First round interview screening done for every applicant

The first round is the first real conversation after an application, usually run by a recruiter rather than the hiring manager, and its job is narrow: confirm the candidate meets the non-negotiable requirements, find out whether they actually want this role, and take a first structured read on one or two competencies that matter. It is not the round where you decide to hire someone. It is the round where you decide who is worth a hiring manager hour.

Teams get into trouble when the round quietly expands. A 30 minute screen that tries to assess technical depth, team fit and long-term potential produces a shallow read on all three and takes 50 minutes. Keeping round one narrow is what makes it fast enough to run on everyone, and running it on everyone is what stops good candidates being cut on a resume keyword.

Twenty to thirty minutes. Under 15 and you cannot get past stated facts into any behavioral evidence, so you have run a questionnaire with extra steps. Past 35 and completion drops, particularly among employed candidates who are fitting the call into a lunch break, which biases your pool toward people with flexible schedules.

A workable shape is two minutes of role context, four to six questions with follow-ups, five minutes for the candidate to ask questions, and a clear statement of the next step and its date. The last item is free and is the strongest single lever on whether your best candidates are still available two weeks later.

There is no reliable universal figure, and any number quoted without a denominator should be ignored. The rate depends entirely on how much filtering happened upstream: a role with a tightly written posting and a narrow sourcing channel might advance one in three, while an open posting on a large job board can run below one in twenty. Publishing a benchmark for this would be inventing one, so we do not.

The useful metric is not the pass rate but the agreement rate: of the candidates you advanced, how many did the hiring manager consider worth interviewing? When that number is low, round one is measuring the wrong thing, and the fix is in the rubric rather than in the pass threshold. Track it per req for a quarter and the pattern is usually obvious.

The arithmetic is what breaks first. Five hundred applicants at 25 minutes each, plus scheduling and no-shows, is well over 250 recruiter hours for a single req. No team has that, so in practice the round gets rationed: a resume screen picks who gets a conversation, and the resume screen is the least reliable stage in the whole process.

Removing the per-candidate hour changes what is possible. Interviews run in parallel and on the candidate's own schedule, every applicant answers the same questions, the agent follows up where an answer is thin, and each response is scored against anchored criteria before a recruiter sees anything. Recruiters open a ranked shortlist with transcripts instead of a queue, and they still make every advance decision. The volume version of this is covered on our high volume hiring software page, and the campus version, where the whole pipeline arrives in three weeks, on campus recruiting software.

Four to six, the same four to six for everyone on the req, weighted toward behavior rather than opinion. Ask for a specific past instance and then probe for what the candidate personally did, which is where the difference between two similar resumes actually appears. Ask the must-have logistics questions of every candidate identically, because asking them selectively is where discrimination claims start.

Write three anchored levels per criterion before the first application arrives, so scoring describes the content of an answer rather than how confident the person sounded. Our pre-screening interview questions guide has a general set to adapt, interview scorecard template covers the scoring structure, and for student and intern pipelines the questions differ enough to be worth their own set in our internship interview questions guide.

It depends on what you are comparing it to. The honest baseline is not a careful structured interview with every applicant, because almost nobody does that at volume. It is a resume screen that cuts 80 percent of the pool on keywords and school names, followed by improvised calls whose consistency depends on who was free that afternoon. Measured against that, a round where everyone gets the same job-related questions and a scored transcript is a clear improvement.

Two conditions have to hold for that to be true. The questions must be job related rather than proxies for background, and nothing should be inferring competence from face or voice quality. Beyond that, the duties are the employer's: bias audit and notice where Local Law 144 applies, notice and consent in Illinois, and a human making every advance decision. Our AI hiring compliance page sets out the rules state by state.