How to Evaluate a Recruiter's Track Record Before You Engage Them

Why references tell you so little

Ask an external recruiter for references and you will get two or three happy clients. That tells you the recruiter has placed at least two or three people successfully, which was never really in doubt.

What you actually need to know is the shape of their whole record, not its best slice: how many roles they take on versus how many they close, how relevant their submissions typically are, how quickly they respond when a hiring manager has a question, and whether companies come back to them a second time.

None of that is visible in a reference call, and very little of it is visible in a pitch deck. It has to come either from structured questions you ask yourself, or from a platform that records it as a by-product of real work.

The questions worth asking directly

How many roles like mine have you worked in the last six months, and how many did you close? The ratio matters far more than the absolute count. A recruiter who closes three of five roles in your niche is more useful than one who has worked fifty across every function.

How many candidates do you typically submit per role? An unusually high number often signals volume-spraying rather than screening — and it shifts the screening work back onto your team.

Who will actually work my role? In consultancies, the person pitching is frequently not the person sourcing. Ask for the name and talk to them.

What is your most common reason for failing to close a role? A recruiter with a considered answer has reflected on their own misses. A recruiter who claims they always close is telling you they do not track it.

What do you need from us to be effective? Strong recruiters have clear asks — feedback within a set time, access to the hiring manager, clarity on non-negotiables. Vague answers here usually predict a vague engagement.

Signals that predict performance, and signals that do not

Weak predictors, despite how often they are cited: years in business, headcount, a long client logo wall, and the size of their database. A large database that has not been touched in two years is a liability, not an asset.

Better predictors: specialisation depth in your specific function and seniority band; submission-to-interview conversion; responsiveness during the process; and repeat engagement from the same clients, which is the clearest market verdict available.

The most useful question of all is often the simplest: how many of your clients have given you a second role? Recruiters rarely volunteer this and it is difficult to dress up.

What Hiringseed records, and how it is calculated

On Hiringseed, a recruiter accumulates a trust score on a 0 to 100 scale, derived from how they actually perform on the platform rather than from anything they tell you. It is recalculated every night.

Five inputs are weighted into it: selection rate contributes 30 per cent, resume relevance as rated by companies contributes 25 per cent, response time 15 per cent, consistency of activity 15 per cent, and candidate experience ratings 10 per cent. A dispute-rate penalty is then subtracted, capped so that it cannot sink an otherwise strong record outright.

The score only appears once a recruiter has at least ten submissions on the platform. Below that threshold there is not enough evidence for a number to mean anything, so none is shown — which is itself useful information when you are deciding who to engage.

Recruiters can also carry badges, each tied to a specific threshold rather than being awarded editorially: Top Performer requires a trust score of at least 80 across at least 20 submissions; High Success Rate requires a selection rate of 40 per cent or better across at least 15 submissions; Responsive reflects sustained fast response times; Highly Rated requires an average of 4.5 or better from at least five company ratings; and Veteran marks 100 or more submissions.

Run a paid trial before you commit

However good the signals look, the only reliable test is a live role. Give a new recruiter one requisition with a clear brief and a defined window, and watch four things: how long until the first submission arrives, how relevant that first batch is, how they respond to rejection feedback, and whether they keep you updated without being chased.

A pay-on-placement arrangement makes this trial close to free. If they do not deliver, you have lost the elapsed time but not the fee. That is also the reason it is worth running two recruiters in parallel on a difficult role rather than committing exclusively to one on the strength of a pitch.

Whatever you conclude, write it down. A short internal note on each recruiter — what they are good at, what they are not, which roles to send them next time — compounds into something far more valuable than any vendor list.