This is part of GoPerfect Labs — where we publish findings from our data team. We analyze screening decisions across our platform, then share what we find so recruiting teams can build more consistent, reliable hiring processes.
We ran a controlled study across 34 companies using GoPerfect's inbound screening module. For each company, we identified roles where multiple recruiters independently evaluated the same pool of applicants — same role, same criteria, same candidate set. Then we measured agreement.
The question was simple: when two recruiters on the same team screen the same candidates, how often do they agree on who should move forward?
.png)
The consistency gap nobody measures
Recruiting teams track time-to-hire, cost-per-hire, and candidate satisfaction. Almost none of them track screening consistency — whether different recruiters on the same team, evaluating the same candidates, would make the same decisions.
We found they wouldn't. When we compared 12,400 parallel candidate evaluations, the average shortlist overlap between two recruiters was 51%. Meaning: for every 10 candidates one recruiter approved, the other recruiter would only agree on about 5 of them.
.png)
Where inconsistency is highest — and lowest
The 51% average masks a wide range. Some role types produce much tighter recruiter alignment. Others are almost random.
.png)
The pattern is clear: the more subjective the evaluation criteria, the wider the disagreement. Roles with hard, measurable requirements (certifications, specific programming languages) produce the tightest alignment. Roles where "fit" and "potential" dominate the criteria are essentially a coin flip between recruiters.
The cost of inconsistency isn't just unfairness
Screening inconsistency is usually framed as a fairness issue — and it is. But the operational cost is larger than most teams realize.
Cost 1 — Missed hires
When recruiter A advances a candidate and recruiter B wouldn't have, the reverse is also true: recruiter B would advance candidates that recruiter A skipped. In our data, an estimated 23% of ultimately hired candidates would have been screened out if a different recruiter had done the initial review. Your hire depends partly on which recruiter opens the file.
Cost 2 — Wasted interview slots
Inconsistent screening doesn't just miss good candidates — it lets mismatched candidates through. In our dataset, roles with the lowest recruiter agreement (below 40%) had 34% higher interview-to-offer dropout rates. The pipeline looked full, but the quality was unreliable.
Cost 3 — Untrustable pipeline data
If your screening varies by who's doing it, your pipeline metrics don't mean what you think they mean. A "strong pipeline" might reflect loose screening, not good sourcing. A "weak pipeline" might mean a strict screener happened to catch that role. You can't optimize a process where the measurement tool changes every time.
Inconsistent screening doesn't just slow down hiring. It makes the data you use to manage hiring unreliable.
What happens when AI provides the baseline
GoPerfect's AI screening doesn't replace the recruiter's decision. It provides a consistent starting point — a 1–5 score with explainable reasoning — that the recruiter can agree with, challenge, or override.
When we compared screening outcomes at companies using GoPerfect's AI as the initial screen vs. those relying on human-only review, the consistency gap shrank dramatically.
.png)
With AI providing the baseline, recruiter agreement jumped from 51% to 82%. Not because the AI forced a decision — but because it gave both recruiters the same structured evaluation to react to. Instead of starting from a blank slate and their own biases, they started from a consistent data point.
.png)
What this means for your screening process
.png)
