דילוג לתוכן
Articles and Knowledge Centre / Backup and Recovery

Building a resilient backup policy: from critical data to verification

A good backup policy begins with a simple question: if something stops working, what does the client need back — and how quickly? For managed IT providers, information protection and business continuity relate to infrastructure, to revenue, and to customer trust. Here are five steps to building…

12.10.2026 · 6-minute read
Illustration on building a resilient backup policy: from critical data to testing.

Good backup policy starts with a simple question: if something stops working, what does the customer need back — and how quickly? For managed IT providers, information protection and business continuity are connected to infrastructure, revenue, and customer trust. Here are five steps to build a policy that not only looks organized but helps on the day it’s needed.

Plan according to the business

  • Important information and services
  • Time objectives and data loss
  • Recovery tests
Key principles at a glance

What really must be restored to operation?

Divide the information into groups: system information, application settings, and operational information. Now ask what is most important for the business operation and how quickly each group needs to be restored. Not every folder is urgent to the same degree — even if someone named it 'Extremely Important Final'.

The source suggests handling user information and active, or 'hot,' data on an hourly or daily basis, and content that changes less, for example part of the website content, on a less frequent basis. Record the information, and review and update the list at regular intervals. This way you can ensure that the information you need is indeed included in routine backups.

Decide how much time and how much information can be lost

Set objectives according to the business's activity, values, and goals. RTO, the recovery time objective, defines how quickly an infrastructure or service must be restored. RPO, the recovery point objective, defines the acceptable data loss window for the business.

To understand the difference between RPO and RTO and to set appropriate objectives, examine three things:

  • The cost of downtime for the business.
  • The budget available to you.
  • The actions and resources required for disaster recovery.

This examination will help determine time limits and appropriate recovery procedures, as well as the frequency at which data needs to be backed up. It is also important to understand the difference between backup and disaster recovery to complete the picture.

Choose a method according to the need, not according to the title

Choose a solution that fits the needs and IT environment. Options include full, incremental, and differential backups. A cloud management console can help the team view and test the environment remotely. Before choosing, it is important to understand how each option serves the required recovery.

When checking backup methods, follow the 3-2-1 rule: keep three copies of the data, meaning the original data and two backup copies, on two types of media, with one copy stored offsite for disaster recovery purposes.

Turn decisions into work rules

Create a backup policy that supports the success of the process. The backup should meet your service level requirements and also include previous versions of the data. The source suggests backing up critical information in real-time or daily, and other information at least once a week. These examples should be adapted to your business objectives and actual risk.

Also prepare recovery scenario manuals, with actions that allow you to restore according to the established expectations. Implement the policy in the infrastructure, train the staff, and restrict changes to authorized personnel. Keep suitable copies of the backup settings: during an event, it is better to follow a plan rather than try to remember where each setting was.

Make sure the plan works in reality

Check the backup target, whether local or in the cloud, and the results of the tasks. If something does not meet expectations, return to the settings and look for configuration issues that prevent backups from being created. A green message is a good sign; checking what is saved gives a better picture.

As part of backup tests, check the information, IT infrastructure, and security settings. Also check the systems in relation to recovery time objectives and recovery point objectives. If a system change is required, start gradually. Check each change, and over time look for ways to improve backup systems and develop alternative scenarios for a more resilient policy.

Start with the first step, and check along the way

There is no need to wait for an event to start. Use the five steps to protect business operations and customer information, today and in the future. A resilience policy is built from planning, implementation, and ongoing improvement — not from a one-time decision placed in a folder.

Practical recommendations for backing up virtual machines

Backup is evaluated by the ability to return to work, not just by a success mark.

The process at a glance
3 copies of data01

2 types of media02

1 offsite copy03
Original illustration of the process — not a photo of the product interface.

Work stages

  1. Independent copy

    Snapshot can help with a short-term change, but it does not replace a separate backup from the production environment.

  2. Copy Planning

    Implement the 3-2-1 policy: source and two backup copies, on two types of media, and at least one copy offsite.

  3. Information Protection

    Enable encryption and manage its keys. Adjust frequency and retention to the amount of data that can be lost and the required recovery time.

  4. Efficiency and Consistency

    Use CBT in VMware when it is supported. For applications and databases, check application-consistent backup, licensing, and compliance.

  5. Recovery Practice

    Regularly perform machine and item restore on a test server, document the result, and address failures.

How do you know when we’re done?

  • The separate copy is accessible in case of failure.
  • Backup keys are available to authorized personnel.
  • Files, applications, and recovery times have been tested.
Sources and Versions

The original manufacturer’s guide

Continue reading