Adequacy looks like success

There’s a thought experiment nobody in HR seems interested in running. Take every candidate you rejected in the last five years. Track them. Watch where they landed, what they built, what they became. Now compare them to everyone you hired.

You won’t do this. Nobody does. That’s the problem.

The system that cannot fail

Recruiting has a structural advantage most professions don’t: it is, by design, nearly impossible to grade your own work.

When a surgeon loses a patient, there’s a record. When a deployment breaks production, there’s a post-mortem. When a recruiter passes on a candidate who goes on to become a VP at a competitor, or a CTO, or the exact hire that pulls another company ahead, nobody writes that down. There is no post-mortem for the road not taken.

This isn’t unique to recruiting. It’s a well-documented cognitive phenomenon: survivorship bias. The tendency to form conclusions based only on the data that made it through a filter, while ignoring everything the filter removed. You see the ships that came home. You count the bullet holes on the fuselage. You never get to study the planes that didn’t make it back.

In hiring, the filter is the hiring process itself. The only data points that survive it are the people you said yes to.

What gets measured

Here’s what a recruiter’s performance is typically evaluated on. How quickly a role gets filled. How many candidates move through the pipeline. Offer acceptance rate. First-year retention. Hiring manager satisfaction.

These are real metrics. They’re not useless. But notice what none of them measure: the quality of the candidates who were rejected.

Quality of hire, the composite score built from performance reviews, ramp-up time, and retention data, tells you how your yes decisions performed. It says nothing about whether your no decisions were correct. A recruiter can score perfectly on every measurable KPI while consistently passing on the most talented people who applied.

The industry has quietly acknowledged this in its own way. In data science terms, a hiring decision is a hypothesis test. False positives, the bad hires, are visible. They underperform. They leave. They get managed out. The data exists.

False negatives leave the building and go succeed somewhere you’re not watching.

Adequacy as the bar

There’s a narrower version of this problem worth naming.

Imagine a hire who is perfectly adequate. Shows up, doesn’t cause problems, meets expectations, never surprises anyone in either direction. From the recruiter’s perspective, this is a success. Quality of hire scores reflect it. The hiring manager is satisfied. The role is filled.

But what if the four candidates screened out at the phone interview stage would have been exceptional? What if one of them would have done the job twice as well, flagged risks nobody else saw, or grown into a role nobody knew was going to be needed? You’ll never know. There is no mechanism in standard recruitment tracking to capture this.

The hire was adequate, so the decision was correct. That’s the logical trap. Adequacy doesn’t prove correctness. It proves sufficiency. Sufficiency is not the same thing as optimal.

The scoreboard has half the data

Recruiter scorecards, the formal frameworks used to evaluate individual recruiter performance, typically include time-to-fill, funnel conversion rates, offer acceptance rates, and workload difficulty. The more thoughtful ones add candidate experience scores, measuring whether people who went through the process felt it was fair and professional.

What’s absent from essentially all of them is any measurement of the counterfactual. Nobody is tracking whether the rejected candidates from six months ago went on to outperform the person who was hired. Nobody is systematically following up with the people they said no to.

The HR industry has its own language for this. It recognizes that when you only study your successful hires, you’re missing half the story. That focusing on people who made it through the process produces an incomplete picture of what actually predicts success. The stated awareness is there. The tooling and accountability structures don’t reflect it.

A fair defense, before you dismiss it

This is not an argument that recruiters are careless, or that the industry is full of bad actors. Most people in talent acquisition are doing a genuinely difficult job with incomplete information, under time pressure, managing volume that makes deep evaluation nearly impossible.

When 180 applicants apply for a role and a recruiter reviews the top 20, the problem isn’t malice. It’s scale. The process has to filter somehow. The structural question is whether the filter ever gets examined. Whether anyone ever asks if the calibration is right, and how they would know if it wasn’t.

False negatives in hiring cause companies to spend more time interviewing than they need to, and to miss valuable talent. They’re documented. They’re named. But there’s no organizational feedback loop that brings them back to the surface. A false positive costs money, causes friction, leaves a paper trail. A false negative just quietly goes to work for someone else.

The loop that doesn’t close

There is a version of a recruiter scorecard that could account for this. It would require tracking rejected candidates over time, not invasively, but as a data discipline. It would require asking, with some regularity, where the people you passed on ended up. It would mean building a feedback loop that doesn’t stop at offer accepted.

Some hiring teams are starting to advocate for exactly this. Analyzing the full candidate pool rather than just the hires. Reviewing why past hires failed as well as why they succeeded. Being honest about the parts of the process that produce noise instead of signal.

That’s harder than tracking time-to-fill. It requires the kind of intellectual honesty most performance measurement systems don’t reward. It requires accepting that a high score on every visible metric might still represent a pattern of systematically missing the best candidates.

The question isn’t whether recruiters are good at their jobs. Most are. The question is: how would they know if they weren’t?

Adequacy, it turns out, is a very comfortable place to misread as success.