Split Fee is live for UK recruitment agencies.

← Back to blog

Why Overlap Beats Size in a Split Fee Network

Split Fee Team ·

Ask what makes a split fee network useful and most people answer with a number: how many agencies are on it. It is the wrong measure, and it is worth knowing why before you judge any network by it.

A network needs overlap, not volume

The instinct is that a network needs thousands of members before it is worth joining. It does not. It needs your candidate to sit next to someone else’s vacancy.

That is a much narrower requirement, and it turns on specialism rather than scale. The UK has over 30,000 recruitment businesses, the vast majority of them small and focused on one or two sectors. A twelve-person agency doing IT contract work in the North West does not need twelve thousand counterparties. It needs a handful of agencies working adjacent desks with clients it does not have.

Overlap between two specialist agencies is worth more than breadth across a hundred generalists. Ten agencies working the same sector produce more usable matches than a thousand spread thinly across every industry in the country, because a match needs two specific things to line up, not two hundred vague ones.

So the useful question is not “how many agencies are on it”. It is “how many are working near me”, and that number starts mattering far sooner than the headline count.

The arithmetic underneath it

Here is the thing that is true of every agency at once, which is why the model works at all.

Your database fits more roles than your client list contains. You have people you could place tomorrow if the right vacancy existed, and the right vacancy does exist, at an agency that has never heard of you. That agency is in the same position in reverse: their client wants someone your database already holds.

Neither of you is short of ability. You are both working with a partial view of the market, because that is what a client list is. The gap between what your candidates could do and what your own clients happen to need is not a failure of your desk, it is arithmetic, and it applies to all 30,000 of those agencies simultaneously.

A split fee network is the mechanism for closing that gap without either agency giving up the relationship it owns. What split fee recruitment is covers the model itself in more depth.

What adding a record actually involves

This is the part that decides whether the theory is worth acting on, and it is usually skipped.

Adding a candidate takes a couple of minutes. You enter the professional detail, the job title, skills, seniority, salary range and location, alongside the name and contact details you already hold. What another agency sees is narrower: the professional facts, and nothing that identifies the person or your client. Those identifiers are stored rather than shared, and stay hidden until both sides accept a match. That is pseudonymisation rather than anonymisation, and it is what keeps the early stage low risk.

The matching runs automatically from there, scoring your record against every live vacancy on the platform and continuing to score it against every new one as other agencies add theirs.

That last part is worth sitting with. A record you add is not a one-time query against today’s inventory. It stays in the pool and gets scored again every time somebody adds the other half, so the work of finding the match is not yours and does not need repeating.

And the commitment is small by design:

  • You choose what goes on. Nothing is shared automatically. A retained candidate, or a client with contractual restrictions, simply does not get added.
  • Nothing identifying is visible. Names, contact details and client identities stay hidden until both agencies have accepted a match, so an unaccepted record tells another agency nothing about who you work with.
  • You can decline everything. A match is a proposal. If you do not like it, that is the end of it and no data has moved.

How to judge a network

If headcount is the wrong measure, here is what to look at instead.

Concentration, not size. Are there agencies working your sector and your region, or is the membership spread thin? A directory of thousands across every industry is worth less to a specialist desk than a dozen in the same niche.

Whether entry is controlled. Every agency on Split Fee is verified before its account goes live, because a network you share candidates into is only as good as the agencies in it.

Whether matching actually runs. Plenty of arrangements are directories: they introduce two agencies and leave the matching, the terms and the invoicing to you. Automatic scoring is a different proposition from a member list and a forum.

What happens to your data before anyone agrees. The point at which identifying details change hands is the point at which your risk starts. Later is better.

The short version

Judge a split fee network on whether anyone in it works your kind of desk, on whether entry is controlled, and on how much it does for you after the introduction. Those decide whether it produces placements. Headcount mostly decides how impressive the homepage looks.

Create your account and add one candidate. That is the whole test, and it takes about as long as reading this did.