Most businesses hire one virtual assistant and stay there
Expansion patterns across 113 client companies, and why the average is the wrong number to look at.
This article was contributed by Alon Pearl, Founder and CEO of VA Masters. He founded the firm in 2021 after hiring a virtual assistant remotely to solve his own bandwidth problem, and now writes on hiring at volume, drawing on the 166,350 applications the firm processed between January 2025 and July 2026.
At a glance
- Expansion is not gradual. 83.2% of clients run exactly one assistant and stay there, while a thin tail builds a small team.
- Recruiting again is not the same as growing. 31.9% of clients have recruited more than one role, but only 16.8% run two or more assistants today.
- 12% of clients hold 30% of all active placements, at a median of 371 hours a month. That is roughly two and a quarter full-time equivalents each.
- The median client uses 84 hours a month, almost exactly the half-time engagement most clients start at. The typical client does not scale up.
All figures cover 113 client companies with at least one active placement, as of September 2026. This is the most heavily qualified analysis we publish, and the limitation is set out in full below rather than in a footnote.
The number everyone quotes, and why it misleads
Agencies in this market like to talk about how clients grow their offshore team. It is a flattering story and the average engagement size appears to support it.
We looked at the actual distribution across 113 client companies with at least one active placement, as of September 2026. The average is 1.34 assistants per client. The median is one.
Those two numbers are doing very different work, and the gap between them is the whole finding.
How many assistants a client actually runs
| Active assistants | Share of clients |
|---|---|
| One | 83.2% |
| Two | 8.0% |
| Three | 5.3% |
| Four or more | 3.6% |
113 clients with at least one active placement, September 2026. The largest single client in the group runs nine.
The distribution is the finding. This is not a bell curve with most clients somewhere in the middle. It is a large group sitting at exactly one and a thin tail stretching out.
For anyone selling into this market, or buying in it, that matters. The common pitch is that a first hire naturally leads to a second.
In this data it mostly does not. Expansion either happens or it does not, and there is very little in between.
Recruiting again is not the same as growing
| Measure | Share of clients |
|---|---|
| Have recruited two or more roles | 31.9% |
| Run two or more assistants today | 16.8% |
113 clients with at least one active placement, September 2026.
Close to a third of clients have opened more than one role with us. Only half of those are running more than one assistant right now.
That gap is worth sitting with, because it is the number most easily presented as something it is not. A meaningful share of repeat recruitment is a client replacing or changing a role, not adding to a team.
Reported as a single expansion rate, those two behaviours look identical, and they are not.
We publish both figures rather than picking the flattering one. If you see a single expansion percentage quoted anywhere in this market, it is worth asking which of these two things it counts.
Where the volume actually sits
| Client group | Avg assistants | Median hours a month | Share of all placements |
|---|---|---|---|
| Highest engagement | 3.21 | 371 | 30% |
| Standard | 1.25 | 167 | 20% |
| Small | 1.00 | 84 | 36% |
| New | 1.05 | 40 | 15% |
113 clients with at least one active placement, September 2026. Groups are an internal engagement classification assigned automatically, not a commercial tier clients select. Placement shares are rounded and total 101%.
12% of the client base holds 30% of all active placements and runs at a median of 371 hours a month.
At the other end, the largest group by count runs exactly one assistant at a median of 84 hours a month, which is the standard half-time engagement almost unchanged from where those clients began.
The overall median across all clients is 84 hours a month and the mean is 151. The distance between those two numbers is the concentration described above.
It lines up with what we see at the point of hiring: in our analysis of what businesses actually outsource, 46% of roles are opened at exactly 80 hours a month. Most clients start at half-time and most stay there.
The interesting question, then, is not how to lift the average. The average is pulled almost entirely by a handful of companies.
The question is what separates them, and this data cannot answer it. Company size, sector and the kind of role recruited first are the obvious candidates, and we do not hold clean enough data on the first two to test them.
The limitation, stated plainly
This analysis covers clients currently active. It holds no record of companies whose engagement has ended, so every expansion figure here is calculated over businesses that are still with us.
That will overstate expansion relative to the full history of the business. We have not attempted to correct for it and we publish no retention or churn figure.
There is a second effect on top of it. A large share of our current clients signed within the last year, and a company that signed three months ago has had less opportunity to add a second assistant than one that signed two years ago.
These figures measure opportunity as much as behaviour.
We are publishing anyway because the concentration finding holds regardless of both. A small group accounting for a disproportionate share of volume is not an artefact of who stayed or how long they have been here.
The expansion percentages should be read as indicative rather than precise, and we would rather say that than let a reader assume otherwise.
What we looked for and did not find
We could not test whether company size predicts expansion. The employee-count field in our system is not currently clean enough to support the comparison.
Work to resolve it is underway and this cut should be possible in a future edition. It is the most interesting missing piece here.
Number of roles recruited does not cleanly predict number of assistants running. As the section above shows, the two figures diverge, and a client who has recruited three times may be running one assistant today.
We could not test retention or churn. For the reason set out in the limitation above. No figure of that kind appears anywhere in this analysis.
We report these because a relationship that turns out not to exist is still a result, and because it shows which questions were asked rather than only which ones produced an answer.
About this data
This report describes what passes through a single recruitment pipeline, not offshore hiring as a whole. Every dataset of this kind reflects the business that produced it.
It also reflects the age of the relationships inside it: this is a snapshot of a young and fast-growing client base, not a mature one.
Combined with the survivorship limitation above, this is the most heavily qualified analysis in our current set. We publish it because the concentration finding is clear and holds regardless of those caveats.
Treat it as a detailed view of one pipeline at one moment, not a measurement of the market.
Monthly hours are the hours billed in the most recent complete month, which is the closest available measure of current engagement size.
Expansion figures are calculated only over clients who have completed at least one placement, since a client without a first assistant cannot have a second.
Roles and records belonging to the agency itself are excluded throughout, as are client names and any client-level financial data.
Full methodology sits alongside our Filipino virtual assistant workforce dataset, and our analysis of why candidates do not get hired covers the recruitment side of the same pipeline.
Independent
