A fallback process is easy to describe and surprisingly difficult to use during an outage. Staff may not have the latest contact information, may disagree about who records requests, or may create a second source of truth that survives long after the tool returns.
Choose the work that must continue
Identify essential tasks and the tasks that can safely wait. A fallback does not need to reproduce every feature of the normal system. It needs to preserve the customer commitment and the information required to reconcile the work later.
Use an approved location for temporary records, with appropriate access and retention. Do not move sensitive information into personal messaging accounts or an unrestricted spreadsheet simply because the usual tool is unavailable.
Run a small exercise
For a hypothetical booking team, simulate an unavailable scheduling tool using safe data. Ask the operator to receive a request, avoid promising an unverified slot, record the next action, and hand the item to a colleague. Note which information they cannot access and which decisions remain unclear.
Explain to participants that this is an exercise so no real customers receive misleading outage messages. Keep test records identifiable and remove or archive them through the approved process afterward.
Plan the return to normal
Decide how temporary records will be entered into the main system and how duplicates will be recognized. Assign someone to confirm that each action is represented once. A restored login does not mean the backlog has been reconciled.
Review the exercise with the people who performed it. Shorten instructions that were difficult to use and correct missing permissions or references. Repeat the drill when the workflow changes materially.
A useful fallback is a tested operating method, not a document stored somewhere the team cannot reach when the system is down.

