How to find a virtual assistant that actually fits your business

Why a virtual assistant who looked right on paper can still be wrong for your business

If your first virtual assistant arrived with a strong CV, interviewed well and still had you redoing their work within a month, you are not alone, and you are probably not a poor judge of people. The second attempt is the hard one. You already know what a bad match costs: evenings spent correcting invoices, the same instructions explained twice, and a quiet suspicion that delegation simply does not work for a business like yours.

It usually does work. What tends to fail is the order of events. Most owners search for a person first, then try to work out what that person should do once they arrive. Time pressure explains why. The 2026 SME Business Barometer from American Express and Small Business Saturday UK, based on 1,000 owners, found that admin and finance tasks take around 11 hours a week, roughly six working days a month, against just 3.6 days spent on sales and growth. With a backlog that heavy, hiring fast feels like the fix. Defining the job first is the fix that holds.

Fit starts with a definition, not a search

Fit is the word everyone uses and almost nobody defines. In hiring conversations it tends to mean a vague sense of chemistry: the candidate seemed friendly, replied quickly and had once worked for a business that sounded similar. None of that tells you whether they can process your lettings paperwork accurately at 8am on a Monday.

A more useful definition is narrower. An assistant fits your business when they can do your actual tasks, at the hours you need them done, inside the tools you already use, in a communication style you can work with, and with someone behind them when life gets in the way. Every one of those can be written down before you meet a single candidate. Most of them can be tested within a fortnight.

Write it first. Search second.

The five-part fit test that decides whether an assistant will last

Each part of the test answers a question that owners usually discover the hard way, several weeks too late. Taken together, they replace the interview hunch with something much closer to evidence.

Notice what is missing from the list. CV length, a polished interview, a stack of certificates and the lowest hourly rate all feel reassuring, yet none of them predicts whether the arrangement still works in month three. Charm in an interview is a skill. It is not the job.

How to write a fit brief before you look at a single profile

A fit brief is a single page. It takes about an hour to write and saves weeks of guessing on both sides, because the assistant knows what success looks like and you know exactly what to test. This is not a formal job description. It is an honest picture of the work as it actually arrives on your desk.

Start with last month, not the ideal month. Open your inbox, calendar and task list, then note every recurring job you did that someone else could have done. That list, sorted by how often each job comes round, is the first draft of the role.

Testing fit on real work instead of in an interview

Interviews measure how well someone talks about work. A trial measures the work itself. For an owner who has been burned once, that difference matters more than anything else in the process, because the previous hire almost certainly passed the interview.

A useful trial draws on live tasks from the brief, runs long enough to include a busy day and a quiet one, and ends with a short review against the 30-day signs you wrote down. Two weeks is usually enough to see the early signs. Watch the questions the assistant asks in the first three days. Good questions are the earliest reliable sign of fit.

Whichever route you choose, insist on the same sequence: brief, match, trial, review.

What a good fit looks like in four different businesses

The five-part test produces different answers in different businesses. An assistant who thrives on a founder's inbox might struggle on a busy customer service desk, not through lack of ability but because the rhythm of the work is so different. Seeing the test applied makes the point concrete.

Same test, four different assistants. That is the whole case for writing the brief before you look.

Signs the fit is working, or not, in the first 30 days

Even a well-defined role needs a few weeks to settle. The first month is where small habits either build trust or quietly wear it down, and most of the warning signals are visible long before anything goes seriously wrong.

Review against the brief at the end of week two and again at day 30. Keep each review short, specific and two-way, so the assistant can tell you where the brief itself was unclear.

The right assistant is the one your business was ready for

A second search feels riskier than the first, because you have seen how it can go wrong. It is also, quietly, the stronger position to search from. You now know which tasks drained you, which habits frustrated you and which hours actually mattered. That knowledge is the raw material of a good brief.

Owners who make delegation work rarely stumbled on a rare candidate. They described the job honestly, tested it on real work and kept the conversation going after day one. Fit was something they built on purpose.

For owners who would rather have the matching, trial and replacement handled for them, a managed provider such as Umbrella Connect follows that same sequence by design. Either way, the brief comes first. Write the page, and the right assistant becomes far easier to recognise.