How many candidates does it take to make one hire? Real pipeline data

The whole funnel, in one table

These are the cumulative totals across 46 filled vacancies in 18 sectors. Every figure is a count of real candidates on real roles that closed as filled.

Read as conversion rates between stages, the same data says:

• 15% of everyone sourced is worth presenting
• 61% of presented candidates get a client interview
• 36% of client interviews end in a hire
• 22% of presented candidates are eventually hired

What a typical role looks like

Totals are skewed by the hardest searches, so the medians matter more. For a typical filled role in this set:

• 26 candidates sourced and screened
• 4.5 presented to the client
• 3 client interviews
• 1 hire

(Medians across 46 roles, so a half appears where the middle falls between two roles.)

The range around that median is enormous, and it is the most useful thing in this article. One role in the set was filled from a single sourced candidate — a Registered Manager in Manchester who was the right person first time. Another needed 152 candidates worked to produce one hire. Both were filled on the same fixed fee.

What each drop-off is actually telling you

Sourced to internal interview (36% survive)

Two thirds of the people who look right on paper do not survive a conversation. They have moved, the commute does not work, the salary expectation is 20% above the brief, or the CV describes a job they held for two months. This is the stage that clients almost never see, and it is most of the work.

Internal interview to presented (42% survive)

This is the screen that decides whether a shortlist is worth reading. Sending everyone who interviews well is not shortlisting — it moves the work to the client and calls it service.

Presented to client interview (61% survive)

If this rate collapses, the brief and the search have come apart: the candidates are competent but not what was actually wanted. Two or three consecutive rejections at this stage is a signal to re-take the brief, not to send more CVs.

Client interview to hire (36% survive)

Roughly one in three client interviews ends in a hire. Which means a role needing three interviews to close is normal, not a problem — and an employer who expects to interview one person and hire them is planning for the exception.

Why this matters for cost

A percentage fee is charged on the outcome. A flat fee is charged on the same outcome, but the 1,542 conversations behind it cost the same whether the role is easy or hard.

The economics are simple. A role that takes 26 candidates and a role that takes 152 both produce one hire. On a 20% fee they are billed identically — which is why a hard role is the one that quietly slips down a consultant's list. On a flat fee, the difficult search is absorbed at our end, and the refund if the role is not filled is what makes that a real commitment rather than a slogan.

Using these numbers on your own hiring

If you are running the process in-house, the funnel gives you a planning model:

1. Work backwards from one hire. Expect to need three or four genuinely qualified people in front of you, not one.
2. To get four qualified people in front of you, expect to screen in the region of 25, and to have sourced or received considerably more than that.
3. If you are seeing plenty of applicants but nothing worth interviewing, the problem is upstream — the advert, the title or the salary — not the volume. Why am I getting no applications? covers the nine usual causes.

And if that volume of screening is not something you have the hours for, that is precisely the work being bought when a role is handed over — 26 conversations so that four are worth your diary.

Reading your own funnel

You can run this diagnostic on your own hiring without any special tooling. Count four numbers for the last role you filled — applicants or people approached, people you screened, people you interviewed, offers made — and compare the shape rather than the size.

A wide top and a narrow middle means reach is fine and relevance is not. The advert, the title or the salary is attracting the wrong people. Fixing the advert is cheaper than any change of supplier.

A narrow top all the way down means reach is the problem. Either the role is not visible where the right people look, or the right people are not looking at all.

A healthy middle and no offers is the most expensive shape, because you have paid for all the work and have nothing. It usually means one of three things: the interview process is losing people through delay, the offer is below what the market pays, or the brief and the reality of the job have drifted apart.

Offers made and declined is worth counting separately. Two declined offers is not bad luck; it is market feedback about money or about how the process felt.

One caveat about our numbers

Every figure here comes from roles that were filled. It is a funnel of successes, and it does not include searches that were abandoned or roles pulled by the client — so it describes what filling a role took, not the odds of any given role being filled at all.

That is the right dataset for planning a hire. It would be the wrong dataset for claiming a success rate, which is why we have not.

Frequently asked questions