Key Takeaways
- O*NET lists 23 tasks for executive secretaries and executive administrative assistants, including scheduling, travel, correspondence, research, and office coordination.
- In O*NET work-context data, 99% report daily telephone conversations, 97% report daily email, and 80% report constant contact with others.
- BLS counted 489,300 executive secretaries and executive administrative assistants in 2025 and projects essentially unchanged employment through 2035.
- A Harvard Business School time-use study of 27 CEOs covered nearly 60,000 hours, showing why an allocation system needs to protect executive attention as well as move tasks.
- No U.S. government series separately measures virtual-assistant workload shares, so any universal EA-to-VA percentage split would be unsupported.
EA and VA workload distribution statistics: the evidence boundary
An executive assistant and a virtual assistant can both schedule, prepare a briefing, coordinate travel, and keep work moving. The useful allocation question is not which title is better. It is which work needs immediate executive context, sensitive-system access, in-person judgment, or a defined response window.
The evidence supports that distinction more clearly than it supports a universal percentage split. U.S. government data describes executive-assistant tasks and work context, but it does not publish a separate occupation or time-use series for virtual assistants. This page therefore separates observed statistics from a practical allocation model. The model is a decision aid, not a claim that every organization should divide work in the same proportions.
What official occupation data says about executive support work
The U.S. Department of Labor's O*NET profile for Executive Secretaries and Executive Administrative Assistants lists 23 tasks. The role covers high-level support, research, reports, information requests, correspondence, visitors, calls, and meeting scheduling. Its task list is broader than calendar administration alone.
| Observed measure | Result | What it means for allocation |
|---|---|---|
| Listed O*NET tasks | 23 | Support work spans scheduling, travel, documents, research, correspondence, records, and coordination. |
| Tasks labeled core in the O*NET profile | 15 | The occupation has a wide recurring workload, not one narrow specialty. |
| Daily telephone conversations | 99% | Live stakeholder coordination is a normal part of the role. |
| Daily email use | 97% | Inbox rules, escalation paths, and writing standards need ownership. |
| Constant contact with others | 80% | A handoff that ignores relationship context can create rework. |
| A lot of freedom to determine priorities | 73% | Assistants need decision rights, not only a task list. |
The high-contact figures do not prove that this work must remain on site or with one person. They do show why a task's communication and escalation needs should be assessed before it moves to another role or location.
Labor-market context for the EA role
The Bureau of Labor Statistics Occupational Outlook Handbook reported 489,300 executive secretaries and executive administrative assistants in 2025. It projects employment for that detailed occupation to be essentially unchanged from 2025 to 2035. The handbook reports a 2025 median annual wage of $76,590 for the occupation.
Those are U.S. occupation figures. They do not measure virtual-assistant contracts, assistant quality, tasks completed, or savings from a staffing change. Use them as labor-market context, not as a quote for a particular hire.
Why executive time still belongs in the allocation decision
In Harvard Business Review's report on the Harvard Business School CEO time study, researchers summarized nearly 60,000 hours logged by 27 CEOs. The study is about CEOs, not assistants or virtual assistants, and its sample is not a national average. It remains useful because it treats executive attention as a constrained resource rather than a free byproduct of a job title.
An allocation plan should therefore ask two questions at once:
- Can this task be completed accurately by the proposed owner?
- Does moving the task reduce executive interruption or simply create another review loop?
For broader executive-time context, see our CEO time management statistics.
A practical EA and VA allocation model
The table below is a planning model derived from the task and work-context evidence above. It is not a dataset and it does not prescribe staffing levels.
| Work characteristic | Typical first owner | Why |
|---|---|---|
| Requires executive judgment, confidential relationship knowledge, or authority to speak for the executive | EA or executive | The task needs trust, context, and an explicit decision right. |
| Repeats against a documented process with a clear quality check | VA or operations support | The work can be taught, sampled, and measured with a service-level expectation. |
| Needs same-day live coordination with senior stakeholders | EA, with VA preparation support | The relationship owner can handle exceptions while preparation is delegated. |
| Requires research, document formatting, list building, or data gathering | VA prepares, EA or executive approves | This preserves a review step where the decision carries risk. |
| Involves regulated, financial, legal, or personnel records | Named owner under least-privilege access | Sensitivity, not geography or title, should govern access. |
| Has no documented outcome, deadline, or escalation path | Keep with the requesting team until defined | Delegation cannot repair an undefined request. |
The most productive split is often layered rather than binary. A VA can prepare a meeting brief and maintain the source list, while an EA checks stakeholder context and an executive makes the decision. That is a workflow with visible checkpoints, not a transfer of accountability.
Allocation metrics worth tracking
Use measures that reveal whether the handoff is working. Do not rely only on hours moved off an executive calendar.
| Metric | Simple calculation | Reason to track it |
|---|---|---|
| First-pass completion rate | Tasks accepted without rework divided by tasks submitted | Shows whether instructions and quality checks are sufficient. |
| Executive review time | Minutes of executive review per completed task | Reveals whether delegation is actually reducing interruption. |
| Escalation rate | Tasks escalated divided by tasks assigned | Helps distinguish appropriate exceptions from unclear ownership. |
| On-time completion rate | Tasks finished by agreed deadline divided by tasks due | Tests the service window that the team actually needs. |
| Access exceptions | Access requests outside the approved role divided by all access requests | Identifies where the workflow or permissions need redesign. |
Set a baseline for two to four weeks before changing ownership, then compare the same task categories after the change. A calendar task and a client escalation should not be pooled into one average because the cost of delay is different.
Security and access caveat
Remote support can be appropriate for sensitive work only when access is designed for the task. The NIST SP 800-53 access-control family provides a useful control reference: assign access by role, limit it to what the task requires, and review it. This is not a substitute for legal, privacy, or industry-specific obligations.
No source on this page establishes that a VA is inherently more or less secure than an employee. Security depends on the systems, permissions, devices, contract terms, supervision, and incident process used in the specific engagement.
Source notes and caveats
- O*NET OnLine, Executive Secretaries and Executive Administrative Assistants supplies task and work-context data. The occupation profile is not a virtual-assistant survey.
- O*NET occupation update record shows that different data elements have different update dates. Treat task and context figures as occupation descriptors, not current market forecasts.
- BLS Occupational Outlook Handbook supplies U.S. employment, wage, and outlook context. It does not report VA utilization or workload allocation.
- Harvard Business Review's CEO time-study report summarizes a 27-CEO study. Its findings should not be generalized to every executive or company size.
- NIST SP 800-53 is a security-control reference, not evidence of staffing productivity or cost savings.
Frequently asked questions
What percentage of work should an EA delegate to a VA?
There is no authoritative universal percentage. Start with work that has a documented outcome, a repeatable process, an agreed response window, and permissions that can be limited. Keep executive judgment, sensitive relationship management, and exception ownership with the person who has the needed authority.
What tasks are best for a virtual assistant to prepare?
Research, meeting preparation, scheduling inputs, document formatting, follow-up tracking, and routine coordination are common preparation candidates when the process and quality check are defined. O*NET's task list supports that these activities are part of executive-support work, but the right split depends on context and risk.
Are executive assistants and virtual assistants the same occupation in BLS data?
No. BLS publishes data for executive secretaries and executive administrative assistants, among other occupations. It does not publish a separate national virtual-assistant occupation series, so comparisons must state that limitation.
Related Reading
Tags
Ready to put this into practice?
Book a free 15-min match call
Tell us what role you're filling. We'll match you with a pre-vetted virtual assistant - or tell you honestly if we're not the right fit.
Book a free call →