What One-Way Video Interviews Actually Select For

Every filter selects for something. The trouble is that it isn't always the thing you built it to select for.

A one-way video interview is meant to find people who'd be good at the job. The pitch is simple: send candidates a set of prompts, let them record whenever it suits them, and watch the recordings whenever it suits you. Nobody schedules anything, and no recruiter time gets spent until there's a shortlist. After 2020 it became the default first screen for a lot of high-volume hiring, which makes sense. It promised to solve the problem every hiring team has, which is how to screen more people without more recruiters.

It also loses a lot of people. Enterprise talent teams using these tools have reported completion rates between 40 and 60 percent, varying by role and sector. Suppose 200 people apply for a role and 100 of them are worth screening. At those rates, somewhere between 40 and 60 of the 100 never record anything. They just move on.

The usual explanation is that those people weren't very interested. They applied to too many jobs, or couldn't be bothered. That explanation is convenient, because it means nothing is wrong with the tool. I don't think it holds up.

Think about what a one-way video asks of someone. They sit alone in front of a camera and answer questions, knowing they'll be watched and judged later by people they've never spoken to, who haven't shown any sign they're still interested. There's no conversation and no feedback. They're performing for an audience that may never reply. A strong candidate with other options can look at that and reasonably decide it isn't worth it. A candidate with fewer options is more likely to do it anyway. So the format tends to keep the people with the fewest alternatives, which is close to the opposite of what you wanted.

It isn't equally hard for everyone, either. Someone who's employed can't easily find a quiet room to record in during business hours. Someone answering in a second language has more to lose in a stiff, one-take recording than in a conversation where they can ask what a question means. Some people are just less comfortable on camera, which has little to do with how well they'd do the job. I'd expect all of these groups to drop out more often, and none of them is less qualified. What the screen ends up measuring is closer to ease on camera than competence.

The reason this goes unnoticed is that the missing people don't show up anywhere. If 200 applied and 90 submitted videos, the system shows you 90 candidates. The 110 who didn't aren't in any report, and nobody asks whether the 40 or 50 qualified people among them would have been excellent hires if you'd reached them some other way.

It's the same mistake as crude keyword screening in an ATS: the filter runs perfectly and filters for the wrong thing. Keyword screens reward keyword density, one-way video rewards willingness to perform on camera with no one on the other end, and neither tracks job performance very well. Both hand the recruiter a manageable stack and the feeling that the pipeline has been screened.

That feeling is the part I'd worry about. The recruiter watching those 90 videos trusts the process because it produced a reasonable number of candidates. But the list is short partly because about half the applicants decided this employer wasn't worth the trouble. It looks like a screened shortlist when it's really a self-selected one.

You can't fix this by putting a recruiter on every first screen, since that's the capacity problem one-way video was supposed to solve. What you need is a first step that's easy on the candidate and still works across hundreds of applications.

That's the gap we built Asendia for, and it's easiest to describe from the candidate's end. A few hours after applying, while they still remember the job and are half expecting to hear back, their phone rings. The voice on the line asks the same structured qualification questions a recruiter would, listens, and follows up on what they actually said. If they're only free in the evening or on a weekend, that's when the conversation happens. There's no camera to set up and no quiet room to find, and a phone call feels like a first step rather than an audition, so I'd expect it to lose far fewer people than a video prompt. The candidate who skipped your video because they were between meetings will often take a call that evening without thinking twice. What reaches the recruiter is a ranked shortlist with qualification summaries, key quotes and notes in the ATS, drawn from the whole applicant pool instead of the half that was willing to perform on camera.

This matters more now that so many applications are written with AI, which has weakened the written parts of screening as well. The post on AI-generated applications overwhelming text-based ATS screening explains why the first real contact with a candidate has become the most important part of the funnel.

One-way video isn't going away, and for some roles it's a reasonable tool. But I'd stop treating it as a neutral screen. When roughly half of applicants drop out, that tells you less about their interest than about what the format selects for. If you use it, try comparing the people who finished your video step with the people who didn't, using whatever you know about both groups from their applications. If the ones who dropped out look just as strong on paper, you've found out what your filter was really filtering.

Ready to transform your hiring strategy? Schedule a Demo with our founders today!