How to Vet a Virtual Assistant Company: The Questions Most Small Businesses Forget to Ask

A founder I worked with years ago, I’ll call her Renee since it’s not the kind of story she’d want attached to her real name, signed with a virtual assistant company after one sales call. The rep was sharp. The deck was clean.

They talked about “world-class talent” and “rigorous screening” and she nodded along, because who doesn’t want that. She signed the contract that afternoon.

Three weeks into onboarding, she asked her assigned VA a basic question about the CRM they’d told her the VA already knew. Silence.

Then a vague answer told her everything. Nobody had actually tested this person on the tool. The sales rep hadn’t lied, exactly. He’d just never been asked to prove it, so he never did.

Renee ended up doing the vetting herself, after the contract was signed, which is the most expensive time to find out a hire isn’t ready.

I think about Renee’s story every time I hear a founder describe how they chose a staffing partner.

Most people approach vetting a virtual assistant agency the way they’d pick a restaurant off a review site. They check the brand’s reputation.

They compare pricing. Maybe they read a testimonial or two.

None of that tells you anything about how the specific person who gets assigned to your business was actually screened. A five-star agency can still hand you a VA nobody properly tested, because the reputation belongs to the company, not to the individual sitting across from you every day.

This gap matters more than people think it does. CareerBuilder research found that a bad hire costs employers close to $17,000 on average, with roughly three in four employers admitting they’ve hired the wrong person for a role in the past year.

That number was built around traditional employees. When you’re hiring through a staffing company specifically to avoid that risk, and the company hasn’t actually done the screening it claims to have done, you haven’t reduced the risk. You’ve just added a middleman to it.

So, this is really the whole question of how to choose a reliable virtual assistant company: not what the brand says about itself, but what it can prove about the person it’s about to hand you.

Here’s what I’d ask if I were sitting across from a VA company deciding whether to sign up, the questions to ask a VA company before hiring that most sales calls never invite. The short version, before the detail:

Who Specifically Vets Candidates, and How

Not “we have a screening process.” Who does it, and what does it consist of. Is it a recruiter with a checklist, or someone who actually understands the skill they’re testing for?

A company that vets a bookkeeping VA the same generic way it vets a social media VA is telling you something important: the process is designed to move people through, not to catch mismatches. Ask for the specific steps.

If the answer is vague, that’s a virtual assistant company red flag, not a small thing to shrug off.

What Happens If the Assigned VA Isn’t a Fit

Every placement has some chance of not working out, and that’s fine; the question is what the company does about it.

Is there a replacement policy, and how fast does it kick in? Do you pay again to restart, or is a swap built into the agreement?

A company confident in its own screening will have a clear, fast answer here, because replacements should be rare enough that they’re not afraid to commit to one in writing.

How Communication and Reporting Actually Work, Day to Day

This is the part most sales calls gloss over. Who do you talk to when something goes wrong, the VA directly, or an account manager buffer? What does weekly reporting look like, and who’s checking that the work is actually getting done, not just that hours are logged?

I’ve seen founders go months assuming a task was handled because nobody flagged otherwise, only to find out later that “no news” meant “nobody was watching,” not “everything’s fine.”

What Tools and Systems the Company Already Tests For

This one is closer to table stakes than people treat it. If your business runs QuickBooks, Asana, and a specific CRM, ask whether the company tests those tools directly, or just asks candidates to self-report their comfort level.

Self-reported skill and demonstrated skill are two different things, and the gap between them is exactly where placements like Renee’s go wrong.

The One Question Almost Nobody Asks

Here’s the one that separates a company that’s kept up with how work actually happens now from one that’s still running a 2019 playbook: how do you confirm AI-tool fluency before placement, not just list it as a skill on a profile?

I wrote about this from the other direction in Your Next Hire Should Know AI Better Than You Do, the core argument there is that a resume line claiming AI fluency and tested AI fluency are two completely different things, and most staffing companies still don’t distinguish between them. That gap has only gotten more expensive to ignore.

It’s not hard to see why the claim spread so fast:

That first number comes from industry research, and the pattern holds wherever you look: candidates can claim the skill far more easily than they can demonstrate it. The claim has become cheap. The proof hasn’t.

The middle number comes from Monster’s AI Resume Trends Report, tracking how fast that claim spread in just two years. The wage number comes from PwC’s 2026 Global AI Jobs Barometer, which is exactly why the incentive to claim the skill, whether or not it’s true, keeps climbing.

When the reward for claiming a skill goes up, more people claim it, whether or not it’s true. A serious VA company should have an answer ready for exactly how they separate the two, a specific test, a specific scenario, a specific thing they ask a candidate to actually do with an AI tool in front of them.

If the answer is “we ask about it in the interview,” that’s not a test. That’s a conversation, and a company that can’t get more specific than that is showing you a virtual assistant company red flag hiding behind a buzzword.

Why This Isn’t Optional Anymore

This isn’t just a hiring nicety. A Time etc survey of entrepreneurs at growing businesses found they spend an average of 36% of a 45.5-hour work week on administrative tasks, the exact work you’re hiring a VA to take off your plate.

If the person taking on that work isn’t actually fluent in the AI tools that now handle a meaningful share of that admin load, you haven’t delegated the work. You’ve just relocated it to someone slower at it than the tools would allow.

And that’s before accounting for what it costs to redo the search: ZipRecruiter’s 2026 employer survey found AI fluency has become one of the fastest-rising expectations hiring managers screen for, which means a VA company still treating it as optional is already behind where the market has moved.

A company that can’t answer these four questions specifically, who vets, what happens if it doesn’t work out, how communication runs day to day, and how AI fluency actually gets tested, isn’t lying to you, most of the time. It’s guessing along with you, and hoping the guess holds. You’ll find out either way. The only choice you have is whether you find out before you sign or after, and after is always more expensive.

If you’re earlier in this process and haven’t decided whether a VA makes sense for your business yet, I’d start with What You Need to Know Before You Hire a Virtual Assistant. It’s the readiness conversation that naturally comes before this vetting one.

When you’re ready to see how we actually handle these four questions at Simpalm, our staffing process walks through it step by step. Or just book a consultation and ask me directly. I’d rather answer the hard question upfront than have you find the gap three weeks into onboarding.