Virtual Assistant Vetting Process: How to Evaluate VAs Before You Hire

The gap between a VA who transforms your operations and one who creates more work than they save is almost entirely explained by the vetting process. Business owners who hire quickly - based on a resume and a 20-minute interview - get inconsistent results. Those who screen systematically get consistently better ones.

This guide documents a rigorous VA vetting framework, based on what we've learned from screening over 250,000 applicants and accepting fewer than 1 in 112. The principles apply whether you're hiring through a managed service, a freelance platform, or directly.

---

Why Most VA Hiring Fails

The interview problem. Most hiring decisions for VAs are made based on interviews. Interviews are good at revealing communication style, enthusiasm, and the ability to say the right things. They're poor at revealing actual work quality, reliability under pressure, or what happens when a task is ambiguous and there's no one to ask.

The self-representation problem. On freelance platforms, VAs describe their own skills. There's no independent verification. "Proficient in HubSpot" might mean "I have a HubSpot account I used once" or it might mean "I've managed 500 deals in HubSpot over three years." The description looks identical.

The failure cost problem. A bad VA hire costs more than the rate you paid them. It costs:

• The hours you spent screening and onboarding
• The weeks of substandard output before you recognized the problem
• The time to re-source, screen, and onboard again
• The opportunity cost of not having good support during the period

At Stealth Agents, we estimate the total cost of a failed VA hire that gets three months in before replacement is 8–14 weeks of equivalent executive time. At $100–$200/hr, that's $30,000–$55,000 in total impact for what looked like a $12/hr hire.

---

Stage 1: Define Before You Screen

Vetting doesn't start with evaluating candidates. It starts with knowing what you're vetting for.

Write the task list first. Every task you'll delegate, the estimated weekly hours, the tools involved, the output expected. This takes 30–60 minutes and prevents the common failure of hiring a generalist when you needed a specialist, or expecting phone support from someone you hired for writing tasks.

Define the must-haves from the nice-to-haves. For a customer service VA, strong written English is a must-have; experience with your specific help desk software is trainable. Know the difference before you screen - otherwise you'll screen out qualified candidates on the wrong criteria or ignore legitimate red flags.

Define the communication requirements. Same-day response? US business hours? Async with next-day turnaround? Your answers determine whether geography and time zone are must-have criteria.

---

Stage 2: Application Review

Before any interview or test, the application tells you a significant amount if you know what to look for.

What to evaluate:

Consistency of work history. Are the experiences described consistent with the role's requirements? A VA claiming customer service expertise but with a work history of only data entry roles has a credibility gap. Gaps in work history aren't automatic disqualifiers, but unexplained gaps warrant questions.

Specificity vs. generality. The best VA resumes and profiles describe specific things they did - "Managed 80+ customer support tickets daily in Zendesk for a SaaS company with 4.8/5 CSAT rating" - rather than generic descriptions like "responsible for customer service." Specificity signals genuine experience; generality signals surface familiarity.

Error rate in written materials. If the application itself contains grammatical errors, formatting inconsistencies, or unclear language, this is a signal - not a disqualifier for a single error, but a pattern of errors predicts future output quality.

Application effort. Did the applicant read the job description and respond to it specifically? Or did they send a template response that could apply to any VA role? Applicants who engage with the specific requirements of your role are more likely to engage with the specific requirements of your work.

---

Stage 3: Written Sample

Before any interview, request a written sample. This is the most information-dense screen you can run without scheduling time.

The prompt that works: "Please write 150–200 words describing what you would do in the first two weeks of starting as a VA for a new client. Be specific."

What you're evaluating:

• Sentence structure and grammar
• Clarity and logical organization
• Specificity vs. vague generality
• Evidence of genuine thought (not template responses)
• Appropriate professional tone

Red flags:

• Generic, templated language that could have been written by anyone ("I would work hard to meet all deadlines and exceed expectations")
• Poor sentence structure, excessive comma splices, or non-standard grammar
• Vague descriptions with no specific steps or concrete examples
• Responses that are notably short (suggests limited engagement) or excessively long and padded (suggests an attempt to obscure poor writing with volume)

Green flags:

• Specific, ordered steps that reflect genuine operational thinking
• Language that acknowledges the setup investment required ("In the first week, I'd focus on understanding your tools and workflows before taking on independent tasks")
• A tone that's professional without being stiff

At Stealth Agents, this stage eliminates approximately 40–50% of candidates who passed skills assessment. Written communication is the most reliable predictor of overall VA performance we've identified.

---

Stage 4: Skills Assessment

A skills assessment is a task-based test designed to verify the specific capabilities listed in the role requirements. The test should cover the actual tasks you'll assign - not abstract puzzles or generic aptitude tests.

Effective skills assessments are:

• Task-based. A customer service VA test includes sample tickets to resolve; a research VA test includes a research brief with deliverable; a bookkeeping VA test includes a sample transaction set to categorize.

• Time-bounded. Set a realistic time limit. A 2-hour assessment completed in 4 hours tells you something about time management.

• Evaluated against clear criteria. Before grading the output, define what an excellent response looks like. Grade against that standard, not against your gut.

• Specific to your tools, where possible. If the role requires HubSpot, include a HubSpot task in the assessment. If the role requires Canva, include a Canva deliverable.

Our approach: At Stealth Agents, skills assessments are custom-built by role category (admin, customer service, social media, bookkeeping, SDR). Each assessment tests both technical proficiency and judgment in ambiguous scenarios. The ambiguous scenarios are the most revealing - they show what the candidate does when the answer isn't obvious.

---

Stage 5: Structured Interview

After the written sample and skills assessment have screened the candidate pool, the interview validates what the earlier stages revealed and tests for judgment that's harder to assess on paper.

Avoid: Unstructured conversations that end up being likeability assessments. Likeable VAs who can't deliver independently are expensive.

Structure the interview around behavioral questions:

• "Tell me about a time you made a mistake in a task and how you handled it." (Evaluates self-awareness and integrity - do they own mistakes or deflect?)

• "Describe a situation where you received unclear instructions. What did you do?" (Evaluates communication - do they proceed and guess, or clarify before starting?)

• "What would you do if you completed a task and weren't sure the output was right, but your manager wasn't available to review it?" (Evaluates judgment under uncertainty)

• "Tell me about the most complex administrative task you've managed independently." (Evaluates actual experience vs. described experience)

Communication fluency assessment: For remote, async-heavy roles, assess how clearly the candidate communicates in real time. Not fluency - clarity. Can they explain their thinking? Can they ask good questions? Do they pause thoughtfully or fill silences with filler?

The question quality test: Pay attention to the questions they ask you. Candidates who ask no questions about the role have low curiosity or are trying to seem low-maintenance. Candidates who ask smart questions about your workflow, your expectations, and how you'll measure success are demonstrating the same intellectual engagement you'll want when they're working for you.

---

Stage 6: Reference Checks

References are underused in VA hiring. Most hiring managers call the reference, ask if the candidate "worked there" and whether they'd "rehire them," get affirmative answers to both, and call it done.

This produces almost no useful signal.

Reference checks that work:

• "What types of tasks did [candidate] handle independently, without supervision?"
• "Tell me about a specific situation where [candidate] demonstrated good judgment."
• "What is the one thing you wish [candidate] had been stronger at?"
• "On a scale of 1–10, how likely would you be to rehire [candidate], and what would need to be different for it to be a 10?"

The specificity of the answers is the signal. References who provide specific examples (including the fourth question's specific gap) are giving you real information. References who answer everything with "oh, they were wonderful, very hardworking, very professional" without specifics are either not recalling the candidate clearly or are providing a courtesy reference.

At Stealth Agents: We contact two professional references for every candidate who reaches Stage 4, using a standardized questionnaire with the questions above. References that score poorly on the "would rehire" scale or provide multiple hesitation signals result in candidate elimination regardless of performance in earlier stages.

---

Stage 7: Paid Task Test (Before Final Commitment)

After all previous stages have been passed, a 4–8 hour paid task test is the final verification. You pay the candidate at the rate you've discussed. They complete 4–8 hours of actual work. You evaluate the output.

What to evaluate:

• Brief adherence: Did they do what you asked, the way you described it?
• Output quality: Is the result usable without significant rework?
• Clarifying questions: Did they ask the right questions before starting (showing engagement with the brief), and did they ask before proceeding when something was unclear?
• Delivery: Did they deliver on time? Did they communicate proactively if anything came up?
• Self-review quality: Did they catch their own errors before submitting?

A task test that produces output you could use tomorrow is the strongest signal available. A task test that produces output requiring significant correction is a clear outcome signal - regardless of how the earlier stages went.

---

Red Flags That Should Stop an Engagement Early

Week 1 red flags:

• Silence on a missed task or deadline (no proactive communication)
• Output that didn't follow the brief despite clear instructions
• Inability to access or configure tools after setup
• Significant grammar or professional writing errors in client-facing communications

Week 2–3 red flags:

• Repeated escalation on questions that should have been covered in the brief
• Consistent need for instruction on tasks that should now be independent
• Pattern of partial completion without noting what was incomplete

Long-term disqualifiers:

• Dishonesty about task completion status
• Client data mishandled (shared, stored, or accessed inappropriately)
• Reliability pattern - consistent late delivery, repeated communication gaps

The most important principle: give specific, direct feedback the moment a concern appears. "I noticed this was delivered a day late without communication - can we talk about why and what the protocol should be?" Most concerns can be corrected with clear feedback in Week 1 or 2. Concerns that persist after clear feedback indicate a fundamental mismatch.

---

How Long Does a Rigorous Screen Take?

For most roles:

• Stage 1 (Definition): 1–2 hours (your time)
• Stage 2 (Application review): 20 minutes per candidate
• Stage 3 (Written sample): 10 minutes per candidate to review
• Stage 4 (Skills assessment): 1–2 hours to design, 10–15 minutes to review per candidate
• Stage 5 (Interview): 30–45 minutes per candidate
• Stage 6 (Reference checks): 15 minutes per reference (2 per candidate)
• Stage 7 (Task test): 30–60 minutes to review

Total investment for a well-screened hire: 6–10 hours across 2–3 weeks.

This compares favorably to the 30–50 hours typically consumed by a failed hire before replacement - plus the re-screening investment for the replacement.

The alternative to investing in screening is absorbing the cost of bad hires. At the rate most business owners absorb it, screening is significantly cheaper.

---

The Managed Service Alternative

Managed VA services run their own screening process on your behalf. You benefit from their screening investment without spending the time yourself.

At Stealth Agents:

• 250,000+ applicants screened since launch
• 0.89% acceptance rate - fewer than 1 in 112 applicants become active VAs
• 7-stage vetting process covering application review, skills assessment, writing sample, video interview, reference verification, background check, and performance monitoring during placement
• Replacement guarantee: If your matched VA isn't working out, we replace them without additional fee

The premium over a self-sourced hire covers the screening investment and the replacement protection. For business owners whose time is valuable, the economics typically favor managed services - the total cost of ownership, including screening time and failure rate, is lower even at a higher hourly rate.

---

The Bottom Line

VA vetting is not difficult. It's systematic. Most hiring failures are not the result of bad luck - they're the result of skipping stages, making hiring decisions based on interviews alone, or treating task tests as optional.

The business owners who get consistently good VA results do the same things every time: define the work before screening, test communication before skills, run a paid task test before committing, and give specific feedback the moment something is off.

The screening investment is the most cost-efficient thing you can do before an engagement. It's far cheaper than the alternative.