Choose the work that cannot wait
A backup drill starts with the work that cannot wait for the primary assistant to return. Choose two or three tasks with a daily or weekly deadline, such as time-sensitive inbox replies, payment requests, or client status updates. Do not test the entire role at once. A narrow drill reveals more than a broad rehearsal.
Write the failure you are testing. The primary may be sick, on leave, or without a working connection. The backup may be a teammate, a supervisor, or a provider's coverage person. Name the person and the date before the drill begins. A drill without a named backup tests an idea rather than a plan.
Set the standard for success in observable terms. The backup should find the current instructions, open the required tools, complete the sample task, and record the result. A drill that only asks whether the backup knows the work is a conversation, not a test. Observable steps show where the plan breaks.
Prepare the access before the drill
Check whether the backup has a named account and the minimum permission needed for the task. A shared password is not a backup plan. If the backup needs temporary access, record how it is granted, how long it lasts, and who removes it. The drill should test the grant and the removal, not only the login.
Confirm that the backup can reach the instructions from the place they will actually look. A document buried in the primary assistant's personal folder is not available. Put the current runbook in the team's shared workspace and link it from the task. Test the link with the backup's account.
Check the recovery path for the tools that require a second factor or a client approval. A backup who can log in but cannot complete the approval is not covered. Record the client contact and the expected response time. The drill should reveal that dependency before a real absence.
Run the drill on real or realistic work
Use a real task if it is safe, or a realistic sample with invented records. The backup should complete the work without asking the primary for help. A short question is allowed only if it is the kind of question a real absence would permit. Record every question and every delay.
Time the drill from the start signal to the completed output. Record the steps that were unclear, the tools that were missing, and the decisions that needed someone else. Do not stop the drill at the first problem. The purpose is to find the whole path, not to prove that the backup is capable.
Ask the backup to write a short note describing what they did and what they could not do. The note is the evidence for the review. Compare it with the runbook and the primary's version of the task. Differences between the two are the gaps the drill was meant to find.
Review the gaps by cause
Group the gaps into access, instruction, tool, decision, and capacity. An access gap needs a permission or a recovery fix. An instruction gap needs a clearer runbook. A decision gap needs a named owner who can act during an absence. The cause decides the repair.
Fix the smallest gap first and repeat that part of the drill. A missing link or an expired permission can be repaired in a day. A decision that no one can make without the primary may need a policy change. Do not mix the two in one action list.
Record the time the backup needed and compare it with the service standard. A backup who can complete the task but takes three times as long may still miss the deadline. The comparison shows whether the coverage is real or only theoretical. Adjust the standard or the plan before the next absence.
Test the return and the handback
A backup plan also needs a handback. When the primary returns, the backup should transfer the open items, the decisions made, and the access used during the absence. Test that handback in the drill. A return that leaves work in two places creates a new risk.
Revoke any temporary access at the end of the drill and confirm that it is gone. Record the removal date and the person who verified it. A drill that leaves extra access behind has created the problem it was meant to prevent. The cleanup is part of the test.
Ask the primary to review the backup's output against the normal standard. The primary may find errors that the backup did not see. Record the errors and update the runbook or the sample. The drill should improve the instructions for the next person, not only measure the current one.
Set the cadence and the owner
Run the drill on a fixed cadence, such as once a quarter for critical roles and twice a year for others. Put it on the same calendar as the access review. A drill that is scheduled only after an absence is a reaction rather than a control. The cadence should match the risk of the work.
Name the person who owns the drill and the person who approves the result. The assistant can run the drill and prepare the evidence. The manager should decide whether the coverage is adequate. Do not let the person being tested approve their own result.
Review one part of the plan each cycle and change it if the work has changed. A new client, a new tool, or a new deadline can make an old backup plan obsolete. Record the change and the next drill date. The plan should be a living document, not a binder on a shelf.
Sources
Open the original publisher and confirm the current version, definition, and scope before using a source for a business decision.
Turn the reading into a role brief
Bring the recurring tasks, tools, schedule, access limits, and owner-only decisions. Start with one work sample and one clear review path.
Plan a Philippines-based role