Bullshit Hiring KPIs: When Recruiting Metrics Stop Meaning Anything

Recruiting metrics are useful when they help us understand a process, identify problems or make better decisions. Tracking how long candidates wait between stages, for example, can give a team a concrete delay to investigate.

The problem starts when a metric becomes a performance claim without enough information to interpret it. A percentage might be correct while the story attached to it is much less certain.

That sounds like a major improvement. But a reduction is only meaningful when we understand the starting point and whether the comparison is fair.

• Compared with which period?
• Were the same types and seniority of roles compared?
• Did hiring volume change?
• Did the market change?
• Did quality or candidate experience suffer?

A quarter of familiar, frequently hired roles is different from a quarter of specialist leadership searches. A faster average could reflect that change in the mix, even if the process itself stayed the same.

Agree on the clock, too: when does it start, when does it stop, and are pauses counted? Changing the definition can change the number without changing a candidate's experience.

Faster hiring can be valuable. But speed alone does not tell us whether hiring became better.

If a company made three offers and all three candidates accepted, the offer acceptance rate is indeed 100%.

That is mathematically correct. It is not necessarily evidence of an exceptional recruiting process.

With three offers, a single different outcome would change the percentage dramatically. Show the count alongside the rate and name the period: “3 of 3 offers accepted this quarter” is easier to interpret than a standalone 100%.

Offer acceptance can also hide candidates who withdrew earlier in the process. If people leave before an offer is made, their experience never enters that denominator. Look at withdrawals across the process as well as the candidates who reached the offer stage.

A satisfied hiring manager can tell you something useful about communication, alignment or support. But before treating this number as proof of success, ask:

• Who was surveyed?
• How many people responded?
• What questions were asked?
• Was the survey anonymous?
• Was this measured once or consistently?
• Does satisfaction correlate with hiring outcomes?

Does 97% mean the share of respondents who selected “satisfied”, or an average score converted into a percentage? Those are different measurements. Missing responses and unclear questions can make either one difficult to interpret.

Stakeholder satisfaction can be useful. But precision like “97%” can create an illusion of scientific accuracy when the underlying measurement is weak. A happy stakeholder is a perspective on the process; it does not, on its own, establish the quality of a hiring decision.

The more useful question: did qualified applications increase?

Application volume alone can be a vanity metric. A broader campaign or an easier application form might bring in more people without bringing in more candidates who meet the role's requirements.

More applicants can even create more screening work and longer waits without improving hiring outcomes. Look at the number and proportion meeting agreed criteria, where they came from and how they progress.

Define “qualified” before celebrating the result. A hiring scorecard with observable criteria helps a team agree what relevant evidence looks like.

Check these five things before making a metric the headline of your next reporting meeting.

Definition

Everyone understands exactly what is being measured.

Context

Role type, geography, seniority, hiring volume and market conditions matter.

Comparison

A baseline, historical period, benchmark or meaningful target exists.

Sample size

We know whether the metric represents 3 cases or 300.

Decision value

The metric helps someone decide what to change, investigate or improve.

If a metric cannot help someone decide what to investigate or change, ask why it is on the dashboard.

Recruiting analytics should help teams investigate reality rather than decorate dashboards. A number is a starting point for a conversation about how hiring works and what needs attention.

A good dashboard does not simply tell us: “Time-to-hire is 32 days.”

It helps us ask:

• Where are candidates waiting?
• Which stages create delays?
• Which roles behave differently?
• Where are candidates dropping out?
• Did a process change actually improve the outcome?

Use recruiting data
to make better hiring decisions.