Your recruitment dashboard's biggest number is its least useful one

Recruitment analytics sells a clean promise: track your metrics, watch the dashboard, and the numbers will tell you how hiring is going, with time-to-hire sitting at 38 days, cost-per-hire at £4,500, a single figure for each and updated live.

The trouble is that the headline number is the least useful thing on the page. A blended average of every role, level and team throws together situations that have nothing to do with each other, so it either flatters you or panics you, usually for the wrong reason.

The metrics worth tracking, and what good looks like

Most guides tell you which metrics to collect, usually time-to-hire, cost-per-hire, source-of-hire and offer-acceptance, but almost none tell you what a good number actually is. That's the part they skip, and it's why a single figure for any of them can mislead you.

Time-to-hire. Across roles filled on cord the median is 30 days, but it splits from 26 days at a startup to 39 at a scaleup, and from 28 for a mid-level role to 42 for a lead. A blended 30 tells you nothing until you know which role it's for.

Cost-per-hire. The average on cord is £940, though a graduate role and a principal role sit a long way either side of it, so on its own it's a starting point rather than an answer.

Application quality. One in five applications on cord leads to an interview, and even that moves by role.

Each of these is only useful when you read it against a comparable segment. On its own, it's just a number you write down.

The published benchmarks have the same flaw

Industry benchmarks don't help much either. SHRM puts the average time to hire at 44 days, Workable reports a global engineering average of 62, and broader tech figures land somewhere near 52, so whichever you pick you're only comparing your blend to theirs: two averages hiding the same variation, passing in the night.

And those figures describe how slow hiring usually is, not how fast it can be. The 30-day median on cord is roughly half the tech average, but even that isn't a target until you know which role it's for.

The bottom line

The biggest number on your recruitment dashboard is the one to trust least, because it averages away the very differences that decide whether you're fast or slow.

A 30-day time-to-hire is good. Whether yours is good depends on what you're hiring, and no blended average will ever tell you.

A blended benchmark can't tell you whether you're fast, but your own segments can, so it's worth seeing how your roles compare on cord.