How to know if your candidate screen is working

Knowing whether your candidate screen works means checking whether it lets the right people through and discards those who don’t fit, without cutting too much or too little. A screen that’s too permissive doesn’t filter: it shifts all the work onto the later stages. One that’s too strict leaves out good candidates for reasons that may not matter for the role. The good screen isn’t the toughest or the softest: it’s the one that measures what the position really needs and gets it wrong less often.

The candidate screen is the part of the process that’s least questioned and that decides the most. Every application you discard or let through shapes who reaches the end, but we rarely stop to ask whether that screen is doing its job well. And a screen that fails, fails in silence: you never see the ones it discarded, so you don’t notice the good ones that slipped away.

The two ways to fail

A screen can go wrong in two opposite directions, and both cost:

Failure	What happens	The cost
Lets too many through	Approves almost everyone	The interview fills with candidates who don’t fit; time is lost
Cuts too much	Discards too many	Good candidates are left out by criteria unrelated to the role

The treacherous part is that the second failure is invisible. When you let too many through, you notice it: the interviews become a waste of time. When you cut too much, you don’t see the ones you discarded, so you never know how many good ones you lost.

How to assess whether it’s working

There’s no single number, but there are signals that, together, give an honest picture:

1. Look at what happens afterward to the ones you approved. The most honest signal arrives late: do the people your screen let through do well once they’re in? If many turn out not to fit, the screen is letting too many through. This connects directly with the quality of the hire.
2. Check the advance rate. Of the candidates who pass the screen, what percentage advance in the interview? If almost none advance, the screen is approving people the later stages discard right away: it isn’t separating well.
3. Ask the people who interview. Do they feel relevant candidates are reaching them, or that they have to discard most of them? It’s a qualitative signal, but a fast and revealing one.
4. Watch the volume balance. A screen that approves almost everyone or almost no one is probably miscalibrated. The healthy point depends on the role and the sector; there’s no universally correct percentage.

The root: why the screen filters

More important than how much it lets through is why it does so. A screen that cuts by the prestige of the CV, by a first impression, or by requirements the role doesn’t really need —exact years of experience, a specific degree— gets it wrong by design, even if its approval percentage looks reasonable.

A good screen rests on a comparable signal, connected to the competencies the role actually requires, and applies it equally to everyone. This way, separating those who fit from those who don’t is based on evidence and not on intuition. That’s the difference between a screen that discards the right people and one that just discards fast.

Read it alongside the other metrics

The quality of the screen isn’t measured on its own. A screen that cuts a lot can lower your time-to-hire while also leaving out good candidates: fast isn’t the same as right. That’s why the health of the screen is read by cross-referencing volume, advance rate, and, above all, how the people it let through end up performing. Metrics describe what’s happening; your team decides whether the screen is calibrated for the role.

In short

Knowing whether your candidate screen works means checking whether it lets the right people through and discards those who don’t fit, without cutting too much or too little. It can fail in two directions: letting too many through (the interview fills with people who don’t fit) or cutting too much (good candidates are left out, and that failure is invisible). To assess it, look at how the people you approved do, the advance rate in interviews, the perception of those who interview, and the volume balance. But what’s decisive is why it filters: a screen that cuts by prestige or by requirements unrelated to the role gets it wrong by design. A good screen rests on a comparable signal connected to the competencies of the position, and is always read alongside the quality of the hire and time-to-hire.