Async Video Interviews Are Completing at 62%. You're Not Screening Candidates — You're Screening for Compliance.

Async video interviews — where candidates record answers to preset questions and submit the footage for later human review — are now completing at around 62% in competitive hiring markets, down from over 80% at peak pandemic adoption in 2021 [1]. Most hiring teams look at that number and wonder what's wrong with their candidate pool. The better question is what's wrong with the format.

What the Completion Rate Is Actually Measuring

The candidates who abandon async video screens are not the disengaged ones. They're the candidates with enough inbound interest that they can afford to skip formats they find demeaning. Top performers in software engineering, enterprise sales, operations leadership — the people most hiring teams are trying to reach — have calibrated that recording themselves for a faceless review queue is a poor use of their time. Candidates who complete at the highest rates tend to be those with fewer parallel options, more willingness to accept process friction, and — this matters — more time to spare.

That's not a quality signal. That's selection pressure you didn't design.

Research on candidate decision-making shows that high performers are significantly more likely to abandon hiring processes they perceive as one-sided — where they're being evaluated without any evaluative return [2]. Async video is the most extreme version of this: the candidate performs, the employer watches. There's no conversation, no follow-up, no moment where the candidate learns something about the role or the team. The transaction is purely extractive from the candidate's perspective. The ones who reject that arrangement first are often the ones you most want to hire.

The Signal Problem Nobody Fixed

Beyond the completion rate, async video has a structural signal quality problem that is genuinely underreported. When a candidate records an answer to "describe a time you managed competing priorities," they do so under ideal conditions: quiet environment, written notes, multiple takes, days of preparation. That's the opposite of the context in which they will perform the actual job.

A live recruiter conversation — even a routine one — catches the candidate in real time. The follow-up question is unscripted. The candidate has to adjust mid-sentence, maintain a thread under mild pressure, and demonstrate whether their thinking is actually coherent or just well-rehearsed. That's signal relevant to nearly every professional role. A polished 90-second monologue tells you almost nothing about that. It tells you whether the candidate knows how to film a 90-second monologue.

The companies that adopted async video at scale were solving a legitimate problem — recruiter bandwidth doesn't scale linearly with application volume — with a format that generated the appearance of structured process without the substance of real evaluation. The pipeline looked like it was moving. The shortlists kept disappointing.

The format also introduced a second, quieter selection effect: candidates who complete async video screens disproportionately skew toward roles where the candidate has fewer external offers and more anxiety about losing this one. For competitive roles in tight talent markets, that skew runs exactly opposite to what you're trying to optimize for.

How Asendia AI Solves the Volume Problem Without the Format Problem

The original case for async video was sound: you cannot have a recruiter on a live call with 400 applicants in 72 hours. The format was the wrong solution to a real constraint. Asendia AI addresses that constraint differently — with live, adaptive voice conversations that run 24 hours a day, seven days a week.

When a candidate applies at 10pm, Asendia calls them that evening and conducts a structured qualification conversation. Not a recorded monologue prompt. A real-time exchange where the AI follows what the candidate says, asks relevant follow-ups based on their actual responses, and captures signal that resembles what a skilled recruiter would surface in a strong screening call — how they explain their reasoning, how they handle a question they weren't fully prepared for, what the actual texture of their thinking is.

That conversation lands in your existing ATS: qualification summary, ranked shortlist, and verbatim excerpts, before your recruiters arrive in the morning. No separate platform. No review queue of 200 video files. Recruiting agencies use Asendia to absorb volume spikes without adding headcount — three recruiters running campaigns that previously needed six, because the AI handles every first conversation at any hour. The team inherits a vetted shortlist of candidates who've already been genuinely engaged, not a compliance queue of whoever sat through the format long enough to record their answers. The practical test: if your async video completion rate is sitting at 62%, you're making shortlist decisions with data from 62% of your applicant pool. The 38% who opted out aren't randomly distributed — that's a structural gap in your candidate visibility that your ATS isn't surfacing. For more on where these pipeline gaps compound into KPI blind spots, this breakdown of what to actually measure in a post-AI hiring process is worth the read.

Final Word

Async video interviews solved the optics of "we reviewed everyone's submission" without solving the underlying problem of hiring at volume with real signal. The completion rate collapse is the market's verdict on that bargain. Candidates with options opted out first. That's not a candidate engagement problem — it's a format design problem that's been compounding into every shortlist your team has celebrated for the past three years. The companies still relying on async video as their primary early-screen layer are not getting a representative sample of the available talent. They're getting the fraction willing to perform for a camera with no audience. Replacing that with a live voice conversation — one that actually requires the candidate to think, not recite — recovers both the completion rate and the signal quality at the same time.

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