
A business continuity plan template: what needs to be on paper
A business continuity plan is not a technical document. The IT recovery plan says how systems come back; this one says how the organisation keeps working until they do, who makes that call and who gets told what. Below is a template you can fill in, and the one place it most often fails: it lives on the server that has just stopped.
The IT recovery plan says how systems come back. The business continuity plan says who decides, how work carries on in the meantime and who is told what.
The recovery order comes from the business: how long each process can stand still and how much work can be lost. That is a board decision, not a setting the administrator picks.
The most common failure is also the simplest: the plan sits on the file share that goes down. A printed copy in two places and a copy on the crisis team’s phones do work.
The business continuity plan template, section by section
A business continuity plan describes how the organisation keeps going when something essential fails. IT failure is one scenario in it, and in practice the most frequent one. The technical part, which server comes back from which copy, belongs in a separate IT disaster recovery plan. This document gives that plan its brief.
The template, in this order:
- When this plan applies, and who may invoke it.
- The crisis team: roles, names, deputies.
- The contact list, held outside your own systems.
- Processes in order of importance, each with how long it can stand still.
- The fallback procedure per process: how work continues without the systems.
- Communication to staff, customers, suppliers and regulators.
- How you return to normal, and who decides that.
- Owner, date of the current version and date of the last test.
That fits on a handful of pages. Anything longer goes in an appendix, because nobody reads twenty pages in the middle of an outage.
When the plan applies, and who decides
The first line of the plan matters most: when do you open it? Ten minutes of downtime is not a crisis, a day is, and somewhere in between someone has to say that this is it. Write down the trigger (for example: the main system has been unavailable longer than the process can cope with, or there is a suspected attack) and who is allowed to call it.
Then the crisis team, as roles rather than just names. Someone who decides, someone who coordinates the technical work, someone who communicates, and someone who logs what was decided and when. In a small organisation that is two or three people wearing several hats, which is fine as long as each role has a deputy. People are on holiday, usually in the week it goes wrong.
Also write down what someone may do without asking. May the IT coordinator disconnect the internet line? Call in a supplier outside the contract? Anyone who needs a call with the managing director first is waiting for a call that may not be answered.
A contact list that still works
Your address book lives in your email, and your email lives in the system that has just failed. That is why the contact list belongs in the plan, and the plan belongs outside the environment.
- Your own people: the crisis team and deputies, with a personal number, because the work phone may depend on the same environment.
- Suppliers: your IT partner, your internet line provider, your telephony provider and the main application vendors, with contract numbers and the number that answers out of hours.
- Beyond IT: your insurer if you hold a cyber policy, your landlord, and whoever gives you legal advice.
- Regulators: where you report a personal data breach and, if you fall under NIS rules, where you report an incident.
Put the date each entry was last checked next to it. A list with your previous supplier’s number on it is worse than an empty one, because it costs you time.
Which process first, and how long it can wait
The plan does not rank systems, it ranks processes: taking orders, sending invoices, running payroll, keeping customer service reachable. For each process you ask two questions.
- How long can it stand still before it really hurts? That is the recovery time you accept, the RTO (recovery time objective).
- How much work can be lost? A morning of orders keyed in again, or nothing at all? That is the data loss you accept, the RPO (recovery point objective).
The board answers those, not the administrator. It is a trade-off between what downtime costs and what it costs to make it shorter, and that is a decision about the business. So do not fill in a figure because it looks tidy; fill in what you can genuinely do without, and let IT work out what that requires.
The recovery order follows from that list. Write the reason next to it: the order will change, and with the reason you can derive it again rather than reinvent it. Watch for processes that lean on each other. Invoicing seems able to wait until you notice the accounts package also produces the payroll payment files.
Keeping work going while IT comes back
A recovery time of eight hours is only acceptable if there are eight hours you can bridge. So for every important process the plan needs a fallback procedure: how does the work carry on without the system?
- On paper. A printed order form, a list of yesterday’s open appointments, a copy of the price list. Old-fashioned, and it works.
- Somewhere else. Can people work from home if the office is unavailable, and on which devices? Is there a second site, and who has a key?
- On spare equipment. If laptops are infected or encrypted, where do the replacements come from, and how long before they are ready to use?
- Staying reachable. Can the main number be forwarded to a mobile? Who knows how, and is it written down?
Also write down how work from the fallback period gets back into the systems. Orders taken on paper still need entering, and without an agreement on that, a pile always goes missing.
What you say, to whom, and through which channel
During an outage everyone wants to know what is going on at once, and if nobody tells them, the story writes itself. The plan sets out who communicates and through which channel, and that channel is not your own email.
- Staff: a messaging group or text list that runs outside your own environment, with an agreement on who may post in it.
- Customers: a few prepared messages, short and without speculating about the cause. What is happening, what it means for them, when they will hear more.
- Suppliers and partners: especially where your systems connect to theirs, because they will want to know whether to disconnect.
- Regulators: in the UK a reportable personal data breach goes to the ICO within 72 hours of becoming aware of it. Write down who files the report, so recovery and reporting do not get in each other’s way.
Agree in advance what is not said while the cause is unknown. An early statement that “no data has been lost” is hard to take back. The NCSC publishes practical guidance on incident planning that is worth reading alongside your own plan.
Where it fails, and how the plan stays current
The plan sits on the server that fails. That is the classic, and it happens at organisations that have everything else in order. Keep a printed copy in two places, for instance at the office and at a crisis team member’s home, and a copy on the crisis team’s phones that opens without a network.
The plan is written for the person who wrote it. Internal abbreviations, steps that “go without saying”. Have it read by someone who did not help write it; what they do not understand, the deputy at three in the morning will not understand either.
The plan is never tried. A tabletop exercise, where the crisis team walks through a scenario with only the plan in hand, takes an afternoon and nearly always finds a gap: a number that is wrong, a decision nobody has the authority to make. Do it at least once a year.
The plan goes stale. Update it after every exercise, after every real outage, when a supplier changes and when the crisis team changes. One named owner and a date on the cover; a plan without an owner ages exactly as fast as the organisation changes.
Where our work stops
We write and maintain the IT part: the recovery plan, the backup and failover, the contact details of the suppliers we know, and we take part in the exercise. The business continuity plan itself stays yours. Which process may stand still for how long, who may decide and what customers are told are choices about the business, and the board makes them.
We are not a crisis communications agency, a lawyer or an insurer either. Whether an incident is covered by your policy or is reportable is for the party responsible for that to judge. And if you want the plan certified, against ISO 22301 for example, that is done by a certification body; we certify no one.
Questions we get about this
The ones that come up most when the plan is on the table.
What is the difference between a business continuity plan and a disaster recovery plan?
A business continuity plan describes how the organisation keeps working when something fails: who decides, how work carries on and who is told what. A disaster recovery plan is the technical part that says how IT systems come back and in what order. The continuity plan gives the recovery plan its brief, because it sets how long each process can stand still.
What should a business continuity plan template include?
When the plan applies and who invokes it, the crisis team with deputies, a contact list kept outside your own systems, processes ranked by importance with their accepted recovery time and data loss, a fallback procedure per process, communication to staff, customers and regulators, and an owner with the date of the last test.
Where should you keep a business continuity plan?
Outside the environment that can fail. A plan on a file share that is down does not exist at that moment. A printed copy in two places and a copy on the crisis team’s phones that opens without a network both work.
How often should a business continuity plan be tested?
At least once a year with a tabletop exercise, where the crisis team walks through a scenario with only the plan in hand. On top of that, update it after every real outage, when a supplier changes and whenever the crisis team changes.
Will you write our business continuity plan?
The IT part, yes: the recovery plan, the backup and failover, and we take part in the exercise. The continuity plan itself stays with the organisation, because how long a process may stand still and who may decide are choices about the business. Certification against a standard is done by a certification body, not by us.
Further reading
The technical recovery plan, and what happens when it goes wrong.
Shall we walk through your plan once?
Send us what you have, or tell us how things would go if the systems were down for a day. We will go through the IT part and tell you which questions the plan does not yet answer.
Practical IT knowledge in your inbox
New guides on management, security and the workplace, written by the people doing the work. No sales talk, and you can unsubscribe in one click.