Stockertim Dictionary · Artificial intelligence · Business
Manual fallback
Also called manual workaround, fallback procedure, manual fallback process
In short
A manual fallback is the documented way people carry on a process by hand when its automation is paused, broken or rolled back, so work continues without the system.
- Beginner
- Process
- Checked
In detail
Once a task is automated, the knowledge of how to do it by hand fades. Staff who keyed invoices move to other work, templates go out of date and new joiners never learn the old way. Then the automation stops, perhaps because a supplier changed a system, a cap was reached or a fault forced a pause, and nobody is sure how to keep the work moving. A manual fallback prevents that gap.
The fallback is a short written procedure: who does the work, with which screens, templates and checks, how the items affected are identified, and how the backlog is handled. It also covers the return to normal running, so that items done by hand are not processed again when the automation restarts and items caught mid-run are not lost.
Capacity is part of the plan. If the automation handles hundreds of items a day, a fallback may not keep up with all of them, so the procedure sets priorities, such as urgent or high-value items first, and says how long the organisation can run that way before other measures are needed.
Like any contingency arrangement, a fallback that has never been used is uncertain. Teams practise it, for example by running a day by hand each quarter, which also keeps the skills alive and tests the runbook's switch-over steps.
Worked example
Stonegate Veterinary Practice automates appointment reminders for about 120 clients a day.
- Fallback owner: the reception lead.
- If the automation stops, reception phones clients with next-day surgery appointments first, then texts the rest from the practice phone using a saved template.
- Clients reminded by hand are marked in the diary so that the restarted automation skips them.
- The fallback is run for one day each quarter as a rehearsal.
When the reminder service fails, surgeries still go ahead with the most important clients reminded, and no one receives two reminders.
In practice
Teams write the manual fallback before launch, name an owner, set priorities for when volume exceeds manual capacity, link it from the runbook, and rehearse it on a schedule so the skills and templates stay current.
Watch out
- Restarting the automation without marking hand-processed items can duplicate work.
- A fallback nobody has practised may fail when it is first needed.
Learn it properly
Official sources
- SP 800-34 Rev. 1, Contingency Planning Guide for Federal Information Systems · National Institute of Standards and Technology
- Incident management · National Cyber Security Centre
Facts last checked on 9 November 2026.
Education, not advice: the Stockertim Dictionary explains what terms mean. It never recommends buying or selling anything.