The backdoor hire: how often it happens, and whether you can collect

You find out months late. Someone you submitted in March is on the client's team page in September, and nobody called you.

The instinct is to assume you got sloppy. The only measured study we could find says otherwise, at least at the agencies it looked at: across 56 databases, all of them at firms with more than 50 employees, 84% of cases started on the client's side and 16% were agency process failures.

Cause and recoverability point in opposite directions, though. They caused it. Whether you can do anything about it runs through three things on your side, and this is the order to take them in, because the first two can end the question before the third begins.

Contract and licensing rules vary by US state and this is not legal advice. Every legal passage below ends in a question for a lawyer in your own state, and that is deliberate.

What decides whether you can collect

First, are you licensed where your state requires it. Employment agencies are regulated at state level in the US, and in at least one state the consequence of getting it wrong is severe. Washington's RCW 19.31.245 makes registration or licensing a prerequisite to suit by an employment agency, and gives anyone who paid an unregistered or unlicensed agency a cause of action against it for treble damages plus attorney's fees.

Two things about that before you either relax or panic. The cause of action runs to whoever paid the agency, which in a backdoor hire is not the client refusing this fee, but could well be the same client on placements it did pay for. And whether an employer-paid contingency search firm falls inside that chapter's definition of an employment agency at all is exactly what a defence lawyer would argue about first.

That is one state's statute, not a national rule. The route to your own answer is short: your state's labour or licensing department publishes what it regulates and who has to register, and a search for your state plus "employment agency licensing" reaches it. Do that before you need it. It is the cheapest thing on this list to get right and the only one that can end the argument before it starts.

Second, were your terms in place at the introduction, and does the hire fall inside them. The validity period is the part that gets forgotten: an introduction has a shelf life set by your own terms, and a hire after it expires is a much weaker claim however clearly you made the introduction. Nobody publishes what a normal period is, so the only number that matters is the one in your agreement. Go and read it, today, before you have a case.

Third, were you the efficient procuring cause of the hire. This is the limb an employer's lawyer will actually argue, and the one least likely to be in your head when you write the first email. The next section is entirely about it.

And one more question, early, because it has a deadline in it. Contract claims carry limitation periods, they are set by state law, and they vary. Washington, to stay with the same state, gives six years on a written contract under RCW 4.16.040. Others are shorter, and when the clock starts is its own question. Ask a lawyer in your state what yours is, because everything about this subject encourages delay: you find out late, the sum is large enough to think about, and the section below is going to give you real reasons to sit on a live account. Waiting can be the right call. Waiting without knowing what you are running out of is not.

Worth separating one thing from these three. Illumini, whose data runs through the rest of this article, names its own three conditions for whether a case turns into cash: terms in place, hire inside the validity period, and whether the agency is willing to open a dispute with a client it still trades with. Two of those are the same as the second condition above. The third is not legal at all, it is commercial, and Illumini's read is that it decides more cases than the paperwork does. Hold it separately, because you will agonise over it separately.

"But for" is their argument, not yours

US courts have applied a test called efficient procuring cause to whether an employment agency earned a placement fee. The formulation quoted in the column below, from a letter written by an employer's lawyer, is that the recruiter must establish that it caused the candidate and the employer to negotiate and that the candidate was hired as a result of its efforts.

The case named for that proposition is Michele Matthews, Inc. v. Kroll & Tract, 275 N.J. Super. 101 (App. Div. 1994), and later New Jersey opinions cite it at page 106 for the standard. We have not read the opinion itself, only the citing record and a trade column in which a lawyer's letter invokes it, so take it as a place for your lawyer to start rather than a holding we have verified. One case in one state is not the law of the country. Worth knowing too that the doctrine is not recruiting-specific: the two later New Jersey opinions citing it are real-estate broker commission disputes running the same test.

The placement-law writer Jeff Allen made the practical point about it in an ERE column first published in December 2014, answering a reader's question. Recruiters reach for a "but for" argument, meaning the hire would never have happened but for their introduction, and Allen's position is that this is the employer's argument rather than theirs. The column quotes an employer lawyer's reply: "the correspondence between our clients does not clearly support the allegations that [the recruiter] or his office were the efficient procuring cause of [the candidate's] employment."

The mechanism is what he calls intervening acts, the things that happen between your introduction and the hire. An employee referral, a resume already sitting in the client's database, a conversation at a conference. Allen splits them in two, and the split is the part that matters. An act that flowed from your introduction is a dependent one and does not break the chain. An act with an independent origin is what a client's lawyer will point at and say caused the hire instead. So the question is never whether things happened between your CV and the offer, because they always do. It is whether each of them came out of what you started.

Which tells you what to write down at the time, and it is a short list: the date the CV went over, the client's acknowledgement of it, and every subsequent step that traces back to that introduction. Interviews arranged, conversations you brokered, feedback relayed. Those are your dependent acts, and they are the answer to the defence.

Allen's own recommendation is to argue substantial cause rather than "but for". Note when weighing it that the same column sells his $125 fee-collection guide. The underlying observation costs nothing to accept either way: the fewer unexplained gaps between your introduction and the hire, the harder your claim is to break.

The sister-company gap

Here is the specific hole that produces cases nobody expected.

In the Illumini data, 23% of client-side cases involved the candidate being hired by a different legal entity in the same corporate group as the client the CV went to. Client-side cases are 84% of the total, so that is 19% of all cases in the study, just under a fifth. You submitted to the subsidiary; the parent hired.

Go and look at how your own terms define "client". If it is the company that signed them, then the entity that actually did the hiring is not covered. The fix is one clause: define the client to include affiliates, parents, subsidiaries and any entity under common control, and say that an introduction to one is an introduction to all of them. It costs nothing at the point where you are already negotiating terms. Adding it later is worth doing too, it just will not reach introductions you have already made.

When to claim, and when not to

A claim is a commercial decision with legal content, and the arithmetic is not only the fee.

Claim. When the client has stopped giving you work, because there is no relationship left to spend. When it is a group-entity hire and your terms cover affiliates, because that is the one fact pattern where the paperwork usually wins outright. And when the pattern looks repeated rather than accidental, because a client that does this twice will do it again and the relationship is worth less than you think.

Do not claim, or not yet. When your introduction sits outside the validity period, or your terms were not in place before the CV went over. Both put you at the weak end of an argument you are paying for with the account, and the honest position is that neither is automatically fatal and neither of us can tell which it is from here: whether an unsigned set of terms binds anyone, and whether anything survives an expired window, are questions a lawyer in your state answers in an hour and this page cannot answer at all. So the rule is not "give up". It is: get the position checked before you spend the relationship, not after.

And be slow about a live account. Illumini's read is that willingness to open a dispute decides more cases than the paperwork does, which is another way of saying most agencies quietly eat it. That is not irrational. Weigh a year of that account against a single fee, honestly, before you send anything, and remember the limitation period is running while you decide.

How often this happens, at your volume

The reason a solo desk should read all of the above and then not build a system around it.

Illumini, a company that sells backdoor-hire detection software, published a study on 7 August 2026 with a method section. It ran across 56 agency databases, matching CV send records against current employment data and confirming billing status with each agency: a one-year lookback per database, cases collected over two years.

What it found: 89% of those databases contained at least one missed fee, the median agency had three recovered in a single year, and the rate was one case per 2,139 CVs sent. The average recovered claim was GBP 13,200. That figure is in pounds because the study is, and it is left in pounds on purpose: converting it would imply a US benchmark nobody has measured.

Read the sample before the headline. Every agency in it has more than 50 employees, and the study describes them as typically sending over 10,000 CVs a year. Put your own volume through the same rate and the picture inverts.

That is one dataset's rate applied to your volume, not a forecast, and the rate comes from agencies nothing like yours. Treat it as an order of magnitude, and the order of magnitude is the point. A large agency builds a monitoring process because cases arrive continuously. At 800 CVs a year you are budgeting attention for something that may happen three or four times in a decade, and is worth a full fee each time.

So the answer is a habit rather than a system. Send your CVs in a way that produces a dated record with the client's own acknowledgement in it, because a dated introduction the client acknowledged is the first thing you will be asked to produce. Then twice a year, take the candidates you submitted in the last twelve to eighteen months who you know did not get hired through you, and check where they are now. If your sends sit in a searchable system that is a few minutes and a spot-check. If they live in your sent folder it is an afternoon, which is itself an argument for moving them.

What this rests on, and what would change it

One dataset. Illumini's study is the only measurement of backdoor-hire frequency we could find that publishes a method, and Illumini sells backdoor-hire detection software. We are not suggesting the numbers were shaped to fit the product. We are saying you should know who counted, and that there is no second count to check them against. The sample is 56 agencies, all its own clients, all over 50 employees, described only as global, denominated in pounds. Every number here that comes from it inherits all of that: the rate, the recovery figure, the causes and the group-entity share alike.

The legal side is narrower than it looks: a named doctrine, a case citation we checked against the record without reading the opinion, and two sections of one state's code. That is why every legal passage above ends in a question for your own lawyer. It is the honest edge of what those sources support, not throat-clearing.

So if you can produce a second measurement of how often backdoor hires happen, or a US case that moves the procuring-cause analysis for recruiters in your state, send it to us and we will update this page and say plainly what changed. We would rather be corrected than be one more page repeating a number nobody checked.

The wider fee mechanics are in our guide to recruitment agency fees. Which engagement model is worth taking at all is in contingency versus retained recruiting. And if you are setting up a desk from scratch, how to start a recruiting agency in the US is the wider checklist your terms of business sit inside.