Staffing Agency Metrics That Matter

Two families, and only one tells you anything about next quarter. Leading indicators measure flow through the pipeline: roles taken, candidates submitted, submission-to-interview ratio, interview-to-placement ratio, time to submit. They move before revenue does, which makes them the ones to manage a desk on. Lagging indicators, including placements made, revenue billed and average fee, describe what already happened. They matter for reporting and for pay, but a consultant behind on them cannot fix them directly this week. The vanity category is anything counting effort without connecting to an outcome: calls made, emails sent, CVs uploaded, connections added. Activity numbers feel productive and are easy to game. A desk can make two hundred calls and submit nobody. One rule keeps a dashboard useful: every metric needs an owner, a decision attached to it, and a plausible number that would change what that person does tomorrow. If nobody would act differently on the figure, take it off the report. Recruitment metrics are for decisions, not decoration.

• Leading: roles worked, submissions per role, submission-to-interview, time to submit
• Lagging: placements, revenue, average fee, gross margin
• Risk: client concentration, redeployment rate, consultant dependency
• Vanity: raw call and email counts with no stage conversion attached

Submission-to-interview is the number of candidates you send a client divided by the number the client agrees to interview, measured per client and per role rather than across the whole business. It is the fastest read on qualification quality you have, because it tests whether your understanding of the role matches the hiring manager's. A low ratio almost never means the candidates were bad. It usually means the brief was thin, the client changed their mind after the brief, or the consultant submitted speculatively to look busy. Measure it per client, since the same consultant can be excellent with one hiring manager and lost with another, and an aggregate hides that completely. Watch the trend after any change to the intake conversation. If the ratio improves when consultants run a structured intake, you have found something worth enforcing across the desk. Track submissions inside your candidate management software rather than a personal spreadsheet, or the denominator quietly becomes whatever the consultant remembers.

• Measure per client and per role, never as one agency-wide number
• A low ratio usually indicts the brief rather than the candidates
• Compare consultants on the same client before drawing conclusions
• Watch the trend after any change to how intake is run

Interview-to-placement measures how many client interviews it takes to produce one placement, and it isolates a different failure from the submission ratio. Candidates are reaching the client but not converting. Causes cluster into three: the shortlist is technically fine but poorly matched on motivation, the client's own interview process is slow or inconsistent, or the offer stage keeps collapsing on money. Each has a different fix, and the metric alone will not tell you which one you have. Add a mandatory reason code at rejection and at offer decline, then read the codes monthly. If most losses happen after a second interview, the problem is usually the client's decision-making rather than your sourcing, and that is a conversation to have with them directly and with evidence. If losses cluster at offer, qualification on salary expectations is happening too late in your process. Watch it per consultant too: strong submission ratios with weak interview conversion usually means someone is selling the role harder than they are qualifying the candidate.

Fill rate is the share of roles you were given that you actually filled, and its honesty depends entirely on the denominator. Count every role you formally accepted, including ones the client cancelled, put on hold or filled internally, and report those outcomes separately rather than deleting them. Agencies that quietly remove dead roles produce a number that looks excellent and predicts nothing. Two versions are worth keeping. Overall fill rate covers everything accepted. Qualified fill rate covers only roles that reached a proper intake and stayed live, which is a fair measure of delivery. The gap between the two is itself the insight, because a wide gap means you are accepting work you should decline. Fill rate also belongs at client level, since one chaotic account can drag a whole desk average down while the rest of the book performs well. Use it to decide which clients get priority, and be willing to hand back roles that consistently die. Filling fewer roles at a higher rate usually earns more.

• Count every role you formally accepted, including the ones that died
• Report cancelled, on-hold and filled-internally as separate outcomes
• Keep overall and qualified fill rate side by side
• Read the gap between them as a verdict on what you accept
• Break it down by client before blaming a consultant

Time to submit measures from role accepted to first shortlist sent. Time to fill measures from role accepted to offer accepted. The first is almost entirely inside your control, the second is shared with the client, and confusing them causes arguments that go nowhere. Manage the metric you own. If time to submit is slow, the cause is usually a thin talent pool in that niche, a consultant juggling too many roles, or an intake that never really happened. If time to submit is fast but time to fill is long, the delay lives in the client's process, and you now have evidence rather than a complaint. Bring elapsed days per stage to the account review and ask which stage they want to shorten. For context, the SHRM 2025 Benchmarking Report puts average time to fill at around 45 days across employers, though it varies substantially by role and sector. A time to hire calculator is a quick way to establish your own baseline before arguing about anyone else's.

• Time to submit: role accepted through to first shortlist sent
• Time to fill: role accepted through to offer accepted
• Record the client's own open date separately if they want it reported
• Bring elapsed days per stage to the account review, not a complaint

Gross margin, not revenue, is what pays the business. For a permanent placement the arithmetic is simple: the fee invoiced minus direct costs attributable to winning it, such as advertising or assessment fees. For contract and temporary work it is the bill rate minus the pay rate minus employment on-costs, and the on-costs are where inexperienced agencies lose money, because taxes, insurance, holiday accrual and any statutory contributions belong in the calculation. Express contract margin in currency per hour as well as a percentage, since a low percentage on a high bill rate can out-earn the reverse. Roll the same number up per consultant to measure productivity honestly. Placements per consultant rewards whoever works the easiest roles. Gross margin per consultant, set against the fully loaded cost of employing them, tells you whether a desk pays for itself. Review it quarterly rather than monthly so one lumpy month does not trigger a bad decision, and keep an eye on the cost per hire picture your clients are comparing you against.

Three numbers predict whether the business survives a bad quarter. Client concentration is the share of gross margin coming from your largest client, and from your top three. A single account carrying a large share of margin turns a routine procurement review into an existential event, and it weakens you in every negotiation with that client. Redeployment rate applies to temp and contract desks: the share of workers who move straight from one assignment into another with you rather than leaving the bench. Redeployment costs almost nothing to source and is the closest thing a staffing business has to recurring revenue, so a falling rate is an early warning about consultant follow-up. Consultant dependency is the third, the share of margin produced by one biller. Measure all three quarterly, write down a ceiling you are not willing to cross, and act while there is still time. The fix is always slower than the discovery, because diversifying a client book takes quarters rather than weeks.

• Share of gross margin from your largest client, and from your top three
• Redeployment rate on every contract and temporary desk
• Share of margin produced by a single biller
• A written ceiling for each, agreed long before you approach it

Most agencies do not need analysts. They need one system of record and the discipline to update it. If submissions, interviews, offers and outcomes are logged in an applicant tracking system for staffing agencies as they happen, every ratio in this guide becomes a report rather than a project. The failure mode is not missing software. It is partial adoption: two consultants log everything, one keeps a private spreadsheet, and the aggregate becomes fiction. Start with four numbers, not forty. Submissions per live role, submission-to-interview, interview-to-placement, and gross margin per consultant will surface almost every operational problem a small agency has. Review them at the same meeting every week so the data has a job to do. Add mandatory reason codes on rejections, because the why is worth more than the count. Once the basics are stable, recruitment analytics software can add trend and forecast views, but reporting cannot repair data nobody entered.