Staffing KPIs: The Four Numbers a Firm Owner Should Check Every Day

Most staffing dashboards report what already happened. Fill rate, time-to-fill and placements are all outcomes, and an outcome cannot be influenced by the time it appears on a screen. Four staffing KPIs move earlier: live submissions, interviews scheduled, offers in play, and receivables past 60 days. Together they forecast next month before next month arrives.

The four staffing KPIs, defined

1. Live submissions this week

Candidates submitted to clients, counted weekly. It is the only one of the four where more is unambiguously better, and it constrains everything after it. A pipeline problem always shows here first.

2. Interviews scheduled

Read as a ratio against submissions, never as a raw count. Twenty submissions producing two interviews is a very different week from twenty producing eleven, and the raw number hides the difference.

3. Offers in play

Candidates at offer stage across all desks. Multiply by your historical offer-to-join rate and you have a revenue forecast that is roughly four to six weeks ahead of your P&L.

4. Receivables past 60 days

The most ignored number in staffing and the one that kills otherwise healthy firms. Placements made, invoices raised, money not arrived.

Why fill rate and time-to-fill are the wrong daily numbers

Both are useful. Neither is actionable daily, and the distinction matters.

Time-to-fill is measured from a start point you often do not control to an end point the client entirely controls. A client who takes three weeks to schedule an interview doubles your time-to-fill while your team does everything correctly. Reviewing it every morning produces anxiety, not decisions.

Fill rate is a quality signal over a quarter. Checked weekly it moves on noise, because the denominator is small and one unusual role distorts it.

Use both in monthly reviews and in client conversations. Do not use them to run the week.

Why leading indicators get ignored

Every staffing owner knows in principle that leading indicators matter. Almost none run their week on them. Three reasons, and they are worth naming because each has a fix.

The dashboard was built by a finance mindset

Most reporting in staffing is built to answer "what did we do last month", because that is what the accounts need. Those reports then get repurposed as management tools, which they were never designed to be. Revenue reporting and operating reporting are different jobs and should be different screens.

Leading indicators are uncomfortable

Fill rate is retrospective and therefore safe. Live submissions is a number about this week that somebody is accountable for right now. Teams unconsciously prefer metrics that cannot be acted on, because metrics that can be acted on create obligations.

The data is not trustworthy enough to act on

This is the real one. If submissions are logged manually and inconsistently, nobody believes the number, so nobody uses it. The metric problem is downstream of a data capture problem, and fixing the dashboard without fixing the capture just produces a prettier number nobody trusts.

How to read each staffing KPI when it moves

The pattern that should worry you most is submissions holding steady while the interview ratio falls. That means the team is busy and the work is not landing, which is the most expensive failure mode in staffing because it consumes capacity and produces nothing. It is almost always a definition problem rather than an effort problem, which is why the scorecard sits upstream of it.

The ten-minute daily routine

Same time every day. Four numbers, one screen.

1. Check direction, not value. A single day's number is noise. Three days in the same direction is a signal.
2. When something moves, ask one question of one person. Not a review meeting. Daily checks are for spotting, weekly reviews are for solving.
3. Write nothing down. If the routine requires a spreadsheet, it will stop within a fortnight. The value is in developing a feel for normal.

That last point is the real argument for doing it daily rather than weekly. You are not looking for a number. You are calibrating what normal looks like on your desks, because that is the only way you will notice when it stops being normal.

Setting the thresholds for your own firm

The four numbers are universal. The thresholds are not, and a borrowed threshold is worse than none.

Spend three weeks recording the four numbers without acting on them. At the end, you have a baseline. Set your alert levels from that baseline, not from anything you read.

The last row deserves a note. Expressing overdue receivables as a percentage of billings makes a single large stuck invoice look small. An absolute figure does not, and a single stuck invoice is usually the whole problem.

What to do when a number moves

Submissions down three days running. Look at recruiter time before recruiter effort. Something has usually been added to the week: a new process, a client escalation, a system that stopped working. The admin load is the most common silent culprit.

Interview ratio down. Pull the last ten rejections and read the client's reasons. Clustered reasons mean the brief was wrong. Scattered reasons mean the screening was.

Offers in play down. This is a four-to-six-week warning. Act now, because by the time it appears in revenue it is unfixable for that quarter.

Receivables ageing. One call from the person who holds the relationship resolves most of it. The failure is almost always visibility rather than willingness.

The wider KPI set, and what to check when

Four numbers run the day. They do not run the business. Here is where the rest belong.

The common failure is checking monthly metrics weekly. Small denominators make them swing, the swing gets treated as signal, and the team gets managed on noise.

The opposite failure is worse. Client concentration checked once a year is how a firm discovers that 60 percent of revenue sits with one account in the same week that account leaves.

Four mistakes people make reading these numbers

Reacting to one day

A single reading is noise. Nothing in staffing moves cleanly day to day, and acting on one data point trains your team to manage the metric rather than the desk.

Counting submissions that were never going to convert

If a recruiter can raise the submissions number by lowering the bar, the number stops being a leading indicator and becomes a target to game. Pair it with the interview ratio, always, and never review one without the other.

Using industry conversion rates in a forecast

Your offer-to-join rate is a property of your clients, your market and your screening. A published average from a different market will produce a confident forecast that is wrong.

Hiding receivables from the people who can fix them

Finance owns the invoice. The recruiter owns the relationship. Keeping the ageing visible only to finance guarantees that the one person who could resolve it in a phone call never knows it exists.

Where four numbers is not enough

This is a daily operating routine, not a management system, and it has real limits.

It tells you nothing about margin. A desk can hit every one of these four numbers while placing at a fee percentage that loses money, and nothing in the daily check will surface it. Margin belongs in a monthly review against desk economics.

It tells you nothing about individual performance. Four aggregate numbers hide a top biller carrying two people who are not producing. That is what a weekly desk review is for.

And it is a lagging indicator of your own capacity. If submissions are flat because every recruiter is at their ceiling, no daily check will show you that. That is a structural question about how many roles one person can carry, which for our customers has moved by roughly 2x once sourcing and screening stopped being manual.

Two numbers that look leading and are not

Both get proposed constantly. Both fail as daily metrics.

Calls made

It is genuinely leading, and it is trivially gameable. The moment a team is measured on calls, call volume rises and call quality falls, and you have bought activity theatre. If you track it at all, track it as a diagnostic when submissions fall, never as a target.

CVs sourced

Sourcing volume feels like progress and correlates weakly with placements, because most sourced profiles were never viable. A firm sourcing 250 profiles a day and submitting four has a screening problem that the sourcing number actively conceals.

The test for any candidate metric: could a recruiter improve this number tomorrow without improving anything real? If yes, it belongs in a diagnostic, not on the daily screen.

Frequently asked questions

How to start tracking these four

Put the four numbers on one screen and check them at the same time every day for three weeks. Do not act on anything in week one.

The value is not in any single reading. It is in knowing, by week three, what a normal Tuesday looks like on your desks. Everything useful follows from that, and nothing useful precedes it.