Retained vs. Contingency Search: What Actually Differs

The usual explanation of retained versus contingency search stops at the invoice: retained is paid in installments, contingency is paid on placement. Accurate, and it misses the part that actually changes the work.

The consequential difference is where the risk sits. And because research is the most expensive part of a search and happens before anyone knows whether a placement will occur, who carries that risk determines how much research each model can afford.

The structural difference

	Retained	Contingency
When the firm is paid	Across the search, on an agreed schedule	On placement only
Who carries the risk	The client	The search firm
Exclusivity	Normally exclusive	Normally not — several firms may work the same role
Fee, relative	Higher — priced against exclusivity and a funded process	Lower — priced against competition and placement risk
What the client is buying	A search process	A successful placement
Where the incentive points	Coverage of the market	Speed to a submittable candidate

Actual percentages vary enough by market, sector and seniority that any single figure would be misleading — ask the firm you are talking to, and ask what it is a percentage of.

The last row of that table explains almost everything else.

Why the payment timing changes the research

Under a retained arrangement, the firm is paid across the search regardless of outcome. Research hours spent in week two are funded whether or not the placement closes. That makes it economically rational to map the market properly — to build the pool out before evaluating anyone, to work the adjacent sectors, to spend hours on someone who will turn out not to qualify.

Under contingency, every research hour is speculative, and it is speculative against competition — another firm may fill the role first, in which case the hours return nothing. Follow the incentive and the likely response is predictable: move quickly to candidates who can plausibly be submitted, which in practice tends to mean candidates who are visible, reachable, and probably already looking.

Neither is a failure of professionalism. Both are firms responding sensibly to how they get paid.

The consequence is predictable. Contingency work skews toward the active market. That leaves out the executive who is not looking: the one doing the job you want them to do, well, with no reason to keep a profile current. At senior levels, that person is often the whole point of the search.

What each model is genuinely for

Retained fits when the pool is small or hard to see, when the role is genuinely senior, when confidentiality matters, when the client needs the market properly mapped rather than sampled, and when a wrong hire is expensive enough to justify the process.

Contingency fits when the pool is deep and visible, when speed matters more than coverage, when the client wants optionality across several firms, and when they are unwilling to commit fee before seeing candidates.

Most of the arguing happens in the middle band — roles senior enough that coverage matters but not senior enough that the client feels the retainer is obviously justified. That is a real judgment call about how visible the pool is, not a question with a general answer.

Follow the incentive and the pattern is predictable, which is not a judgment on any firm: the executive who is not looking is the hardest one for a speculative search to justify reaching.

The exclusivity effect, which is underrated

Exclusivity changes candidate conversations more than most fee comparisons acknowledge.

When several firms work the same role, a candidate may be approached about the same position two or three times, sometimes with slightly different descriptions of it. Some candidates conclude the client is disorganized. Some conclude the role is hard to fill. Either way, the client’s employer brand takes the cost, and it is the kind of cost nobody attributes to the fee model afterwards.

Exclusivity also allows a firm to be honest with a candidate about the mandate, because there is one version of it.

What is the same in both

Worth stating plainly, because the fee-model debate can imply otherwise: the research obligations do not change.

A shortlist from either model should carry the same properties. Claims that trace to nameable sources rather than “public sources.” Criteria evaluated one at a time with confirmed, unconfirmed, or flagged as the possible verdicts. Laterals surfaced with the trade-off stated rather than filtered out silently. Length that follows the evidence rather than an expectation.

The fee model determines how much research is economically viable. It does not lower the standard for whatever research does get done. A contingency search that covers less of the market is defensible; a contingency search that asserts unsourced claims is not.

The question underneath the fee question

The fee comparison is usually a proxy for a different question the client has not asked directly: how much of the market will actually be covered, and how will I know?

That is the answerable version. Ask how the pool is assembled and what it leaves out. Ask whether adjacent sectors get mapped deliberately or incidentally. Ask about someone on a recent shortlist who missed a stated requirement. Ask what a claim in the shortlist is sourced to, and check one.

Those answers tell you what you are buying. The fee structure only tells you when you pay for it — and increasingly, what the research stages cost to run at all is changing faster than the fee conventions built on top of them.

Related reading: how one smaller metro’s firms split on model, how one metro’s firms describe their own model, what a headhunting fee actually pays for, what a retained executive search firm actually does, headhunter versus recruiter, and the executive search process, step by step.