A disaster recovery plan is not a document that proves you have a backup. It is an agreed way to restore the services that matter, in a known order, with people who know what to do. Many plans fail not because the technology is missing, but because the plan was never exercised.

Key takeaways

  • Start with business impact, not with the backup software.
  • Recovery objectives should be agreed in plain language and tested.
  • Dependencies, roles and communication matter as much as restoration.
  • A plan that is not exercised is a hypothesis, not a capability.

Start with what the organisation cannot do without

List the business processes that must continue: paying staff, taking orders, delivering a service, communicating with customers, meeting a regulatory obligation. For each one, identify the systems and data it depends on.

This conversation belongs to the business, not only to IT. A technology team can restore a server; only the business can say whether the service it supports is critical, and for how long it can be unavailable.

Define recovery objectives in plain terms

Two questions are useful. How much data can the organisation afford to lose, measured in time? How long can the service be unavailable before the impact becomes unacceptable? These become the recovery point objective and recovery time objective.

Avoid copying generic targets from a template. A payroll system and a meeting-room booking tool do not need the same treatment. The targets should reflect the actual impact and the cost of achieving them.

Map the dependencies

A business system depends on more than its server: network links, domain services, storage, backup, power, certificates, licences, third-party APIs and people. A recovery plan that restores the server but not the identity service will not deliver a working service.

Document the dependencies and the order in which they must be restored. This becomes the recovery sequence, and it is usually more useful than a long inventory.

Design backup and replication around recovery

Backup frequency, retention and storage should follow the recovery objectives. If the organisation cannot lose more than an hour of data, a nightly backup is not sufficient. If a site loss must be survived, a copy in the same room is not sufficient.

Protect the backup environment separately. Use different credentials, restrict access and keep at least one copy that cannot be deleted by a compromised production account. Test restores on a schedule and record the results.

Write runbooks people can follow

A runbook should state the trigger, the decision-maker, the sequence of actions, the expected result and the escalation point. It should include contact routes and the location of configuration records, licences and recovery media.

Keep it short enough to use under pressure. A 90-page document that nobody opens during an incident is less useful than a two-page guide for the first hour and a detailed appendix for later.

Assign roles and communication

Someone must decide whether to invoke the plan, someone must lead the technical recovery, and someone must communicate with staff, customers and suppliers. These roles should be named and have a deputy.

Communication is often the weakest part of a recovery. Prepare holding statements, agree who can approve them and decide how staff will be told what to do if email or the office is unavailable.

Test, record and update

Testing can start with a tabletop exercise: walk through a scenario and ask what each person would do. It is inexpensive and exposes assumptions quickly. Technical tests can follow, beginning with a non-critical system.

Record what worked, what failed and what changed as a result. Update the plan, the dependencies and the training. A plan that is reviewed once a year but never exercised will drift away from the real environment.

Common reasons plans fail

The plan covers technology but not people. Backups exist but restores are untested. The recovery sequence is wrong. Contact details are out of date. Nobody has authority to invoke the plan. The plan assumes a site that is also affected.

Most of these are fixable without buying anything. They require agreement, documentation and practice.

Note: Recovery objectives and arrangements depend on the organisation, its obligations and its budget. This guide provides a planning approach, not a guarantee of recovery or a substitute for professional advice.

Need a second opinion?

Tell us what you are trying to fix, build or buy. We will give you a clear view of the options and the next step.

Contact Stevicom