A datacenter fire, a ransomware attack, a simple human error… and the whole business stops. The question is no longer “if”, but “when”. A Disaster Recovery Plan (DRP) and a Business Continuity Plan (BCP) turn a potential catastrophe into a controlled interruption. Provided you build them on the right foundations — and test them.
Many organisations think they have one… until the day they need it. Here’s how to build a genuinely reliable plan, starting with the two numbers that structure everything: RTO and RPO.
1. DRP, BCP: what are we talking about?
The BCP (Business Continuity Plan) aims to keep the business running, even in degraded mode, during the incident — the business view. The DRP (Disaster Recovery Plan) aims to bring the IS back into service after a disaster — the technical view.
Key takeaway
The BCP answers “how do we keep working?“, the DRP “how do we get the IS back up?“. The two complement each other: one for the business, one for IT.
2. The 2 numbers that structure everything: RTO and RPO
Before any technical solution, two objectives to set per service: how long can you go without it, and how much data can you afford to lose.
RPO (Recovery Point Objective): the amount of data you accept to lose — driven by backup frequency. RTO (Recovery Time Objective): the maximum time to restore the service — driven by the recovery strategy.
3. Where to start: the 5-step method
1 · Business impact analysis (BIA)
Identify critical activities and their dependencies, then derive an RTO and RPO for each service.
2 · Define the recovery strategy
Backups, replication, recovery site: the right level depending on the target RTO/RPO and available budget.
3 · Document the plan
Step-by-step procedures, roles and contacts, application restart order and dependencies.
4 · Test regularly
An untested plan is fiction. Planned then unannounced drills, to reveal blind spots.
5 · Keep it up to date
The IS evolves, so does the plan: review after every major architecture or application change.
4. The strategy depends on the target RTO
The shorter the RTO, the more sophisticated — and costly — the solution. So size by criticality: vital services first.
5. Backups: the non-negotiable foundation
- 3-2-1 rule: 3 copies, 2 media, 1 offsite / offline.
- Immutability: copies that neither an attacker nor an error can alter.
- Isolation: backups not reachable with the current IS credentials.
- Restore tests regularly: a backup never restored doesn’t really exist.
6. The misconceptions
“We have backups, so we have a DRP”
A backup isn’t a plan. Without procedures, a target site or restore tests, recovery is improvised under pressure.
“The cloud makes us resilient”
The cloud offers building blocks, but resilience must be designed: redundancy, isolated backups, defined RTO/RPO.
“A plan written once is enough”
An untested, unmaintained plan fails on D-day. It lives with the IS and is rehearsed.
“It’s too expensive for us”
Size by criticality: not everything has the same RTO. Vital services first, the rest later.
7. Your DRP / BCP checklist
Frame
- Critical activities and dependencies identified (BIA)
- RTO and RPO defined per service
- Roles and crisis unit designated
Implement
- Immutable 3-2-1 backups tested
- Replication / recovery site sized by RTO
- Documented recovery procedures (restart order)
Verify
- Recovery drills planned (at least annual)
- Plan reviewed after every major change
- Emergency contacts and access up to date
8. Frequently asked questions
Difference between a backup and a DRP?
A backup protects data; the DRP organises the return to service — procedures, target site, restart order. One doesn’t replace the other.
How often should you test the plan?
At least once a year, ideally per critical service, and after every major change. A test always reveals surprises — better found cold than on the day.
Do you always need a second site?
It depends on the RTO. A short RTO requires replication and a recovery site; a wider RTO can rely on offsite, tested backups.
How do you protect the DRP from ransomware?
With immutable, isolated backups: if the IS is encrypted, you restore from a copy the attacker can’t reach — without paying.
Would your recovery plan hold on D-day?
ZENISEC designs, tests and industrialises your DRP/BCP — from impact analysis to the recovery site.
Build my recovery plan →
Discover our Infrastructure expertise →
Case study: Datacenter migration & IS rebuild for a mutual-insurance group →
Comments are closed