Every search firm eventually has the same conversation with a client. Why should I pay you anything before you find someone?
Your answer depends on which model you're working on. Most small firms do better once they know exactly when each one fits, and when to say no to a role that only works on terms you can't win.
How contingency fees work
On contingency, the client pays only if they hire a candidate you introduced. Nothing's due upfront. Fees are commonly 15% to 25% of the hire's first-year base salary, and 20% to 25% is typical for professional roles.
The client takes no risk, so they often brief several agencies on the same role. That's the cost to you. If three firms are working the search, you might do a third of the work for nothing two times out of three. Your fill rate tells you how often that's happening.
How retained fees work
On a retained search, the client pays to engage you and you run the search exclusively. Fees are commonly 25% to 35% of first-year total cash compensation. They're paid in installments, not at the end.
A common split is in thirds:
- 1One third when the client signs the agreement.
- 2One third when you present the shortlist, or on a fixed date such as day 30.
- 3The final third when the candidate accepts the offer.
Retained work suits senior, confidential or hard-to-fill roles, where the client wants a full map of the market and one firm that's accountable for the result. Clients accept paying upfront because they're buying a process, not just a name.
The middle option: engaged or container search
A lot of small firms use a hybrid. The client pays a smaller engagement fee upfront, often $3,000 to $10,000, and it's credited against the final fee. The search is exclusive. The rest is paid on placement.
It filters out clients who aren't serious. If a client won't pay a small engagement fee, they usually weren't going to give you the role exclusively either. And it still feels close enough to contingency that clients used to that model will often agree to it.
Which one should you offer?
- Contingency, when the role's common, the client already has several agencies on it, and you can fill it quickly from your own network.
- Engaged or container, when the role's specialized and you want exclusivity but the client isn't used to paying for search.
- Retained, when the role's senior or confidential, or the client wants a market map and regular reporting.
If you're not sure, ask the client how many other agencies are working the role. The answer usually tells you which model you're really in.
Put the model in writing
Whatever you agree, your terms should spell out the percentage, which compensation it applies to, the installment dates, what happens if the client cancels the search, and your guarantee.
Here's what that looks like in practice. A $150,000 retained search at 30% is a $45,000 fee, invoiced as three payments of $15,000. Each payment needs its own trigger in the agreement, and each one should go out the day that trigger happens. When the shortlist goes out, the second invoice goes out with it.
Track every installment the way you'd track a placement fee. If you'd rather have those dates on the same record as the search itself, that's what recruitment agency software is for.




