What Is an Asynchronous Interview? One-Way Video, AI Avatar Video Interviews and How to Choose

An asynchronous interview is an interview in which the candidate and the interviewer are never present at the same time. The employer sets the interview, the candidate completes it on their own schedule, and the hiring team reviews the output afterwards. The word describes when the interview happens, not how it is conducted. That distinction matters more than it used to, because the asynchronous video interview now has a second layer available on top of it: the AI Avatar video interview, where an AI interviewer conducts a structured, two-way conversation and returns a score, summary and transcript for a human to review.

The base format has not gone anywhere. What has changed is what you can add to it.

Looking for interview preparation advice? This guide is written for recruiters, talent acquisition teams and hiring organisations deciding how to run their screening process. If you are a candidate preparing for a recorded or AI-led interview, this page will not help you. You want a candidate preparation resource instead.

What Is an Asynchronous Interview?

An asynchronous interview is a structured screening method in which the employer and the candidate are never online simultaneously. The hiring team defines the questions and the assessment approach, the candidate completes the interview independently within a deadline, and the recruiter reviews the result when it suits their schedule.

The opposite is a synchronous interview, meaning a live phone screen, video call or in-person meeting where both parties are present at the same moment.

Synonyms you will encounter

• One-way video interview

• Pre-recorded video interview

• Async video interview

• Asynchronous video interview

• Self-paced video interview

• Video screening interview

• On-demand interview

These terms get used interchangeably, but they are not all doing the same job. "Asynchronous" and "on-demand" describe timing, meaning the interview does not require a shared calendar slot. "One-way" and "pre-recorded" describe a format, meaning the candidate records answers and nothing responds. Every one-way video interview is asynchronous. Not every asynchronous interview is one-way, because the AI Avatar video interview layer keeps the same timing property while replacing the recording with a conversation.

Holding those two ideas apart is what makes the rest of this article usable. If you treat "asynchronous" as a synonym for "recorded", you will evaluate the wrong things in a demo.

How Does an Asynchronous Interview Work?

The mechanics are consistent across platforms and typically involve five steps.

1. The hiring team defines the interview. This means a question set, whether behavioural, role-specific or a mix, and the assessment criteria those questions are meant to test. On some platforms the questions are written manually. On others, the system drafts role-specific questions from the job description for the recruiter to edit and approve.

2. The team sets the parameters. Think time before a candidate begins answering and a maximum response length are both standard controls. Retake allowances and the submission deadline are set here too.

3. The candidate receives an invitation link. Modern tools run in a browser on desktop or mobile with no app download and no account creation.

4. The candidate completes the interview in their own time, before the deadline, without booking anything with anyone.

5. The hiring team reviews the result. What "the result" consists of depends on whether the AI Avatar layer is switched on. In a standard asynchronous video interview it is a set of recordings a person has to watch. With the AI Avatar video interview layer it is a score against the defined criteria, a written summary, a full transcript and the recording, so review can start from the structured output and drop into the footage where it matters.

Step five is where the difference lands. Everything else about asynchronous interviewing has been stable for years. What has changed is what arrives on the recruiter's desk at the end.

Want the candidate's side of this? See our guide to candidate experience in asynchronous interviews.

Asynchronous vs Synchronous Interviews

‍
The scheduling row is the one that drives adoption. A recruiter managing 30 open requisitions who needs to speak to 10 candidates for each of them is looking at 300 separate scheduling negotiations before a single offer goes out. Asynchronous interviewing removes that step entirely rather than making it faster.

Asynchronous Interviewing Has Evolved: The AI Avatar Video Interview Layer

For most of the format's history, the asynchronous video interview did one thing. The candidate recorded, the recruiter watched. That solved scheduling and left two problems untouched.

The first is review load. Removing the live call does not remove the viewing time. A team screening several hundred candidates still has to watch several hundred recordings, so the recruiter hours simply move from the calendar to the review queue.

The second is signal quality. A recorded answer to a fixed prompt, with retakes allowed and no follow-up possible, is a rehearsed artefact. It cannot distinguish a candidate who understands a subject from one who prepared a polished paragraph about it, because nothing in the format can ask "why did you make that choice?" This has become more pressing as candidates use AI assistance during unmonitored recorded interviews, an issue we cover separately in our analysis of interview fraud.

The AI Avatar video interview is the answer to both, and it is best understood as a layer rather than a replacement. It sits on top of the same asynchronous video interviewing foundation and leaves the timing property untouched. The candidate still interviews on their own schedule, still with no recruiter present, still from a link. What the layer changes is the middle and the end of the process: instead of collecting recordings, a branded AI Avatar interviewer conducts a structured two-way interview, and instead of returning footage, it returns a structured assessment.

‍

The base format remains valid on its own, and switching the layer on is a per-requisition decision rather than a platform-wide one. Standard asynchronous video interviewing is the simpler, lower-cost option and gives candidates something a conversational format cannot, which is control in the form of retakes. Most enterprise teams end up running all three options at different stages and for different roles: live interviews, asynchronous video interviews, and asynchronous video interviews with the AI Avatar layer applied.

What the Difference Looks Like in Practice

A standard asynchronous video interview asks a question and waits:

"Please record a two-minute response describing how you handled a difficult customer."

The candidate reads the prompt, thinks, records, possibly re-records, and submits. The interview does not react to anything they say.

An AI Avatar video interview conducts a structured conversation instead. The AI Avatar interviewer introduces the role, asks the questions the hiring team has configured or approved, and continues the interview according to that defined assessment approach, responding to what the candidate actually says rather than moving mechanically to the next fixed prompt. Where a recorded prompt collects an answer, a conversation can pursue one.

The trade-off is real and worth stating plainly. Candidates who value the safety of a retake generally prefer the recorded format. Candidates who find recording into a silent camera uncomfortable generally prefer the conversation. Some candidates object to being interviewed by an AI at all, and that reaction attaches to your employer brand rather than to your vendor's. Turning the layer on is a candidate experience decision as much as an efficiency one.

Which Option Should You Use? A Decision Framework

There are no hard rules here, and any vendor offering you one is selling. These are the conditions under which each option tends to be the right call.

Live interviewing is the right call when:

• The role is senior or executive and the interview carries mutual selling weight

• The conversation is confidential or commercially sensitive

• Assessment depends heavily on rapport, negotiation or stakeholder dynamics

• The candidate pool is small enough that scheduling was never the constraint

A standard asynchronous video interview is sufficient when:

• Standardised, comparable responses to a fixed question set answer your assessment need

• No conversational follow-up is required to interpret the answer

• Candidate control matters, because retakes lower the stakes for nervous or early-career applicants

• You want the simplest possible step between application and human review

Adding the AI Avatar video interview layer earns its place when:

• Hiring volume is high and scheduling is the binding constraint on throughput

• Consistency of questioning across recruiters, regions and requisitions matters

• Candidates need to interview at any hour, in their own language

• Your recruiters need structured outputs to work from rather than a queue of full-length recordings

• Interview integrity signals are relevant to the roles you are hiring for

The useful question is not "which is best?" It is "which stage of which funnel is this for?" Most teams land on a mix, with asynchronous interviewing at first round and live interviewing from the hiring manager stage onwards, and the AI Avatar layer switched on for the requisitions where volume and consistency justify it.

When Asynchronous Interviewing Is the Wrong Choice

Asynchronous interviewing is genuinely a poor fit in several situations, with or without the AI Avatar layer, and knowing them protects both your hiring outcomes and your employer brand.

Senior and executive hiring:

‍At this level the interview is a two-way evaluation and often a negotiation. Asking a director-level candidate to record answers, or to speak to an AI Avatar interviewer, signals how you value the relationship. Some organisations do run structured asynchronous steps for senior roles with heavily adapted design. Most should not.

‍

Small candidate pools.‍

If you have eleven applicants, an asynchronous stage adds a drop-off point and saves almost nothing. The scheduling arithmetic that justifies the format only works at volume.

‍

Roles where the interview is a selling conversation.

In competitive skill markets, the first conversation is often your best chance to convince someone to stay in the process. Automating it can cost you the candidates you most wanted.

‍

Accessibility constraints.

Candidates with speech, hearing, processing or anxiety-related conditions, and those without reliable devices or connectivity, may be disadvantaged by a timed recorded or conversational format. An alternative route must be available and easy to request. This is a design requirement, not an edge case.

‍

Contexts where candidate sentiment carries real cost.

Attitudes to automated interviewing vary by market, sector and seniority. Where your candidates are likely to react badly, the efficiency gain may not survive contact with your application completion rate.

What Enterprise Buyers Should Evaluate in an Asynchronous Interview Platform

If you have been handed a platform to trial, these are the criteria that separate products. Use them as questions, not as a feature checklist to tick.

Interview experience

• Whether the AI Avatar video interview layer is available, and whether you can switch it on per requisition rather than per account

• Browser-based on desktop and mobile, with no app download or account creation

• Genuinely no scheduling step anywhere in the candidate flow

• Multilingual interaction, and what "multilingual" actually covers

• Employer branding on the interview experience itself, including the AI Avatar interviewer

• A documented alternative route for candidates who need one

Assessment quality

• Role-specific question generation, and how much control your team keeps over it

• Structured interview design with defined criteria

• Consistent questions and a consistent scoring framework across candidates

• Score, written summary and full transcript, not just the recording

• Human review controls, meaning who can see, override and comment on a score

Integrity

• Which signals the platform captures, named specifically

• How flags are presented to a reviewer, and with what context

• What the candidate is told, and when consent is captured

• Whether a human reviews every flag before any adverse decision

Workflow

• ATS integration, and whether scores and summaries write back automatically

• Recruiter dashboard and reporting across requisitions

• How the product behaves at your actual volume, not the demo volume

Enterprise trust

• Who owns the candidate data, and where it sits

• How recordings and transcripts are stored, protected and retained

• What audit trail exists for each interview

• How the vendor evidences its security and data-handling posture when your security team asks. Ask to see the documentation rather than accepting a badge. Recruitment Smart's own position is set out on our security and compliance page.

Take this into your evaluation. We have turned the criteria above into an asynchronous and AI Avatar video interview evaluation checklist covering interview experience, assessment quality, integrity, workflow and enterprise trust, built to be used in a vendor demo and shared with your security and legal reviewers.

Privacy, Consent, Integrity Signals and Human Oversight

This is where enterprise deployments succeed or stall, and it is worth being precise about two things.

An AI-generated score supports a human decision. It does not make one.

A score is a structured summary of evidence against defined criteria, an input to a recruiter's judgement in the same category as a structured interview scorecard. Any product or process that advances or rejects candidates without a person reviewing the underlying evidence has changed what the tool is, and changes your obligations with it.

‍

Integrity signals are signals, not verdicts.

Platforms that monitor interview integrity typically capture things like copy-and-paste activity, page reloads, changes to the audio or video setup, window focus changes, the presence of additional devices, and anomalies in response patterns. None of these is proof of anything. A page reload is usually a bad connection. A window focus change is usually a notification. These signals exist to tell a reviewer where to look more closely. They should never trigger an automatic rejection, and they should never be presented to a hiring manager as a cheating verdict.

Questions worth putting to any vendor:

• What consent is captured from the candidate, at what point, and in what wording?

• What exactly is monitored during the interview, and is the candidate told before they start?

• How long are recordings, transcripts and scores retained, and who decides?

• Can a candidate request deletion, and what is the process?

• Is a human required to review before any adverse decision, and is that enforced by the product or only by our policy?

• What audit trail exists showing who reviewed what, and when?

• What data flows into our ATS, and what stays with the vendor?

• Can we export or delete everything if we leave?

Obligations around automated employment decision tools, candidate notice, consent and biometric data vary significantly by jurisdiction and change often. Verify current requirements for every location you hire in with your own legal advisers before deploying. That is a legal question rather than a procurement one, and this guide is not a substitute for advice.

For teams moving from evaluation into deployment, our guides on how to run a bias-safe AI interview pilot and on ethical AI video interviewing go considerably deeper on oversight design.

How Asynchronous and AI Avatar Video Interviews Fit Into the Recruiting Workflow

In a typical enterprise funnel, the asynchronous interview sits after application review and screening automation, and before the first hiring manager conversation. It replaces the phone screen, not the interview loop.

The part that determines whether it actually saves anyone time is what happens to the output. If a recruiter has to open a separate tool, watch recordings and retype conclusions into the ATS, the scheduling saving has been spent on administration. The workflow only closes when scores, summaries and transcripts land on the candidate record in the system your team already works in, and when the recruiter's job becomes reviewing evidence and deciding. That is the point at which their judgement enters, and the only point at which a decision is made.

This is where Recruitment Smart's AI video interviewing platform sits. VScreen provides asynchronous video interviewing as the foundation, with the AI Avatar video interview available as an additional layer on the same platform. With the layer applied, a branded AI Avatar interviewer conducts a structured two-way interview on demand, generates role-specific questions from the role, and interacts with the candidate in their own language. Each completed interview returns a score, summary and transcript alongside integrity signals that a human reviews before any adverse decision, and results flow through to the candidate record via our ATS integrations rather than being copied across by hand.

On outcomes, Recruitment Smart's product materials cite figures including up to 80% less time spent on screening, 40% faster time to hire and 20+ hours saved per recruiter per week. These are Recruitment Smart's own reported product figures drawn from customer deployments. They are not independent industry benchmarks, and your results will depend on your volumes, roles and current process. Ask us, and every vendor you evaluate, for the deployment context behind any number you are shown.

Asynchronous Interview FAQ

Is an asynchronous interview the same as a one-way or pre-recorded interview? Not quite, and the difference now matters. "Asynchronous" means the candidate and interviewer are never present at the same time. "One-way" and "pre-recorded" describe a specific format in which the candidate records answers and nothing responds. All one-way interviews are asynchronous. An AI Avatar video interview is also asynchronous, because the candidate still completes it on their own schedule with no recruiter present.

Is an AI Avatar video interview the same as an asynchronous interview? An AI Avatar video interview is a layer applied to an asynchronous video interview, not a separate category. The timing is identical. What changes is that the candidate speaks with an AI Avatar interviewer rather than recording answers to fixed prompts, and the hiring team receives a structured assessment rather than raw footage. Vendors define this inconsistently, so ask specifically whether the interview is genuinely conversational and how much control your team keeps over the questions.

Do we have to use the AI Avatar layer to run asynchronous interviews? No. The asynchronous video interview works on its own and remains the right choice for plenty of requisitions. The layer is something you switch on where volume, consistency or multilingual interviewing justify it.

Does the AI make the hiring decision? It should not, and in a properly configured process it does not. AI Avatar video interviews produce structured evidence, meaning a score against defined criteria, a summary and a transcript, which a recruiter or hiring manager reviews before any decision. If a vendor's product advances or rejects candidates without human review, that is a materially different tool and should be evaluated as such.

What happens if a candidate is flagged by integrity monitoring? A flag routes the interview to a human for closer review. It is not a rejection and it is not evidence of cheating. Most flags have innocent explanations. Your process should define who reviews flags, what additional evidence they consider, and how a candidate can respond before any adverse decision is taken.

Do candidates have to consent to being recorded or monitored? Candidates should always be told what is being captured before the interview begins, and consent requirements vary by jurisdiction. Treat this as a legal question for your own advisers rather than a setting to switch on.

Can candidates opt out or ask for a different format? They should be able to, and a documented alternative route is a reasonable requirement to put to any vendor. This matters most for candidates with accessibility needs and for those without reliable devices or connectivity.

Is an asynchronous interview fairer than a live interview? It is more consistent, which is not the same claim. Every candidate faces the same questions under the same conditions, which removes the interviewer variance that makes unstructured phone screens hard to compare and hard to defend. Consistency is a precondition for fairness rather than a guarantee of it. Fairness also depends on question design, scoring criteria, human oversight and regular review of outcomes.

How long should an asynchronous interview be? Shorter than you think. Completion rates fall as interviews lengthen, so the working constraint is candidate patience rather than assessment ambition. Design for the smallest number of questions that produce a decision, and measure your own completion rates rather than trusting a benchmark.

Where does an asynchronous interview sit in the process? Usually as a replacement for the first phone screen, after application review and before the hiring manager conversation. It is most valuable on roles with large applicant volumes, where synchronous screening at that scale would consume a disproportionate share of recruiter capacity.

The Bottom Line

Asynchronous interviewing is a timing decision before it is a technology decision. Once you separate when the interview happens from how it is conducted, the choice becomes clearer. Live interviewing where the relationship is the point, asynchronous video interviewing where standardised answers are enough, and the AI Avatar video interview layer on top of it where volume, consistency and structured output are what you are short of. Most enterprise teams need more than one of the three.

See how an AI Avatar video interview actually runs. If you want to watch a structured AI Avatar video interview end to end, including the candidate experience, the score and transcript a recruiter receives, and how integrity signals are surfaced for human review, we can walk you through it on your own roles.

Book a walkthrough if you are scoping a requirement, or start with the VScreen white paper for enterprise hiring if you are still building the internal case.

‍