Glossary
BCP and DRP
A business continuity plan (BCP) describes how an organization keeps its essential activities running during a disruption and then restores them. A disaster recovery plan (DRP) is the part that brings the affected systems back into service, usually at an alternate site.
Updated
What are a business continuity plan and a disaster recovery plan?
A business continuity plan answers an executive's question: if an event takes away a site, a team or our systems, how do we keep serving customers? ISO 22301:2019 defines a business continuity plan as documented information that guides an organization to respond to a disruption and resume, recover and restore the delivery of products and services consistent with its business continuity objectives (definition 3.4).
A disaster recovery plan is narrower. NIST SP 800-34 defines it as a written plan for recovering one or more information systems at an alternate facility in response to a major hardware or software failure or destruction of facilities. In an IT department, the DRP is the document that says how the systems come back.
The same NIST publication defines the business continuity plan in terms close to ISO's: a predetermined set of instructions or procedures that describe how an organization's mission and business processes will be sustained during and after a significant disruption.
IT is only one of the resources continuity planning deals with. The French government's 2013 business continuity guide sorts critical resources into five categories: people, infrastructure, information systems, intellectual resources and external suppliers.
What is the difference between a BCP and a DRP?
Scope and timing separate the two.
The BCP covers the whole organization. It deals with fallback premises, staff, crisis communication, suppliers, and the manual procedures that let people work without IT for a few hours. It applies from the moment the disruption starts, including while the business runs in degraded mode.
The DRP covers bringing the technical means back into service. It says which site to restart on, from which backups, in what order, and who confirms that each application is usable again. It is triggered once the disaster is declared and ends with the return to normal operations.
Each needs the other. A DRP without a BCP restarts servers without knowing which ones matter most to the business. A BCP without a DRP promises recovery times that nobody has checked technically.
In a mid-sized organization, ownership follows the org chart: general management or the risk manager owns the BCP, and the IT department owns the DRP and answers for its recovery times. The business impact analysis (BIA), which ranks activities by priority, connects the two. It is covered in the impact analysis entry.
RTO and RPO: what do these two objectives measure?
RTO and RPO are the two figures that size a disaster recovery plan. The definitions below come from NIST SP 800-34, as carried in the NIST glossary.
The recovery time objective (RTO) is the overall length of time an information system's components can be in the recovery phase before negatively impacting the organization's mission or mission/business processes. It answers the question: how fast must we restart?
The recovery point objective (RPO) is the point in time to which data must be recovered after an outage. It answers a different question: how much data can we afford to lose? A four-hour RPO implies a backup or a replication cycle at least every four hours.
The same publication defines a third notion, maximum tolerable downtime (MTD): the amount of time a mission or business process can be disrupted without causing significant harm to the organization's mission. A system's RTO is therefore set below the MTD of the processes it serves.
These objectives are set per application, starting from business needs. Announcing a four-hour RTO for the entire estate decides nothing: email and the billing system do not have the same urgency or the same standby cost.
Why does a recovery plan fail without an up-to-date dependency map?
A DRP is an ordered list of actions, and the order comes from dependencies. The billing application will not start without its database, which will not start without storage, and nobody logs in without the directory and DNS. Shared services go first, and they are the ones application files leave out, because they belong to none of them.
For each application, the plan has to know what it needs: its machines, its databases, the network flows to reopen at the recovery site, the interfaces with neighboring applications. The French government guide asks planners to record, for each process, its critical resources and its critical interdependencies, upstream, downstream and with partners.
Plans rarely fail on replication technology. They fail on a dependency that appeared after the document was last updated: a new interface server, a firewall rule missing at the recovery site, an application moved to the cloud that the plan still describes on premises.
There are several ways to keep dependencies. A spreadsheet attached to the plan is simple, and frozen at the date of the last exercise. A CMDB holds relationships between configuration items when someone maintains them. Replication and recovery orchestration tools know the machines they protect very well, and the flows and the business side less well. A map of the IT estate derived from the sources brings these together and keeps a date on each link.
How do you test a BCP and a DRP?
A plan that has never been run is a hypothesis. ISO 22301 devotes a clause to an exercise program (8.5). The French government guide presents exercises as a sound way to validate that the plan works, each one designed around what you want to verify.
The guide gives examples, from the lightest to the most demanding. A tabletop exercise checks the alert procedure: who calls whom, on which numbers, at what hour. A simulated exercise activates the crisis team. The technical procedures for switching to backup mode are verified, in the guide's words, by real and periodic execution. A last kind of exercise tests coordination between stakeholders around a simulated major incident.
For a DRP, the test that counts is a real failover of one scope, with users working from the recovery site. You measure the time actually taken against the RTO and the age of the restored data against the RPO. Above all you write down what was missing: the flow that was not opened, the forgotten service account, the server nobody had listed in the plan.
Every exercise ends with an update to the plan and to the list of dependencies.
What does NIS2 say about business continuity?
Directive (EU) 2022/2555, known as NIS2, is addressed to Member States. Article 21(1) requires them to ensure that essential and important entities take appropriate and proportionate technical, operational and organizational measures to manage the risks posed to the security of their network and information systems.
Paragraph 2 lists what those measures must include at least. Point (c) reads: "business continuity, such as backup management and disaster recovery, and crisis management".
The text stops there. It names neither a BCP nor a DRP, mentions neither RTO nor RPO, and sets no recovery time. The detail depends on how each Member State transposes the directive and, for certain categories of entities, on the implementing acts that paragraph 5 assigns to the Commission. In practice, an entity in scope should be able to show that it has addressed continuity, backups, recovery and crisis management in proportion to its risks. The general context of the directive is covered in the NIS2 entry.
What Uncia brings, and what it does not do
Uncia reads your sources read-only (connectors to clouds, hypervisors, CMDB, firewalls and scanners; dropped-in files such as an RVTools export, a flow matrix or an architecture document) and reconciles them into a knowledge base. For a given machine it returns the impact of an outage: the applications and flows that go down. The thread runs from business domain to firewall rule in both directions, and every piece of information cites its source. That is the material a DRP needs to build its dependency list, and to reread it after each collection.
Uncia does not write a BCP or a DRP and does not run one. It replicates nothing, fails nothing over, and sets neither RTO nor RPO: those objectives belong to the business and the IT department. It knows only the dependencies present in the sources it has read, and reports them as undetermined when they are missing. Uncia also produces reports structured along NIS2, ReCyF, DORA or ISO 27001, each section tied to its requirement; a structured report is not the same thing as compliance.
Sources
- ISO 22301:2019, Security and resilience, Business continuity management systems, Requirements: public preview of the standard (definitions 3.3 and 3.4, table of contents), accessed
- NIST CSRC glossary, "recovery time objective", from NIST SP 800-34 Rev. 1, accessed
- NIST CSRC glossary, "recovery point objective", from NIST SP 800-34 Rev. 1, accessed
- NIST CSRC glossary, "disaster recovery plan", from NIST SP 800-34 Rev. 1, accessed
- NIST CSRC glossary, "business continuity plan", from NIST SP 800-34 Rev. 1, accessed
- NIST CSRC glossary, "maximum tolerable downtime", from NIST SP 800-34 Rev. 1, accessed
- SGDSN (French General Secretariat for Defense and National Security), guide to building a business continuity plan, 2013 edition (in French, copy hosted by the French Ministry of Armed Forces), accessed
- Directive (EU) 2022/2555 of December 14, 2022 (NIS2), Article 21, EUR-Lex, accessed
Frequently asked questions
What is the difference between a BCP and a DRP?
The BCP covers the whole organization: ISO 22301:2019 defines it as documented information that guides the response to a disruption and the restoration of products and services. The DRP is narrower: NIST SP 800-34 defines it as a written plan for recovering one or more information systems at an alternate facility.
What is the difference between RTO and RPO?
According to NIST SP 800-34, the RTO is the overall length of time a system's components can be in the recovery phase before negatively impacting the organization's mission, and the RPO is the point in time to which data must be recovered after an outage. The first measures a duration, the second a loss of data.
Does NIS2 require a BCP or a DRP?
Article 21 is addressed to Member States, which shall ensure that essential and important entities take risk-management measures. Paragraph 2, point (c), lists among them "business continuity, such as backup management and disaster recovery, and crisis management". The text names no specific plan and sets no recovery time; the detail is left to national transposition.
How do you test a disaster recovery plan?
With exercises chosen according to what you want to verify: a tabletop exercise for the alert procedure, a simulated exercise for the crisis team, and a real failover for the technical procedures. Compare the measured recovery time with the RTO and the age of the restored data with the RPO.
Why does a DRP need a dependency map?
Because the restart order comes from it. An application comes back only after the services it depends on: storage, database, directory, DNS, network flows. A dependency missing from the plan is discovered on failover day.