I’ve watched founders run a retained search for a mid-level IC while three other agencies were working the same req. Six weeks. Zero offers. Forty thousand dollars committed in retainer and milestones before anyone got hired. Nobody told them it was the wrong model for that role — they did what most founders do and trusted the recommendation they were given. The incentive just ran the other way.
Worse: they paid the retainer without locking in real exclusivity. Nothing in the agreement stopped three contingency firms from quietly sourcing the same seat in parallel. That’s not a flaw in retained search — that’s a contract that didn’t do its job. You paid upfront for a guarantee you never actually got.
Most recruiters won’t tell you when retained is the wrong call. I will.
The mechanics, quickly
In technical recruiting, there are two dominant models: contingency and retained. Here’s how each actually works.
Contingency: you pay nothing until someone gets hired. The recruiter works at risk and collects a fee — typically 20–25% of first-year base — when you make an offer and it’s accepted.
Retained: you pay a portion upfront, with the remainder tied to milestones — often split into thirds at kickoff, shortlist, and hire. In exchange, the recruiter dedicates focused time to your search. That arrangement should include written exclusivity. If it doesn’t, you’re paying retained prices for contingency behavior.
On a $220k VP of Engineering role, a 25% contingency fee is $55k. Retained on the same role often runs a few points higher in practice — 28–33% isn’t unusual, sometimes with a minimum fee or a non-refundable kickoff payment built in. So retained can cost more in absolute dollars. The gap is usually smaller than founders expect going in, and it’s not really the point.
What retained buys you is dedicated time, real exclusivity, and a recruiter who’s committed before a single candidate shows up. What it costs you is upfront risk — a third (or more) of that fee is due before you’ve seen anyone. Contingency flips the risk: nothing due until someone’s hired, but no guarantee anyone’s working your req exclusively either.
That’s the structure. Here’s when each one is actually right.
Contingency works — until your process doesn’t
Contingency works when the role is clearly defined and the market for that skill set is reasonably active. Senior backend engineer. Full-stack IC with a specific stack. Frontend lead. Strong mobile developer. These roles exist in volume, and a good recruiter will work them hard because the path to a fee is clear.
It also works when you have a functioning interview process — calibrated interviewers, reasonable timeline, comp that’s competitive for your stage. If you’re not sure whether your process is ready, start here. On an open contingency search, the recruiter’s job is sourcing and screening. Everything downstream needs to be ready on your end.
One more thing worth naming: there’s a real middle option — engaged or exclusive contingency. Same no-upfront structure, but you commit to one recruiter for a defined window instead of running the role through three or four agencies at once. It’s what a lot of startups actually want. Open contingency across multiple firms breaks down because nobody invests real sourcing time in a coin-flip. You get whoever submits fastest, not whoever fits best. Naming the exclusivity, even without paying a retainer, fixes that.
I filled three senior engineering roles this way for a seed-stage startup: one recruiter, exclusive window, no retainer. Under 30 days. 5 profiles, 4 offers, 3 starts. The roles were defined, the founders moved fast, and exclusive contingency was right because the constraint was finding people, not the search strategy itself. You can read the full story here.
Retained is right when the search is the hard part
Retained makes sense when the constraint actually is the search itself.
The role is genuinely hard to fill. Not “we’ve been looking for a while” hard — the intersection of skills, seniority, and domain is narrow enough that a contingency search will produce a shallow pipeline. A CTO who has both technical depth and board-level presence is the classic example. Founders consistently hire the wrong stage of CTO — exactly the kind of mismatch a focused, retained search is built to catch. An engineering manager who has built distributed teams from scratch at a specific stage. A VP of Product with direct fintech or cybersecurity domain knowledge.
The search is confidential. Replacing a senior leader while they’re still in seat, or finding a CTO before you’ve announced a transition, requires focused effort and real discretion — not broadcasting the role across a recruiter’s full network hoping something surfaces.
The hire defines the next chapter. The retainer is a forcing function for both sides. It signals this isn’t a “let’s see what comes in” search. It commits the recruiter’s time. It commits yours.
None of this is just opinion. SHRM’s 2026 benchmarking data puts the median cost-per-hire for executive roles at $15,000 — up from $10,600 the year before. That’s a cost-per-hire figure, not a search-fee figure — it covers everything a company spends to fill a role. But the trend is real: filling senior roles keeps getting more expensive, which is why getting the model right matters more the higher up you go.
When retained is the wrong call
A Series A founder came to me wanting to go retained on a search for two senior engineers. The urgency was real — product deadline, board pressure, the works. But when we got into it, their job description was four sentences, comp was still being figured out with the board, and their technical interviewers were the CTO and one IC who were already stretched thin. I had a 45-minute call with them and told them I wouldn’t take the retainer. We spent three weeks cleaning up the process. Then ran a contingency search and filled both roles in five weeks.
Retained search works when the bottleneck is the search. If the bottleneck is your internal process, your comp structure, or your clarity on what you actually need — a retainer doesn’t fix that. It just adds cost.
If a recruiter is selling you retained on a well-defined IC role where two candidates are already in process, that’s information about their incentives, not your situation.
The actual decision rule
One question: what’s the real bottleneck?
If the market for this role is active, your process is ready, and you can give one firm a window — exclusive contingency. If you won’t give exclusivity, open contingency works, but expect a shallower pipeline. If the search is narrow, confidential, or company-defining — retained. If you’re not sure your process is ready, solve that first.
Whichever model you pick, ask three questions before you sign anything: what happens if the placement doesn’t work out, what happens to an unused retainer if the search stalls, and what happens if you fill the role from your own network mid-search. Those answers should be in writing before kickoff. If they aren’t, keep walking.
If you have a search open now, let’s talk. Twenty minutes and you’ll know which model fits — or whether there’s something else to fix first.
Leave a Reply