---
title: "Backups You Have Actually Tested"
description: "What to back up, where the copies should live, and how to prove a restore works before the day you need one, in plain English for a small business."
date: 2026-11-15
updated: 2026-11-15
topic: "Insights"
author: Wave Agency
canonical: https://waveagency.com/blog/tested-backups/
---

# Backups You Have Actually Tested

Ask most owners whether their business is backed up and the answer is yes. Ask when they last restored something from that backup and the room goes quiet. Somebody set it up years ago. There is a little green check in a settings screen. Nobody has ever pulled a file back out of it.

That is not a backup plan. It is a belief. The only backup you can count on is one you have restored from recently, on purpose, on a quiet day. Here is how to get there without buying anything new this week.

## Start with what you cannot run without

Before you think about tools, make a short list of the things your business would be stuck without by lunchtime. For most local businesses it looks something like this:

- **Customer records**: the contact list, job history, patient or client files.
- **Money**: the accounting file, invoices, payroll records.
- **Paper you signed**: contracts, leases, permits, insurance policies.
- **The shared drive**: price sheets, templates, the forms your team fills in every day.
- **Email**, including the folders where someone files important threads.
- **The website**, its content and its settings, not just the domain name.
- **Phones and laptops** that hold work nobody else has a copy of.

Write down where each one lives today. You may find, as many owners do, that some of it lives in places nobody planned: a desktop in the back office, one employee's personal cloud account, a laptop that goes home every night. Our guide to [scattered business data](/blog/scattered-business-data-where-to-start/) walks through that inventory in more detail.

## Cloud apps are not automatically backed up

This is the most common surprise. If your email, files or accounting live in a cloud service, it is easy to assume the provider keeps a copy of everything forever. Usually they keep the service running. They do not promise to bring back the folder an employee deleted six weeks ago, or the records a bad sync overwrote.

Read the retention settings for each service on your list. Find out how long deleted items stay recoverable, who on your team can restore them, and whether the service offers a separate backup option. If a deleted file is gone for good after a few weeks, decide whether that is good enough for that kind of file.

## Keep copies in more than one place

A long-standing rule of thumb in IT is three copies of anything important, on two different kinds of storage, with one kept somewhere else. You do not need to follow it to the letter. You do need the idea behind it: a single problem should never be able to take out every copy at once.

That means:

- **One copy separate from your everyday systems.** If ransomware locks your shared drive, a backup that sits on the same network, signed in with the same account, can be locked with it.
- **One copy offsite.** A fire, flood or break-in at the office should not reach it.
- **Backups that run on their own.** A backup that depends on someone remembering to plug in a drive on Fridays will stop the first busy Friday.

Our [cloud infrastructure basics](/blog/cloud-infrastructure-for-growing-businesses/) covers how the cloud side of this fits together.

## Prove the restore works

This is the step almost everyone skips, and it is the only one that tells you anything. A green check means the backup job ran. It does not mean the files inside it are complete, readable or recent.

Put a restore test on the calendar a few times a year, and do it like a fire drill:

1. **Pick something real.** A folder from the shared drive, last month's accounting file, one employee's mailbox.
2. **Restore it somewhere safe**, to a new folder or a spare computer, never over the live copy.
3. **Open it.** Check that the files open, that they are the version you expected and that nothing is missing.
4. **Time it.** Note how long the restore took from the moment you started looking. That number is roughly how long you would be down on a bad day.
5. **Have someone else do it next time.** If only one person knows how to restore, your backup depends on that person being reachable.
6. **Write down what went wrong.** A missing password, a folder that was never included, a step that was not obvious.

The first test usually turns something up. That is the point. Better to learn the accounting file was never in the backup on a slow Tuesday than in the middle of a crisis.

## Know how long you can be down

Two questions make every backup decision easier, and only you can answer them:

- **How much work can we afford to lose?** If the answer for invoices is "none from today," a nightly backup is not enough for that file.
- **How long can we be without it?** A dental office without its schedule is stuck within the hour. An archive of old project files can wait a week.

Different data gets different answers, and that is fine. Write them next to each item on your list, and check that what your backups actually do matches.

## Questions to answer this week

- What is on our list of things we cannot run without, and where does each one live?
- Which of those are backed up automatically, and which depend on someone remembering?
- Is at least one copy kept separate from our network and our everyday logins?
- When did we last restore a file, and who did it?
- How long did it take?
- Is the restore process written down where someone other than its owner can find it?

Backups are one item on a bigger list. Our [plain-English cybersecurity checklist](/blog/small-business-cybersecurity-checklist/) covers the rest, from multi-factor login to knowing who to call first.

## How we help

Our [cybersecurity and data protection](/technology/cybersecurity-and-data-protection) work runs in three phases: assess, protect and monitor. We find your exposure, put controls in place, and watch over time, with backup and recovery part of the monitoring phase alongside threat monitoring and incident response. If something has already gone wrong, we help you respond, recover and harden against the next one.

Not sure your backups would hold up on a bad day? [Book a call](/book) and we will look at it with you.
