The recruitment metrics that actually predict agency revenue

Walk into most recruitment agencies and you will find a dashboard on the wall counting calls, emails, and CVs sent. These are activity metrics. They tell you how busy the team is. They tell you almost nothing about whether the business will bill more next quarter than it did this one.

The distinction matters because activity and outcome decoupled a long time ago. A recruiter making 80 calls a day in 2016 was generating conversations. A recruiter making 80 calls a day in 2026 is mostly generating voicemail. The same is true of CV volume: sending more candidates to a client used to signal effort, and now it signals that your screening is not working. When the input metrics stop correlating with the output, continuing to manage by them is not rigour — it is superstition.

What follows is the set of metrics that do predict revenue, why each one earns its place, and roughly where the market sits on them. Treat the benchmarks as orientation rather than targets: they vary enormously by sector, seniority, and perm-versus-contract mix, and the useful comparison is almost always against your own trend rather than against an industry median.

The two ratios that expose screening quality

Submit-to-interview ratio is the number of candidates you submit for every one the client agrees to interview. It is the cleanest available measure of whether your shortlists are credible. Industry reporting puts a healthy agency figure at roughly 3:1 — three submissions per interview secured.

When this ratio drifts to 6:1 or 8:1, agencies usually respond by telling recruiters to submit fewer candidates. That treats the symptom. A poor submit-to-interview ratio almost always means one of three things: the brief was never properly qualified, the search is surfacing the wrong profiles, or the presentation is not making the case for the candidate. Each has a different fix, and you cannot tell which one you have without looking at the rejections.

Interview-to-offer ratio measures what happens once your candidate is in the room. Around 3:1 is a common benchmark. If your submit-to-interview is strong but interview-to-offer is weak, the candidates look right on paper and underperform in person — which points at assessment depth rather than sourcing.

Read together, these two ratios localise the problem to a specific stage. Read separately, they are just numbers.

Fill rate: the metric clients actually judge you on

Fill rate is positions filled divided by positions taken, expressed as a percentage. It is the number your client uses to decide whether to keep giving you roles, and it is the number most agencies are least willing to look at honestly, because improving it sometimes means taking fewer roles.

That is the uncomfortable insight buried in fill rate. An agency working 40 roles at a 25% fill rate and an agency working 15 roles at a 65% fill rate can bill similar amounts, but they are very different businesses. The second one has recruiters who know their roles deeply, clients who get a real service, and a cost base that is not being burned on work that will never convert. The first has a pipeline that looks impressive in a board pack and a team quietly triaging by gut feel.

Track fill rate by client, by consultant, and by role type. The aggregate number is nearly useless; the segmented view tells you which clients to invest in, which to renegotiate, and which to resign.

Time-to-fill, and the two clocks inside it

Time-to-fill is the elapsed days between a role opening and a candidate accepting. Recent benchmark reporting puts the general market figure at around 44 days, though agency perm figures vary widely by sector.

The headline number is less useful than its two components. Speed-to-shortlist is the time from brief to first credible submission — the part you control. Client decision time is everything after that — the part you influence but do not own. Agencies that only measure the total tend to blame themselves for delays that sit entirely on the client side, or vice versa.

Splitting the clock changes the conversation you can have. "Our average time-to-fill is 41 days" invites a shrug. "We shortlist in four days and then wait eleven for feedback" is a specific, evidenced request for a service-level agreement — and it is the sort of thing that gets fixed. We covered the operational side of this in how to reduce time-to-fill.

Revenue per recruiter and the productivity question

Revenue per recruiter is total billings divided by fee-earning heads. It is the metric that tells you whether growth is coming from leverage or from headcount, and it is the one that changes most sharply when technology is working.

This is where the 2026 data gets interesting. Industry research this year found that agencies using AI at some stage of the recruitment process were 3.5 to 4.5 times more likely to have grown revenue in the preceding year than those that were not. That correlation does not prove causation — better-run agencies adopt tools faster, and the causality runs both ways. But the size of the gap is hard to dismiss.

The mechanism, where it holds, is straightforward: when parsing, search, shortlist preparation, and status communication stop consuming recruiter hours, the same head can carry more live roles without the fill rate collapsing. That is leverage. Hiring another recruiter is capacity. Only one of them improves margin.

Response rate: the leading indicator everyone under-measures

Outreach response rate is the earliest signal in the funnel and the fastest to react to changes in the market. Recent sourcing benchmarks put cold email response at around 3.4% across sectors — a number that surprises agencies who have not measured it and confirms what candidates have been saying for years about generic outreach.

Because it sits at the top, response rate moves weeks before billings do. A sustained fall is an early warning that either your messaging has gone stale, your targeting has drifted, or your market has cooled. Treat it as the canary, not the score.

Quality-of-hire and the metric nobody tracks

Quality-of-hire is the placement's performance after they start — typically measured by retention at 6 and 12 months, hiring manager satisfaction, or time-to-productivity. It is the hardest agency metric to collect because it requires the client to tell you something after your invoice has been paid.

It is also the most commercially valuable one you can hold. An agency that can demonstrate its placements stay longer than the client's other channels has an argument that survives a procurement conversation. An agency that cannot is competing on fee percentage, which is a race with only one direction.

The practical route in is the rebate period you already have. You are contractually interested in whether the placement lasts 12 weeks; extend the same check-in to 6 and 12 months and you have a dataset nobody else in the pitch will have.

What to actually put on the dashboard

Eight metrics is more than most teams will look at daily. A workable split is a weekly operating view and a monthly strategic one.

• Weekly, per consultant: live roles worked, speed-to-shortlist, submit-to-interview ratio, outreach response rate.
• Monthly, per client and per consultant: fill rate, time-to-fill split into both clocks, revenue per recruiter, and offer-decline reasons.
• Quarterly, per client: quality-of-hire signals, margin by client, and the roles you took but never filled.

That last item deserves its own line. Unfilled roles are the largest hidden cost in most agencies and almost never appear on a dashboard, because there is no invoice attached to them. Counting them, and attributing the hours spent, usually reframes an entire client portfolio.

A note on benchmarks

Every figure quoted above comes from published 2026 industry reporting, and every one of them hides enormous variance. A contract IT desk and a retained executive search practice have almost nothing in common at the metric level. Executive search fill rates look catastrophic next to volume staffing; volume staffing revenue per head looks thin next to executive search.

Use external benchmarks to spot the metrics you are not tracking at all. Use your own historical trend to judge whether you are getting better. The agencies that improve fastest are not the ones that hit an industry median — they are the ones that measured a stage honestly, changed one thing, and measured it again.