Virtual assistant performance metrics: a simple scorecard for founders
A practical VA performance scorecard with weekly metrics, review rules, source-backed guardrails, and update templates.
Operator sections with practical steps.
Decision points before handoff.
Visible citations and source notes.
Key takeaways
- Use 5 to 7 metrics, not 20. A small scorecard is easier to fill in every Friday and easier to coach from.
- Separate volume, speed, quality, judgment, and access safety so one good number does not hide a risky habit.
- Review 100% of high-risk work at first, then sample 20% to 30% of low-risk work once the assistant is steady.
- Ask for a weekly note with completed work, errors found, stuck items, SOP fixes, and owner decisions needed.
- The scorecard should decide one next step each week: keep the lane, tighten the rule, or add the next safe task.
A virtual assistant can look busy while the founder still feels stuck. Messages move, calendars change, and tasks close, but nobody knows if the work is getting safer or easier to manage.
A simple performance scorecard fixes that. It turns the first 30 days into a small set of numbers, notes, and coaching decisions instead of a vague feeling that the VA is doing well or falling behind.
Weekly VA scorecard
Start with these rows for the first month. Keep the scorecard short enough to finish in 10 minutes.
Two review notes to copy
- Weekly update: "Completed: [numbers]. Missed or delayed: [items + reason]. Errors found: [count + pattern]. SOP updates needed: [changes]. Owner decisions: [questions]."
- Coaching note: "Keep doing [specific habit]. Tighten [one rule]. Next week, do not expand the role until [metric or example] is steady."
Keep, tighten, or expand chart
Measure the role, not the person in general
A VA scorecard should match the work lane. An inbox assistant needs response drafts, labels, escalations, and missed follow-ups. A CRM assistant needs records updated, duplicate fixes, missing fields, and source links. A reporting assistant needs on-time reports, source checks, and owner questions flagged before the deadline.
The U.S. Bureau of Labor Statistics describes administrative assistant work around scheduling, records, correspondence, and office support. That is useful for a founder because most early VA work is visible support work. You can count it, sample it, and improve it without pretending the assistant owns the whole business process.
Start with 5 to 7 metrics. More than that turns the scorecard into another job. The weekly goal is not a perfect dashboard. It is a clear answer to one question: is this work ready for more trust, or does it need tighter rules?
Use five metric buckets in the first 30 days
The first bucket is volume: tasks finished by lane. Count items like 35 emails labeled, 12 follow-ups drafted, 8 calendar conflicts solved, or 60 CRM records cleaned. Volume matters because it shows whether the role is taking real work off the founder's desk.
The second bucket is speed: due dates hit, turnaround time, and aging tasks. The third is quality: error patterns found in review. The fourth is judgment: how often the assistant paused, escalated, or asked before a risky decision. The fifth is access safety: whether they used approved accounts, source links, and red-line rules.
Do not reward speed by itself. A VA who clears 100 records with 18 missing source links creates cleanup work. A better Friday note says: "60 records updated, 5 missing-source errors found in a 20-record sample, SOP step added for source links, and 4 records paused for owner review."
Set review rules before the work expands
For the first 2 weeks, review high-risk work before it leaves the business. That includes refunds, payments, contract wording, sensitive customer replies, payroll changes, legal wording, and anything involving private data. Low-risk work can move faster, but it still needs sampling.
NIST's Digital Identity Guidelines treat authentication and account recovery as security controls. In plain terms, the VA scorecard should include access habits: named accounts, MFA, approved folders, no shared passwords in chat, and no owner-only decisions made alone.
A simple rule works: review 100% of risky drafts, sample 20% to 30% of low-risk finished work, and log the pattern. If 3 of 10 sampled CRM updates miss source links, fix the SOP before adding another task lane.
Use a mini-scenario before you add more tasks
Picture a founder with a VA for 12 hours a week. Week 1 covers inbox labels and calendar cleanup. By Friday, the VA labeled 80 emails, drafted 10 follow-ups, scheduled 7 meetings, and flagged 3 customer questions. The founder samples 25 emails and finds 2 wrong labels and 1 follow-up that needs a softer tone.
Week 2 does not add bookkeeping, research, and support tickets. It tightens the current lane. The SOP gets 6 more examples. The assistant uses this note when stuck: "I paused this reply because the customer is asking for a refund. I drafted two options, but I need approval before sending." By week 4, simple scheduling and routine follow-ups can move to spot checks.
That is the value of metrics. They stop the founder from expanding because the VA is nice, fast, or available. The role expands because the current lane is accurate, documented, and safe.
Connect metrics to a keep, tighten, or expand decision
End each weekly review with one decision. Keep means the same work lane continues for another week. Tighten means the SOP, access, examples, or review rule changes. Expand means one adjacent task is added after the current lane is steady.
Atlassian's roles and responsibilities play is built around clarity on who owns what and how decisions get made. The VA version is simple: write what the assistant owns, what they draft only, what they may never approve, and who answers exceptions.
If the scorecard shows 90%+ on-time work, fewer repeated errors, no silent blockers, and clean access habits for 2 straight weeks, add the next safe task. If not, keep the lane small and improve the system first.
How to use this playbook
Read this article like a work guide. Pick one task, one role, or one quality rule and turn it into a short checklist before you give it to an assistant. Small changes are easier to train and easier to check.
The next step is simple: connect the idea to a role scorecard, SOP, onboarding plan, or weekly scorecard. That turns the article into a tool you can use.
Where this fits
Virtual assistant performance metrics: a simple scorecard for founders fits the same basic path used across OutsourcedU: choose the work, write the rule, train the person, check the result, and add more only when the work is steady. Skipping steps usually creates cleanup work.
If this topic is active in your business, the next supporting page is Remote team management systems. Use that page to connect the article to a broader setup plan, including outcomes, rhythm, quality checks, and management expectations.
FAQ
What metrics should I use for a virtual assistant?
Start with work completed by lane, on-time rate, error patterns, stuck items, SOP updates, escalation quality, and access-safety habits. Keep it to 5 to 7 metrics.
How often should I review VA performance?
Use a short weekly review during the first 30 days. Review high-risk work before it goes out and sample 20% to 30% of low-risk work until the assistant is steady.
What is a good VA scorecard for a founder?
A good founder scorecard is simple: tasks finished, deadlines hit, errors found, owner decisions needed, SOP fixes, and one next step: keep, tighten, or expand.
When can I give a VA more responsibility?
Add one adjacent task after the current lane is accurate, documented, and safe for 2 straight weeks. Do not expand just because the assistant has spare hours.
Sources
- U.S. Bureau of Labor Statistics, Secretaries and Administrative Assistants — Administrative task scope.
- NIST SP 800-63B Digital Identity Guidelines — Access controls.
- Gallup, State of the Global Workplace 2026 — Manager clarity.
- Atlassian Team Playbook, Roles and Responsibilities — Ownership rules.