What should a nanny agency charge? UK fees in 2026, and what you actually keep

Most UK nanny agencies charge a one-off placement fee of 10% to 20% of the nanny's gross annual salary, or a flat fee of roughly £1,500 to £2,500 outside London and more inside it. Temporary placements are usually billed weekly, commonly £100 to £250 a week depending on length.

That is the short answer. The longer one matters more, because the headline fee is not what you keep, and the model you choose changes how your year behaves far more than the number does.

Everything below is drawn from UK agencies' own published fee pages in October 2026. It is not a survey. Agencies that publish their fees are a self-selecting group, and the ones that quote privately tend to be dearer, so treat these as the visible end of the market.

What do UK nanny agencies actually charge?

Three models are in common use, and plenty of agencies run more than one.

At the top of the market the numbers leave that range entirely. London consultancies placing into high-net-worth households publish permanent fees around £5,000, and percentage fees of 20% to 22% appear on worked examples for high-profile placements.

A worked example, because the percentage model is the one people get wrong. A nanny on £35,000 gross, placed at 15%, earns the agency £5,250. The same nanny placed at 12% earns £4,200. Three percentage points is a thousand pounds on one placement, which is why the percentage you quote deserves more thought than it usually gets.

Percentage or flat fee: which should you use?

The honest answer is that it depends on what your placements look like, and most agencies pick by habit rather than arithmetic.

Percentage rewards you for placing well. If you are placing experienced nannies into senior roles, a percentage captures that. It also means a salary negotiation between the family and the nanny moves your fee, which cuts both ways.

A flat fee is easier to sell and easier to run. The family knows the number before the search starts. You can quote it on the phone. You do not have to ask what the nanny is being paid, which some families treat as an intrusion. The cost is that a £5,000 placement and a £45,000 placement pay you the same, and the second one almost always took more work.

The weekly multiple suits part-time and live-out work, where an annual salary is either small or hard to pin down. It is also the model most likely to confuse a family, so it needs a minimum attached, which is why "5x weekly net, minimum £650" is the shape you see.

If you are starting out, a flat fee with a lower London band is the easiest thing to explain and the hardest thing to get wrong. Move to percentage when your placements are consistently senior enough that the flat fee is leaving money behind.

Do you have to add VAT, and when does it start to hurt?

Several agencies add VAT on top of the published fee, which puts 20% on the price a family actually pays. Several others do not, because they are not registered.

This is the single biggest step-change in a small agency's pricing, and it arrives without warning. Once your taxable turnover passes the VAT registration threshold on a rolling twelve-month basis, registration is compulsory, and from that point a £2,000 placement fee becomes £2,400 to the family. Your income has not changed. Your price has gone up by a fifth, against competitors who are still under the threshold.

The threshold moves, so check the current figure on GOV.UK's VAT registration guidance rather than trusting any number in a blog post, this one included. What matters for planning is the mechanic: it is rolling, not per tax year, and it is turnover rather than profit.

Two things worth knowing before you get near it:

• Your fee is the turnover, not the nanny's salary. An agency placing twenty nannies a year at £4,000 is turning over £80,000, not £700,000. Placement agencies tend to reach the threshold far later than they fear.
• Temp work counts, and it accumulates faster. Weekly temp billing adds up quietly, and an agency doing steady temporary cover can cross the line while permanent placements alone would not have.

If you are approaching the threshold, the decision is whether to absorb the VAT, pass it on, or raise fees ahead of time so the jump is less visible. Most agencies pass it on. The ones that plan for it a year ahead handle it better than the ones that discover it in a quarterly review.

What do you actually keep from a placement fee?

This is the number almost nobody publishes, and it is the one that decides whether the business works.

Take a £4,000 flat placement fee. Before it is profit, it pays for:

• The searches that did not place. Industry convention is that a reasonable agency fills somewhere between half and two thirds of the searches it takes on. If you fill two of every three, that £4,000 is really covering three searches worth of work.
• Vetting. Enhanced DBS checks, reference calls, right-to-work verification, qualification checks. The cash cost is modest. The time cost is not.
• The guarantee period. Most agencies offer a free replacement if a placement ends within a set window, commonly eight to twelve weeks. A replacement is a second search at zero additional revenue, and it is the single largest hidden cost in the model.
• Your own time. A permanent search that reaches offer is rarely under fifteen hours of work across sourcing, screening, scheduling, reference-taking and handholding both sides to a signature.

At fifteen hours for a filled search, plus the unfilled ones it subsidises, a £4,000 fee at a two-in-three fill rate is roughly £4,000 for 22 hours of billable-equivalent work. That is about £180 an hour before overheads, which sounds healthy until you take off insurance, software, advertising, and the searches that collapse at offer stage.

The number to watch is not your fee. It is fee divided by hours across every search you started, including the ones that went nowhere. Agencies that track only completed placements consistently overestimate what they earn.

What about temporary and ad-hoc work?

Temp is a different business wearing the same clothes. The margins per placement are small, the admin per placement is nearly as large as for permanent, and the cash arrives weekly rather than in lumps.

The reason agencies do it anyway is that it is the only part of the model with recurring revenue. A family that uses you for emergency cover three times a year is worth more over five years than a single permanent placement, and it costs nothing to acquire them after the first time.

The trap is pricing it like an afterthought. A flat £200 a week sounds fine until you are fielding six messages a day to fill a Thursday. If your temp work is not at least covering the coordination it generates, it is a marketing cost that you are calling revenue.

When should you raise your fees?

Three signals, in the order they usually appear.

You are filling nearly everything you take on. A fill rate above about 80% usually means you are being chosen on price. That is the cheapest time to raise fees, because the demand is already there.

You are turning work away. Obvious, and still routinely ignored.

Your fee has not moved in two years. Nanny salaries move. If your percentage is fixed you are already tracking the market, but a flat fee quietly falls behind it, and a flat fee set in 2024 is worth noticeably less today.

What does not work is raising fees quietly and hoping nobody asks. Families compare, agencies talk, and a fee increase you cannot explain in one sentence reads as opportunism. A fee increase you can explain, usually that your vetting got more thorough or your guarantee got longer, lands fine.

What this means for how you run the agency

Two practical things follow from the arithmetic above, and neither of them is software.

First, track every search, not every placement. Your real rate is set by the work that did not convert. If the only thing you record is filled placements, your fees will drift below what they should be and you will not see it happen.

Second, know your guarantee period's real cost. It is the one term agencies compete on without pricing it. Extending from eight weeks to twelve sounds generous and cheap. It is generous. Whether it is cheap depends entirely on how often your placements fail in weeks nine to twelve, which you will only know if you have been recording why placements end.

That second one is the harder of the two to do by hand, because it needs a reason written down every single time a placement finishes, including the ones that end quietly. It is the reason our own placement tracking asks for a reason rather than offering a notes field: an optional field gets filled in when somebody has time, which is never the week a placement falls over.

Both of those are just record-keeping, and both are the kind of record-keeping that falls apart in a spreadsheet once you are past a few dozen searches a year. That is the point at which agencies start looking for a system, which is the business we are in, and we would rather you arrived at that conclusion from your own numbers than from an advert.

See also: what US nanny agencies charge, which covers the same ground for a market with no VAT but a harder employment-status question.

Figures in this post come from UK agencies' published fee pages, read in October 2026, and from published guidance on VAT registration. They describe what agencies advertise, not what every family pays. If your experience of the market differs, we would genuinely like to hear it: hello@rostavo.com.