Onboarding

Write a first-month check-in script for a virtual assistant

A repeatable check-in script that surfaces blockers, review load, and role drift without turning the meeting into a status recital.

By OnboardingEmployees editorial team | | 6 min read

Write a first-month check-in script for a virtual assistant
On this page
  1. Open with the work that did not go to plan
  2. Ask about the review queue, not only the output
  3. Check the tasks that have drifted from the card
  4. Review the escalation record together
  5. Ask about time, energy, and the shape of the week
  6. Close with one written action and one question
  7. Related reading

A first-month check-in script gives the manager and the virtual assistant a shared set of questions to review each week. The script keeps the conversation focused on blockers, review load, and role drift instead of a recital of completed tasks. It also creates a written trail of what changed from one week to the next. The same questions should be used each time so progress is comparable.

Open with the work that did not go to plan

Start the check-in with the tasks that took longer than expected or needed a second attempt. Ask what the assistant expected to happen and where the work diverged. This opening is more useful than a list of finished items because it exposes the friction the manager cannot see. The assistant should not be asked to justify every delay, only to describe the cause.

Record whether the cause was a missing instruction, a missing permission, an unclear source, or a change in the request. Those categories point to different repairs and keep the conversation from becoming personal. A delay caused by a tool problem is not a performance issue. The category should be written in the shared check-in note rather than left in memory.

Ask what the assistant did next after the problem appeared. The answer shows whether the escalation route works and whether the assistant felt able to use it. A good recovery is worth recording as an example for the next week. A weak recovery may mean the stop rule or the owner list needs revision.

Close this part by naming one change the manager will make before the next check-in. The change should be small enough to complete in a week, such as adding an example or naming a backup approver. An action owned only by the assistant turns the check-in into a performance review. Shared ownership keeps the script focused on the system.

Ask about the review queue, not only the output

Ask how many items waited for review and how long they waited. A role can look productive while work sits in a queue that nobody owns. The waiting time is often a better signal of friction than the number of completed tasks. Record the longest wait and the item that caused it.

Ask whether the assistant knew who would review each item before starting. An unclear reviewer causes rework because the assistant prepares for one standard and is judged by another. If the reviewer changes often, record the names and the reason. A stable reviewer for the first month makes the feedback easier to interpret.

Ask what the assistant would change about the review request itself. The assistant often knows which fields are missing or which attachments are routinely forgotten. That knowledge is useful when the intake form is revised. Write the suggestion down even when it cannot be adopted immediately.

Compare the review queue with the previous week and note whether it grew or shrank. A growing queue may mean the role is taking on work faster than the reviewer can absorb it. A shrinking queue may mean the work is getting easier or that items are being closed without review. The trend is more useful than a single count.

Check the tasks that have drifted from the card

Bring the role-scope card to each check-in and ask which task no longer matches it. Drift is normal in the first month because real requests reveal gaps the card did not anticipate. The question is whether the drift was approved or absorbed by habit. Only the approved version should become part of the role.

Name any task the assistant handled that is not on the card and record how it arrived. A one-off request from a client may be legitimate, while a recurring task from another team may be scope creep. The manager should decide which category it belongs to before the next week. An unrecorded decision becomes an expectation.

Ask whether any task on the card has stopped arriving. A task that no longer exists may be safe to remove, or it may have been quietly taken over by someone else. Removing it keeps the card honest and prevents the assistant from being measured against work that is no longer part of the role. Record the removal date and reason.

Review the escalation record together

Open the escalation log and read the questions in the order they were asked. Ask which answer was clear and which one needed a follow-up. The pattern often shows that a single owner is answering questions outside their authority. It may also show that the same question keeps returning.

Record the time between the escalation and the decision for each item. A slow decision may be a route problem rather than a difficult question. If the delay is routine, the script should include a step to name a faster route. The manager owns that change, not the assistant.

Ask whether the assistant ever continued without an answer because the wait was too long. The answer requires trust, so the manager should respond without blame. A yes is a signal that the stop rule or the backup owner is not working. Record the situation and the reason without writing a narrative about the person.

Ask about time, energy, and the shape of the week

Ask which part of the week felt rushed and which part felt quiet. A role that is overloaded on two days and idle on three may need a different task schedule. The answer also shows whether the assistant has enough time to check work before sending it. A constant rush is a quality risk, not only a comfort issue.

Ask whether the assistant has a reliable block of time for focused work. Interruptions from chat and email can make a simple task take much longer. If the role requires concentration, the manager should protect a window and record it in the agreement. The assistant should know which channel is appropriate for urgent requests.

Ask what the assistant would stop doing if one task had to be removed. The answer often identifies low-value work that exists because of an old habit. It may also reveal a task that belongs to another role. The manager can use the answer to test whether the current workload matches the original plan.

Close with one written action and one question

End the check-in with one action for the manager and one for the assistant. Each action should have a date and a visible place in the shared note. Two actions are easier to complete than a long list, and the next meeting can start by checking them. The assistant should be able to say what done looks like.

Leave time for one question the assistant has not asked yet. The question may be about the client, the tools, or the boundaries of the role. It may also reveal a misunderstanding that would otherwise appear as a mistake. The manager should answer it or record who will.

Write the check-in note in the same format every week so trends are visible. Include the blockers, the review queue, the drift, the escalations, and the experiment. A consistent format is faster to scan than a free-form summary. It also makes the first-month review easier to assemble.

Continue building the workflow

Related support: onboarding operations support, contact us. The U.S. Office of Personnel Management — Training and development guidance on onboarding, training, and employee development.