A Data Recovery Plan for a Small Business: What to Do If the Server Burns Out Tomorrow, Computers Get Encrypted or the Internet Goes Down — How to Decide in Advance What Matters, How Much Data You Can Afford to Lose and How Many Hours You Have to Get Back to Work

Small business owners usually think about backups when it's already too late. The server fails on Monday morning, ransomware encrypts the accounts the day before a filing deadline, a pipe bursts in the office, the director's laptop is stolen, the provider cuts the internet for three days to repair the line. At those moments it turns out there was a backup, but it was out of date, or nobody knows where it is, or only the administrator who left knew the cloud password, or restoring will take a week and clients won't wait. A recovery plan isn't a thick document for auditors — it's a few pages of answers to simple questions, prepared in advance. Let's look at how to put one together.

Step 1. What's critical for you

List the systems and data the company can't work without: the accounting and inventory database, the till and payment terminals, email, the client base or CRM, the website and online orders, shared documents, the client booking schedule, medical records, project files. For each, answer: where is it stored, who's responsible for it, and what happens if it's unavailable for an hour, a day, a week.

Step 2. How much data you can lose and how long you can wait

For each system, decide on two simple measures. The first is how much recent data you can afford to lose: a day's work, an hour's, or not a minute. That determines how often backups must be made. The second is how quickly work must be restored: within an hour, a day, a week. That determines where backups are kept and whether you need spare equipment. For a shop, the till is critical within an hour; for a project archive, a day is acceptable.

Step 3. Backups to match those requirements

• Backup frequency matches the acceptable data loss.
• Backups are kept in at least two places, one of them off-site.
• At least one backup is protected from ransomware — a removable medium or storage with immutable versions.
• A responsible person receives notifications of successful and failed backups.

Step 4. Infrastructure that won't let you down

Businesses stop not only because of lost data. A backup internet line through a 4G/5G router with automatic failover keeps the till and online orders running when the provider fails. UPS units for the server, router and till protect against power cuts and surges. For critical computers, a spare laptop with the software installed lets you move an employee onto it quickly.

Step 5. People, contacts and passwords

The plan must say who makes decisions in an emergency, who technically restores the systems, and how to reach them at weekends and at night. It must contain contacts for the provider, software suppliers, the bank and the IT specialist. Admin passwords for the server, NAS, router, cloud services and domain name must be kept somewhere safe that the manager can access — not just in one employee's head.

Step 6. Simple instructions

For typical situations it's useful to have short one-page instructions: "the internet's down", "the database won't open", "suspected ransomware", "the accountant's computer has broken". They cover the first actions, who to call and where the backups are. In a stressful situation, these sheets save hours.

Step 7. Drills

An untested plan is an assumption. Every six months or year, run a practice restore: bring a database backup up on a spare computer, test the switch to backup internet, and check that contacts are current and passwords work. Each such drill usually uncovers one or two problems that are better found in calm conditions.

What our technician does

1. Works with you to list critical systems and their requirements.
2. Sets up backups, off-site storage and ransomware protection.
3. Installs backup internet, UPS units and, if needed, spare equipment.
4. Prepares a document with contacts, passwords, a network diagram and short instructions.
5. Runs a practice restore and regularly checks that everything works.

A real example

A small chain of three pharmacies in Gldani and Mukhiani kept its records on a single server in the main pharmacy, with no copy outside the building. We drew up a plan with the owner: we decided that the till and records had to be restored within two hours, with no more than an hour's data loss acceptable. We set up hourly backups to a NAS and nightly backups to the cloud, backup 4G internet at every location, UPS units, a spare computer with the inventory software installed and instructions for the pharmacists. Six months later the server's power supply burned out in the main pharmacy, and work was restored on the spare computer within forty minutes without losing any sales.

Don't forget the paperwork either. The provider contract, software licences, warranty documents for the server and NAS, and the domain name and hosting registration details are all needed in an emergency, and hunting for them at the last minute is hard. Keep copies of these documents with the plan. Separately, check whose name the website and email domain is registered in: it's often a former employee or contractor, leaving the company unable to manage it when something goes wrong.

Call a technician

If you want to prepare your business for failures and emergencies, we'll help draw up a plan and set up the systems in any district of Tbilisi. Learn more on our local network setup, data recovery and local network installation pages.