Revealed preference: the JD vs. what your client actually buys
Job descriptions are wish lists written by committee. What a client accepts is revealed by who they reject and who they hire. The rejection pattern is the truth — and reading it is the most valuable thing on your desk.
Economists have a clean idea for a messy problem. You don’t learn what people value by asking them. You learn it by watching what they choose when it costs something. They call it revealed preference, and it is the single most useful lens a staffing recruiter can carry into a client relationship.
Because a job description is not what the client wants. It’s what a committee could agree to write down. It’s padded with must-haves that are really nice-to-haves, a stack list copied from the last req, and a seniority band set by budget rather than need. Treat it as the spec and you’ll source against fiction.
The “no” carries more than the “yes”
A placement tells you one profile worked once. A rejection tells you where the line is. The client who passes on a strong candidate is handing you a coordinate — too senior, too expensive, wrong domain, available too late, a stack their lead doesn’t trust. Stack enough of those coordinates and the real shape of what they buy appears, often nothing like the JD.
The brief says “React, 5+ years, fintech a plus.” Over a quarter the client passes on two senior React devs (“not quite the right level of ownership”) and hires two who were lighter on React but had led a small team through a regulated launch.
The revealed preference isn’t “React.” It’s someone who has owned delivery under compliance pressure. The JD never said that. The rejections did.
A playbook you can run on Monday
- Log the reason, not just the result. “Rejected” is noise. “Rejected — rate €30 over, and they wanted someone who’d stay past the contract” is signal. Capture it while it’s fresh.
- Keep it per client. Revealed preference doesn’t generalise. The same profile that one client calls over-qualified, another calls exactly right. The memory has to be client-shaped.
- Watch the gap. When the stated must-haves and the actual hires drift apart, trust the hires. Then — gently — tell the client what you’re seeing. It’s the most consultative thing you can do.
- Date everything. A ceiling from January may not hold in June. Preferences move with urgency, headcount and the market.
The honest caveat
Revealed preference is an observation to confirm, not a law. One rejection is an anecdote. People change their minds; a new hiring manager resets the pattern; a desperate quarter breaks every rule you thought you’d learned. The discipline is to hold these as hypotheses — “this client seems to reward domain over years; let me check that against the next two” — not as verdicts about what they’ll always do.
The JD is what the client asked for. Revealed preference is what they paid for. When they disagree — and they usually do — the cheque is the truth.