On a normal Tuesday morning the invoicing system stops responding. Sales cannot issue quotes, finance cannot check payments and someone asks in the chat whether anybody knows who to call. Within an hour five people are trying five different fixes. The outage itself is rarely the biggest problem. The lack of a shared plan is.
A continuity plan does not have to be a thick document. For a company of ten to fifty people, one page is enough, as long as it is short enough that people actually read it and specific enough to use under stress.
Pick the top five systems
Start with a simple question: if this stopped for a whole day, what would we be unable to do? Go through the tools in use and choose the five that hurt most. For many small companies the list looks like this:
- Email and calendar
- The system that issues invoices and takes payments
- The CRM or order system
- The shared file storage
- The website or online shop
Your list may differ. A workshop might put the production schedule first, a service firm the ticketing tool. Do not try to cover twenty systems. Five well-prepared systems protect more than twenty described vaguely.
Write one workaround for each
For each system, write a few lines: what can we do by hand for a day or two? The point is not an elegant solution, only a way to keep the important work moving.
| System | Manual workaround | Who prepares it |
|---|---|---|
| Use phone and a messaging app; a printed list of key contacts | Office manager | |
| Invoicing | Issue invoices from a spreadsheet template, number them in a paper list | Finance person |
| CRM or orders | Take orders on a shared form, enter them later | Sales lead |
| File storage | Keep a local copy of current contracts and price lists | IT or the owner |
| Website | Show a short notice with a phone number and an email address | Marketing |
A workaround only helps if the material exists before the outage. A printed contact list, a spreadsheet template and a local copy of the price list take an hour to prepare and save a day when they are needed.
Decide who decides
During an outage, the worst delay is waiting for permission. Name one person who declares an incident and decides what happens next, and one deputy for when the first person is away. Write down in the plan:
- Who declares that a system is down and that the plan is in use.
- Who contacts the supplier or the internal IT person.
- Who informs staff and, when needed, customers.
- Who can approve a spending decision, for example paying for urgent support.
The decision maker does not need to be technical. They need to be available and trusted to choose. If an agreement with the supplier includes a support phone number or a response time, copy it onto the page, because nobody will search the contract in the middle of an outage.
Agree the restore order
When several things are broken at once, people tend to fix whatever they noticed first. Agree the order beforehand. A sensible rule is to restore first what stops money coming in or customers being served, then what stops staff from working, then everything else.
Write the order as a numbered list next to the five systems and add a short remark where one system depends on another. If the CRM needs the login service, the login service comes first. If you are unsure about dependencies, the article on reconciling customers between a CRM and invoicing shows how closely such systems are often linked.
Communicate in two directions
An outage is easier to bear when people know what is happening. Decide in advance:
- Inside the company. One channel for updates, for example a group chat that does not depend on the failed system, and a rule that the decision maker posts a short update at set intervals, even if the update is "no change".
- Towards customers. A short, honest message template: what is affected, what customers can do meanwhile, when you will update them again. Do not promise a restore time you do not control.
If an outage leads to complaints, record them in your complaint log so that the pattern is visible later.
Backups are part of the plan
A workaround covers a day or two. A lost database needs a backup. On the same page, note where backups of each of the five systems are kept, how old the latest one is and who has actually tried to restore from it. A backup that has never been restored is a hope, not a plan.
What the page looks like
Keep it to one sheet, printed and also available outside your main systems:
- The five systems with owner and supplier contact.
- The manual workaround for each.
- The decision maker and the deputy with phone numbers.
- The restore order.
- Where backups are and who can restore them.
- Message templates for staff and customers.
Store a printed copy in the office and a copy on a phone or in a service that does not depend on your own servers.
Run a short drill
Once or twice a year, spend thirty minutes on a drill. Pick one system, announce that it is down and let the team follow the page. Note what was unclear: a missing phone number, an outdated workaround, a template nobody could find. Fix these on the same day. Review the drill in your weekly operations meeting so the corrections do not get lost.
Common mistakes
- Writing too much. A long plan is not opened during an incident.
- Keeping the plan inside the failed system. If the file storage is down, the plan must not live only there.
- No named decision maker. Everyone waits for someone else.
- Workarounds that were never tried. A spreadsheet template that nobody has opened fails on the first day.
- Outdated contacts. People change roles and phone numbers; check the page after every staffing change.
How to start
- Choose the five systems this week.
- Write one workaround for each and prepare the material it needs.
- Name the decision maker and the deputy.
- Agree the restore order and write it down.
- Print the page, share it and book a thirty-minute drill within three months.
Summary
A system outage is a matter of when, not if. One page with five systems, a manual workaround for each, a named decision maker and an agreed restore order turns an hour of confusion into a calm, predictable routine, and the cost is a single afternoon of preparation.
Sources
- Our related articles linked above.
- The approach described here is common practice for small companies, used as a practical guide. It is not legal or insurance advice. No client results or market figures are claimed in this article.
