Behind every busy reception desk sits a set of administrative files: booking calendars, invoice records, supplier contacts, staff rotas and shared inboxes. This guide walks through eight steps for making sure those files can be recovered, in the order you would do them. It borrows the mindset of BackupOps, an informal name for backup and recovery work. It is operational information for office managers, not medical or legal advice, and it avoids anything clinical.
Step 1: Write down what you hold
Open a blank page and list every place where office information lives. Include the obvious ones, such as the booking system and billing spreadsheet, and the easily forgotten ones: the shared mailbox, the phone that holds the supplier chat group, and the website’s contact form entries. The Ops Atlas notes that an “unprotected” online service is one of the commonest finds when people do this exercise, because nobody remembered it needed protecting.
Step 2: Rank the list
Mark each item as top, middle or low priority. Ask: if this vanished today, how many days before the office could not function? Keep the top tier short, ideally under ten items. Everything that follows is applied to this tier first.
Step 3: Give each top item an owner
Write one name beside every top-tier item. Not “reception”, not “IT”. The Atlas is direct on this: when everyone owns the backups, in practice no one does. The owner does not have to be technical; they just have to ask the questions below and chase the answers.
Step 4: Keep a copy that your daily login cannot delete
This is the step that saves you from the worst days. Choose a separate storage location with a separate sign-in, held by a person who does not use it routinely. If an everyday account is misused or compromised, this copy is out of reach. The Atlas calls this an isolated copy, and it is the heart of what makes a backup trustworthy. If your storage offers locking, begin with the flexible mode so mistakes can be corrected, and consider stricter settings only after you understand them.
Step 5: Do the restore arithmetic
Next, work out how long recovery would take. The Atlas gives a larger-scale example: pulling four terabytes back through one stream at 250 megabytes per second works out to about 4.4 hours of waiting. Splitting it into four streams brings the copying down to roughly 67 minutes; adding about 20 minutes for replaying logs and 15 for validation gives about 102 minutes. A clinic’s files are far smaller, but the method carries over: data size, divided by real transfer speed, plus the checking time. If the total is longer than the office could tolerate, separate the essential files from bulky ones so the essentials return first.
Step 6: Switch on the right alert
Many tools alert on errors. Ask for an alert on the absence of success instead, such as “no good copy in two days”. Send it to a phone number or mailbox that does not depend on the main system. Test it once by pausing the job on purpose and seeing who is told.
Step 7: Practise a restore
This is the step that turns hope into evidence. Choose one top-tier item, recover it onto a different computer using only the backup and its sign-in, and open it. Note the time, what was missing, and what confused you. The Ops Atlas BackupOps entry calls an item that clears this kind of test “proven recoverable” and treats the share of items in that state as the number that matters, ahead of job success rates.
Step 8: Put it on the calendar
Book the next test before you finish this one. A sensible rhythm for a small office is a quick single-file restore every month and a fuller exercise each quarter. Add a line to your meeting notes asking for the date of the last successful restore. If that date is more than a quarter old, treat it as an open task.
A simple thirty-day schedule
- Days 1 to 7: steps 1 to 3, the list, ranking and owners.
- Days 8 to 15: step 4, the separate copy and its sign-in.
- Days 16 to 22: steps 5 and 6, timing and alerts.
- Days 23 to 30: steps 7 and 8, the first restore and the calendar entry.
The Atlas itself suggests a thirty, sixty and ninety day programme for larger organisations, starting with inventory, owners and missing-success alerts, then adding policy, isolation and key escrow. The eight steps above are the first month of that, trimmed for a small team.
Typical objections, answered
- “We pay for backup already.” Ask what it covers and when it was last restored.
- “Our data is too small to matter.” Small data is easy to restore; losing it still stops the office.
- “It will cost too much.” Costs are modest at office scale; the Atlas reminds readers to budget for several copies, not one.
- “Staff will forget.” Put the test in a shared calendar with a named person, and review it in a regular meeting.
Why the small stuff matters
It is tempting to think of backups as something for large companies with server rooms. In reality, the smaller the office, the fewer people there are to remember how things were set up, and the more a single failed laptop or a mistaken deletion costs in lost days. A tidy list and one practised restore give a three-person front desk the same basic assurance a large enterprise strives for, just at a human scale. Keep the paperwork simple enough that a new colleague could follow it on their first day.
Money matters nearby
Good records pair well with good budgeting. Our practical look at a budget-app plan for the festive spending rush and the simple piece on keeping a plain budget book for household money habits both show the value of writing things down. The same instinct applies to the files your office depends on.
Where you end up
After eight steps you will own a list, a set of names, one protected copy, a timing estimate, a useful alert and a restore you have seen with your own eyes. That is enough to turn a nervous guess into a calm answer next time somebody asks whether the office would cope.