
- integrity — data cannot be altered;
- availability — recovery is genuinely possible.
An automated, managed approach
In a modern environment, backup is a centralised and automated process. In practice VersaIntel uses the NinjaOne platform, which provides centralised backup management, automated backups, continuous monitoring, early detection of errors and controlled data recovery.
Backup becomes a managed system, not a manual task.
The VersaIntel approach
We don’t look at backup in isolation — it is part of your overall IT and cybersecurity architecture:
- critical data is identified;
- a backup strategy is designed;
- a hybrid architecture is implemented;
- recovery tests are automated and monitored;
- continuous monitoring is provided through sensors and specialised solutions.
The focus is not just on copying. The focus is on being able to actually create the backups and fully recover from them.
Not relying on luck, but staying in control.
If you have even the slightest doubt whether your backups really work, book a consultation. VersaIntel will assess your situation and give you a clear picture of how quickly your data could be restored in practice.
Book a consultationThe result: your organisation
Frequently asked questions about data backup
01What is the 3-2-1 rule?
A widely used data protection model: at least 3 copies of the data, on at least 2 different types of media, with at least 1 copy outside the primary infrastructure. It is the minimum needed to reduce the risk of losing everything at once.
02What are RPO and RTO?
RPO is how much data the organisation can afford to lose; RTO is how quickly operations must be restored. Together they determine how often backups must be made and how quickly systems must be recoverable.
03Are file backups enough?
No. To restore a system in full, backups must also include system configurations, software and its components, automation scripts, and log files and technical information. Without them the data may come back, but the system won’t work.
04How do you protect backups from ransomware?
Backups must be technically and logically separated from the primary system, stored in a geographically separate location and protected against being overwritten. Isolated (air-gapped) copies give the highest level of protection, and the backup system itself must be kept up to date.
05How can we be sure the backups can actually be restored?
Through regular recovery testing. VersaIntel automates and monitors recovery tests and, using the NinjaOne platform, provides continuous monitoring and early detection of errors.
Full portfolio of cybersecurity services
NIS2 and NKDL compliance
Who Latvia’s National Cybersecurity Law and NIS2 apply to, what the obligations are — and how to implement them so they genuinely strengthen your security.
Learn more 02
Cybersecurity management (CISOaaS)
Chief Information Security Officer as a service: cyber risk management, security policies, ISO 27001, NIS2 and NKDL compliance.
Learn more 03
Cybersecurity audit
A cybersecurity audit identifies vulnerabilities, assesses cyber risks and delivers prioritised security improvements based on international methodologies.
Learn more 04
Security Operations Center (SOC 24/7/365)
Versa SOC — 24/7/365 cybersecurity monitoring, threat intelligence and incident response.
Learn more 05
IT platform security
Secure IT platforms from the ground up: secure architecture and configuration of servers, cloud platforms, networks and Microsoft environments, plus business continuity.
Learn more 06
Cyber incident investigation
Cyber incident forensics (DFIR): an evidence-based understanding of what happened, how and with what impact — and clear next steps.
Learn more 07
Microsoft 365 services for businesses
Microsoft 365 licensing, implementation, configuration, Defender security and maintenance — from a Microsoft partner.
Learn more 08IT infrastructure security
Managed 24/7 threat protection: endpoint protection, EDR and a SOC in one system that spots attacks early and stops them before they affect the business.
Learn more 010
Versa GRC — compliance platform
Versa GRC: centralised management of risks, incidents, assets and compliance in line with NIS2, ISO/IEC 27001 and NIST.
Learn moreWe will help you
Write to us with your requirements, and our representative will contact you as soon as possible.


