Virtual Assistant Onboarding Checklist: Pre-Start to Day 30

Pre-start checklist: make the role ready before the person arrives

Onboarding starts before the first call. A new Virtual Assistant cannot become productive when the role is still a collection of ideas in the manager's head. Write down the first-month outcome, recurring work, schedule, reporting line, and decisions that stay with the client. Choose two or three workflows for the first week rather than transferring every possible task at once.

Prepare one finished example for each starting workflow. Show the input, completed output, source of truth, quality check, and next handoff. Remove old instructions and conflicting versions before the new hire sees them.

• Confirm the agreement, start date, hours, timezone, compensation, and payment process.
• Name the manager, backup contact, access approver, and workflow owner.
• Define what complete means for each starting responsibility.
• Schedule the welcome call, first review, end-of-week check, and 30-day review.

Use staged access and least privilege

Give the Virtual Assistant the smallest access needed for the first approved workflows. Use client-controlled accounts, role-based permissions, a password manager, and multi-factor authentication where supported. Do not send passwords through ordinary chat or grant administrator access because it is faster than setting up the right role.

Keep an access register with the system, account owner, permission level, approval date, and offboarding action. Add permissions only when the scope expands. This makes troubleshooting clearer and access removal safer when responsibilities change.

• Create named accounts instead of shared personal logins where possible.
• Separate view, edit, publish, refund, payment, and administrator permissions.
• Use dummy or low-risk records for practice when a workflow handles sensitive data.
• Explain prohibited actions and the process for reporting a security concern.

First-day checklist: explain context and boundaries

The first day should explain why the work exists, who depends on it, and what can go wrong. A tool tour teaches buttons but not judgment. Show where new work appears, which record is authoritative, how priorities are set, and how the manager can see that an item is complete.

Define communication precisely. State which channel is used for urgent issues, normal questions, daily updates, and approvals. Agree on response windows and live overlap. Being proactive does not mean making decisions outside the role.

1. Explain the customer, workflow outcome, and the assistant's place in the handoff.
2. Review hours, meetings, holidays, communication channels, and response expectations.
3. Walk through the first workflow from trigger to completed record.
4. Show decision boundaries and situations that require immediate escalation.

First-week checklist: teach through real repetitions

Use a show, do together, observe, and release sequence. Demonstrate one real example, complete the next one together, then watch the Virtual Assistant complete a safe repetition. Review the actual output immediately. This reveals unclear instructions faster than a long presentation or a week of passive videos.

Keep the first-week scope narrow enough to create repetition. Touching ten unrelated systems once does not show whether a workflow is becoming reliable. Several repetitions of the same queue create evidence about accuracy, questions, speed, and exception handling.

• Day 1: context, access, communication, and one demonstrated workflow.
• Day 2: complete the workflow together and record missing instructions.
• Days 3 to 4: supervised repetitions with fast, specific feedback.
• Day 5: review accuracy, open questions, access gaps, and the next responsibility.

Set 30-day milestones around competence

A useful 30-day plan describes what the Virtual Assistant should run, how accurately, and with what level of supervision. Avoid goals such as learn the business or become proactive. Those phrases do not tell either side what evidence counts.

Expand scope only after the previous responsibility is stable. Specialist or customer-facing work may take longer than 30 days. The plan should show progress and expose support needs, not pretend every role reaches independence on the same date.

• Week 1: complete the starting workflow with supervision and escalate known exceptions.
• Week 2: repeat normal cases accurately and maintain records without reminders.
• Week 3: manage the queue rhythm and communicate blockers early.
• Day 30: own normal work and bring exceptions with enough context for a quick decision.

Give feedback that changes the next output

Feedback should name the standard, show the difference in the completed work, and state the next action. Replace be more careful with the record was saved without the customer reference; check the required fields before moving an item to complete. Specific feedback can be applied on the next repetition.

Use a short daily check during the first few days, then reduce frequency as work stabilizes. Keep approvals and corrections in the agreed system so the assistant does not reconstruct standards from scattered messages. Ask what was unclear or unnecessarily slow because onboarding improves in both directions.

Use a simple onboarding performance scorecard

Measure the workflow with signals that match the role. Useful performance checks include accuracy, on-time completion, response within the agreed window, backlog age, appropriate escalation, documentation quality, and the number of manager interventions needed for normal cases.

Do not reward speed when it creates rework. Review samples of completed work and exceptions. A healthy trend is stable quality, fewer repeated corrections, earlier escalation, and less manager effort for the agreed scope.

• Quality: what percentage of sampled work meets the definition of done?
• Reliability: are normal tasks completed within the agreed service window?
• Judgment: are exceptions escalated at the right time with useful context?
• Independence: is manager intervention decreasing for normal cases?

Day-30 review and next-step checklist

Use the day-30 review to decide what stays the same, what needs support, and what can expand. Compare the original outcome with actual completed work. Review schedule fit, access, communication, quality, workload, and whether the hours match the real queue.

Confirm the workflows now owned, standards that still need work, the next responsibility, access changes, and the next review date. If the role is not working, identify whether the cause is fit, training, unclear scope, missing authority, or unrealistic workload before choosing the response.