Australian cloud for business software — built and run in Sydney since 2002

NetLeverage
NetLeverage

Business continuity, disaster recovery and incident response plans: which document answers which question

Technical article
Backup & Recovery

About this article: This article is reviewed against the cited sources. This is source research, not lab validation. Check the stated versions and your own requirements before making changes.

You need to be able to answer three different questions in a bad hour, and each is answered by a different kind of plan. A business continuity plan (BCP) answers: how does the business keep doing its important work while something is wrong? A disaster recovery plan (DRP) answers: how do we get the computer systems back, and where? An incident response plan (IRP) answers: what do we do when someone attacks or compromises our systems, and who decides?

That does not mean three thick binders. For a small or mid-sized business, our reading of the sources below is that you need three questions answered clearly and owned by named people. They can live in one document with three sections, as long as each has its own trigger, owner and contact list. This guide explains how the two main NIST publications separate the plans, where they overlap, and how to check what you already hold.

Why it matters

Have a plan is advice that hides a choice. If you have a thorough plan for restoring servers after a flood but no plan for what staff do while the servers are down, you have answered one question and missed another. If you have a plan for restoring from backup but nothing about deciding whether the attacker is still inside the network, you may restore into the same problem.

NIST's contingency planning guide says the plans are "often confused in their use". It also says "universally accepted definitions" for these planning areas "have not been available", so the scope of real plans "may vary" from its descriptions. (NIST SP 800-34 Rev. 1, chapter 2.) That is a reason to define the terms inside your own documents, not to assume a supplier means the same thing by them.

The three plans in plain terms

NIST SP 800-34 Rev. 1 is the older and broader guide. Its chapter 2 separates continuity from contingency from incident response in one passage: "Continuity planning normally applies to the mission/business itself", while "Contingency planning normally applies to information systems", and "Cyber Incident Response Planning is a type of plan that normally focuses on detection, response, and recovery to a computer security incident or event."

Business continuity plan. NIST says "The BCP focuses on sustaining an organization's mission/business processes during and after a disruption", with payroll and customer service given as examples of such processes. It adds that because those processes depend on information systems, the business continuity planner "must coordinate with information system owners". In everyday terms: which activities must carry on, in what reduced form, using what workarounds, and who tells customers and staff.

Disaster recovery plan. NIST's description is specific and narrower than many people expect. The DRP "applies to major, usually physical disruptions to service that deny access to the primary facility infrastructure for an extended period", and "only addresses information system disruptions that require relocation." In other words, it is about restoring systems at an alternate site after a facility-level loss. NIST distinguishes it from a plan to recover a single system wherever it runs, which it calls an information system contingency plan. That plan contains "roles and responsibilities, inventory information, assessment procedures, detailed recovery procedures, and testing of a system."

Incident response plan. NIST's cyber incident response plan "establishes procedures to address cyber attacks against an organization's information system(s)", covering unauthorised access, denial of service and malicious software. The newer NIST SP 800-61 Rev. 3 (April 2025) describes it as "the roadmap for implementing the incident response capability". It defines an incident as an occurrence that "actually or imminently jeopardizes, without lawful authority, the integrity, confidentiality, or availability of information or an information system". (We have quoted the definition as NIST spells it.) Note the boundary: the guide "addresses only adverse cybersecurity events", so a power failure or flood sits with the other two plans.

What starts each plan

A useful way to tell the documents apart is the trigger: the event that makes someone pick the plan up.

Plan Typical trigger (our illustration) First question it answers Likely owner (our recommendation)
Business continuity Any disruption that stops an important activity, whatever the cause Which activities must continue, and how? A senior business leader, with each process owner
Disaster recovery A major loss of systems or the site that holds them How do we restore systems, and where? The person responsible for IT, with the service provider
Incident response A suspected or confirmed security event Is this an incident, how bad is it, and who decides what to isolate or shut down? A named incident lead with a leadership decision-maker

One event can start more than one plan. A ransomware attack is an incident, so the response plan applies. If it takes the systems down, the recovery plan applies. If staff cannot work, the continuity plan applies. NIST SP 800-61 Rev. 3 captures the link in one recommendation: "Synchronize business continuity plans with incident response plans since incidents can undermine business resilience." (Rev. 3, recommendation R1 under control ID.IM-04.)

Do you need three documents?

Not necessarily. The sources allow for a compact arrangement.

  • NIST SP 800-34 says of the incident response plan: "This plan may be included as an appendix of the BCP."
  • AWS, writing about recovery in the cloud, says a disaster recovery plan "should be a subset of your organization's business continuity plan" (its wording) and "should not be a standalone document." (AWS, Business Continuity Plan (BCP).)

Taken together, our reading is that one coordinated document with three clearly separated sections is consistent with both. The things that should not be merged are the triggers, the owners and the contact lists, because each plan is picked up by different people in different circumstances. Our recommendation is to keep the sections separable so each can be printed, tested and updated on its own timetable.

NIST SP 800-61 Rev. 3 also lists "vulnerability management plans" among the plans relevant to incident response, and asks that plans be based on "the organization's unique requirements, mission, size, structure, and functions." A business with a handful of staff and one hosted system needs shorter plans than one with several sites. The questions stay the same.

Who owns each question

In our view, ownership is the part of a plan most worth settling explicitly, so it deserves its own decision.

NIST SP 800-61 Rev. 3 lists the people involved in response, and the list is wider than IT. It names leadership, who "may have decision-making authority on high-impact response actions, such as shutting down or rebuilding critical services"; incident handlers, who may be on staff, on contract or "available when needed"; technology professionals; legal experts, who can review plans and contracts; public affairs; and human resources. For a small business, our reading is that several of these roles fall to the same two or three people, but the roles still need to be written down.

The guide also says plans should settle "which roles have the authority to confiscate, disconnect, or shut down technology assets." Our recommendation: the most valuable single line in an incident response plan is the name of the person who can decide to take a system offline at three in the morning, and the name of their backup.

For a hosted service, add the provider. The Rev. 3 guide says incident handlers may include "a cloud service provider's incident response team when an incident occurs within that provider's cloud". Your plans should therefore record who contacts whom, and by what route, if the event is on your side of the service versus your provider's side.

Questions to put to your provider

If you rely on a hosted or managed service, the plans above depend on what your provider will and will not do. Ask for the answers in writing.

  • Which parts of my continuity, recovery and response plans do you take part in, and which are mine?
  • How will you tell me about an incident, how fast, and through which contact? Ask who on your side receives it and what they are expected to do.
  • Who can authorise isolating or shutting down my systems, and how do I reach them out of hours?
  • What recovery steps do you carry out, and what must I carry out? Ask whether your recovery depends on staff or systems of yours that would also be affected.
  • How and when have you tested these arrangements with customers? Ask for what was found, not just that it was done.

Checking the plans you already hold

This check reads documents only and changes nothing.

  1. Collect every document that might count: an IT disaster plan, a security policy, an emergency contacts sheet, a supplier agreement, a staff handbook page.
  2. Sort each against the three questions. Mark which question it answers, not what it is called.
  3. Look for gaps. Is there an answer to what staff do during an outage? To who decides to disconnect a system? To where the systems are restored?
  4. Check each for an owner, a date and a trigger. A plan with no named owner or review date is a draft.
  5. Check that the contact details work. NIST SP 800-34 recommends training and testing, including that training "be provided at least annually". Our reading is that an annual walk-through of all three questions is a reasonable minimum for a small business. NIST also says a plan should be "reviewed for accuracy and completeness at an organization-defined frequency or whenever significant changes occur to any element of the plan", and that contact lists "will require more frequent reviews". So review sooner after a material change, such as a new system or supplier, a staff change in a named role, or a move of premises.

A worked example (illustrative)

The following is invented. A firm of twenty has one document called Disaster Plan, written by its IT contractor. It lists server restore steps and a backup schedule. Sorted against the three questions, it answers the recovery question only. Nothing says what staff do for a day without email, and nothing says who decides whether to disconnect the network if a staff member reports odd behaviour.

The owner adds two short sections to the same document. The continuity section lists the three activities the firm must keep running, a workaround for each, and who phones the largest customers. The response section names an incident lead, a backup, a leadership decision-maker and the contractor's emergency number, and states who can order a system offline. The contractor's recovery section is left as it was but gains a trigger line. The firm has three questions answered in one document, with three owners.

Risks and limits

  • NIST's guides are written for United States federal systems. Their definitions are useful vocabulary, but this article is not a compliance guide and does not cite an Australian rule or standard. If a regulation, contract or insurer sets plan requirements for you, those requirements come first.
  • NIST SP 800-34 dates from 2010, so some of its examples, such as alternate sites and tape, reflect older practice. We have used it for its distinctions between plan types, not for its technical examples.
  • NIST SP 800-61 Rev. 3 addresses cybersecurity incidents. It does not give guidance on floods, power failures or other non-cyber disruptions.
  • A plan that has never been exercised is untested. The sources recommend testing; this article has not tested any plan.
  • Recommendations on owners and triggers marked as ours are our reading, not NIST requirements.

Next step

List the documents you hold and sort each against the three questions, then name an owner for any question with no answer. If you are reviewing a hosted service, the Managed Services page is a starting point for that conversation; ask for the scope in writing.

Sources

All accessed 15 November 2026.

Learning notes

  • Continuity is about the work, disaster recovery is about the systems, and incident response is about the attack.
  • One event can start all three plans, so keep the triggers and owners distinct but the documents aligned.
  • A small business can hold all three in one document if each section can be printed and tested alone.
  • The most useful single line in a response plan names who can take a system offline, and their backup.