GUIDE · CONTINUITY AND DISASTER RECOVERY

IT disaster recovery plan: structure and template

After this guide you will know which sections an IT disaster recovery plan must have, which information to collect before writing it and how to make it usable by a technician in an emergency. The document must work even when the person who wrote it is not available.

Updated october 2026
8 min read
Altovar Technical Editorial Team

1. Why a dedicated document is needed

The IT disaster recovery plan describes how to restore servers, data, applications and connections after a serious event. It is the document the technician opens when they have little time and a system does not respond. If it is vague, every restore turns into a search across different tools and personal memories. For this reason it must be readable in a few minutes, even by someone who did not write it.

The document should be kept distinct from the business continuity plan, which decides what to do for the business as a whole. The disaster recovery plan instead describes how to bring the infrastructure back into operation, in order and with verifiable steps. The second must respect the priorities set by the first. For this reason it is best to have them approved together and kept consistent over time, so that they do not tell different stories.

2. Prerequisites before writing

Some elements are needed before you write. The most important is the list of critical processes with their priorities, that is, the output of the business impact analysis. Without it you do not know which system to restore first. You also need an up-to-date inventory of systems, with versions, owners and location. Finally, a map of how they are connected to each other, because a server restored without network or identity is of no use.

Then check that backup copies exist and that the credentials needed for recovery are stored securely and accessible even in an emergency. A plan that requires a password kept only on the system that will not start is useless. Identify who has the authority to declare an emergency and who replaces them if they cannot be reached. Without these names, nobody knows whose decision it is when it really matters.

3. How to write it: the steps

Work in this order, because each step produces a part that the next one uses. Skipping a step forces you to rewrite later. The work is done with the IT manager, a management contact and whoever knows the systems in detail. Complete the steps in sequence even when some parts seem urgent, because badly managed urgency produces a fragile document:

Write the procedures so that they can be followed by a competent person who did not design them, perhaps at night. Use imperative sentences, state the expected result of each step and indicate the point where one should stop and ask for confirmation. Avoid references such as “as usual”, because in an emergency there is no usual. The reader cannot ask for explanations at the moment the procedure is needed.

  1. Appoint the owner of the document and the group that writes it
  2. List the systems covered, starting from the approved impact analysis
  3. For each system, record owner, location, dependencies and the backup copy used
  4. Write the criteria that trigger an emergency declaration and who makes it
  5. Describe the recovery procedure for each priority system, step by step
  6. Define the restore order and the checks to carry out before reopening the service
  7. Collect emergency contacts, with substitutes and alternative channels
  8. Have the document approved and keep offline copies

4. Recommended document structure

A clear structure separates what needs deciding from what needs executing. The decision-making part stays short and is approved by management. The operational part is designed for the technician and can be more detailed. Appendices, such as the inventory and network diagrams, should be kept separate, so that each reader can immediately find what they need. Keep a readable offline version as well.

  • General section: purpose, scope, version and approval
  • Decision section: activation criteria, roles and powers
  • Operational section: procedures per system and restore order
  • Appendices: inventory, network diagrams, contacts and evidence log

5. Common mistakes

The most frequent mistake is writing the plan as if the system were still available. Procedures that begin with “log in to the control panel” without saying with which credentials are of no use in an emergency. Another mistake is not stating the restore order, and the result can be a ready system that depends on another one still switched off. Each technician restarts what they know best, and this does not always coincide with what is needed first.

Beware also of documents that cannot be reached: a plan kept only on the network that has gone down is of no use to anyone. Many plans do not say who has the right to decide on the return to normal, and it is precisely at that moment that the most serious conflicts between departments arise. Define this point in writing before it is needed. A mistake that repeats across several sections should be treated as a sign of a structural issue, not as an isolated oversight.

6. How to check that the document works

A document is verified with a test, not with a re-reading. The simplest way is a paper test: choose a scenario, such as the loss of a file server, and have a person who did not write the procedure follow it to the end. Every unclear or non-executable step should be noted as a correction, with an owner and a date.

The next step is a technical restore test on a non-production system, in which the actual time is measured and compared with the approved objective. If the time is longer, the difference must be explained: either the procedure must be corrected or the objective must be revised with management. Keep the result alongside the procedure, so that the next test starts from real data.

7. What to do after the first draft

After the first version, the document should be linked to the testing calendar and the update process. Every significant change to the systems, such as a migration or a new supplier, must trigger a review of the affected procedure. There is no need to rewrite everything: often it is enough to update the system card and its position in the restore order. Whoever reviews must have access to the current version and the previous one.

8. How to start

This week, create a shared folder, with an offline copy, containing three files: the inventory of priority systems, the list of emergency contacts and an empty draft of the document. Complete the inventory first, because it is the data that unlocks everything else. The first file will tell you which systems matter, the second whom to ask for help, and the third which structure to follow.

Then arrange a short meeting with the IT manager and ask them to state, for each system, where it is located, who manages it and which backup copy protects it. If the answer to one of these questions does not exist, that is the first correction to make, before even writing the procedures. Do not try to complete everything this week: the aim is to have a solid base to work from.

Frequently Asked Questions

Is an IT disaster recovery plan mandatory?

It depends on the sector, size and type of services offered. Some regulations, such as the NIS2 Directive transposed in Italy by D.Lgs. 138/2024, require risk management measures that include recovery after an incident. Even without specific obligations, a recovery document prevents a failure from becoming an emergency without rules. Check with an adviser which obligations apply to you.

How detailed should the document be?

Detailed enough to be followed by a competent technician who did not write it, even under pressure. For the most critical systems precise steps are needed, with credentials kept elsewhere. For secondary systems, a brief card and a reference to the supplier's documentation are enough. More detail does not necessarily mean a better document.

Who should have a copy of the plan?

At least the people who can declare an emergency, the technicians assigned to recovery and their substitutes. An offline copy, printed or on separate media, must be available even if the network is inaccessible. Mind confidentiality: the document contains information about the structure of the infrastructure, so distribution should be limited to those who need it.

Can the IT supplier write the plan for me?

It can draft the technical part and propose a structure, but decisions on priorities and roles remain with the company. The supplier knows the systems it manages, but does not always know which processes matter most to you. Ask that the draft be reviewed with internal managers and approved by management, otherwise it will not truly be your plan.

How often should the document be updated?

It should be reviewed at least once a year and every time systems, networks, suppliers or people in key roles change significantly. After each real test or restore, it should be updated with what emerged. The system cards are the part that changes most often, so it is worth keeping them separate from the main text to update them without effort.

About this content

General informational guide. Regulations and technical requirements change over time: always check the official sources and assess your specific case with a qualified technical or legal advisor.

GET IN TOUCH

Talk to our team.

We'll show how Altovar integrates into your infrastructure. No commitment, no sales pitch. Just a technical conversation.

BOOK A DEMO →