RefolkCandidates
PlaybookInterviewing

The One-Way Video Interview, From Invite to Submitted Takes

You will run a recorded video interview end to end, budgeting prep time and re-records question by question and submitting every answer before the deadline.

15 min readLast reviewed August 12, 2026Read as Markdown

A one-way video interview - also called an asynchronous or recorded interview - asks you to record answers to preset questions on your own, with no interviewer present, before a deadline. This guide is for a candidate who has just received one of those invites and wants to run it start to finish without improvising. It gives you a procedure that budgets the two resources that actually decide the outcome: a fixed prep window and a fixed number of re-records per question.

Most public advice on this format is a listicle about lighting, dress, and smiling. That advice ignores the two constraints that govern everything and, worse, tells you to optimize a signal the leading vendor has already discarded. This guide fixes both problems.

What a one-way video interview actually is, and what scores it

A one-way video interview is a set of preset questions you answer into a camera, alone, inside a deadline, with the recording submitted for later review by a person, a machine, or both. There is no live interviewer, so there is no back-and-forth to save a fumbled answer. The recording is the whole deliverable.

The scoring layer is where the format has changed and most guides have not kept up. HireVue, the dominant vendor, had been scanning candidate faces for 58 behavioral markers including gaze patterns. It then announced it was discontinuing facial analysis; its own internal research showed facial expression data contributed less than 0.25% to job-performance predictions. It shifted to natural language processing that evaluates the competencies of each candidate based on their verbal responses to questions against the requirements of the role.

The practical takeaway is blunt: the words you say and how you structure them carry the weight. Coaching your micro-expressions is optimizing a dead signal.

<0.25%
HireVue's own estimate of how much facial expression data contributed to job-performance predictions
The company dropped facial analysis and moved to scoring verbal responses against the role.

Questions arrive in more than one shape. You'll be prompted to respond to the questions via audio, typing, or in video format, depending on the role. HireVue also uses written assessment questions that appear as text on your screen, where you write an essay response, usually no more than 400 words. So "video interview" can include a typed component. Read the invite for which formats you will face.

The two constraints that decide the outcome

Two settings decide whether this interview is forgiving or one-shot: the prep window before each answer and the number of takes per question. Both are configured by the employer, not fixed by the platform, which is exactly why generic timing advice fails.

Here is what the three common platforms allow. Note the ranges, not the "standard" numbers.

PlatformPrep/think timeAnswer limitTakes/re-records
HireVue (common)~30 secup to 3 min2 total tries
Spark Hire30 sec-5 min (optional)30 sec-3 min1-5 or unlimited
Willoemployer-setemployer-set (3 min common)0 to unlimited

HireVue commonly gives about 30 seconds to prepare and up to 3 minutes to answer, with two total tries, meaning one re-record. Spark Hire lets employers set the response limit anywhere from 30 seconds to 3 minutes and the allowed takes from 1 to 5, or unlimited. Willo ranges from zero retakes to unlimited.

Two traps hide inside that table. First, on Spark Hire, when the employer limits think time, they also limit the candidate to one take. Think-time and single-take are linked, so a generous prep window can quietly mean you get no re-record. Second, on Willo, setting retakes to zero means the candidate's first response is automatically saved and uploaded - you do not get to review it.

Sizing up the interview you were actually sent

Long answer (up to 3 min)Short answer (<=60 sec)
One-shot, short
Script tightly and rehearse aloud; you get one clean pass, no room to explore.
Forgiving, short
Rehearse for concision; use a spare take only if you ran long or the audio failed.
One-shot, long
Outline in the prep window and hit STAR beats; do not aim for perfection you cannot re-record.
Forgiving, long
Budget one deliberate re-record for the two or three highest-value questions.
Few takes (1 take)Many takes (3+ or unlimited)
The two employer-set constraints put every one-way interview into one of four regimes.

How the invite-to-deadline window runs out

The window from invite to deadline is commonly 24 to 48 hours, but it is set per role, so the only number that matters is the one in your invite. Do not treat the format as instant. You want enough runway to set up, run the practice question, and record deliberately.

Within that window, several run-out mechanics can cost you an answer if you have not planned for them:

  • Prep timers auto-start recording. On HireVue, once the allotted prep time is up, the video starts recording your answer whether you're ready or not. On timer-governed platforms generally, when the timer is up, recording begins automatically.
  • Running out of takes submits your last one. On Spark Hire, the most recent answer is submitted when you run out of takes, and you cannot record more takes once you move on from the question.
  • The clock can include your retakes. Willo counts retake time against the timer and keeps it running across browser refreshes as an anti-cheat measure. "Start over" is not free.
A re-record is not a reset. On several platforms the timer keeps running and a wasted take is a take you no longer have.

The mechanics reward front-loaded preparation and punish a last-hour scramble. Get the setup and the practice question done early, then record when you are fresh.

The procedure, from invite to submitted

Run these steps in order. The time estimates assume a typical 5 to 8 question interview with a 24 to 48 hour window.

One-way video interview, end to end

  1. Read the invite and log the constraints
    Open the email link and record the deadline, question count, and any stated time and take limits. Note whether a prep timer auto-starts recording. Done when you know the expiry window and can state the takes and answer length per question.
  2. Set up room, device, and connection
    Use a supported browser such as Chrome, a charged device, headphones, and a plain background. Done when a short test recording plays back with clear audio and framing that keeps your eyes near the lens.
  3. Run the practice question
    Use the platform's unscored practice recording, which the recruiter cannot see. Done when you have confirmed audio and video work and you understand exactly how takes and the timer behave on this specific invite.
  4. Budget prep seconds and takes per question
    Before recording, decide how many takes you will spend on each question and reserve extra takes for high-value questions. If think-time is on, assume one take. Done when you have a written per-question plan.
  5. Record the answer to the lens
    Structure each answer with STAR and look into the camera, not the preview. Recording may auto-start when the prep timer ends, so be talking by then. Done when the answer fits inside the time limit.
  6. Review and decide on the re-record
    Play the answer back and check the audio first. Re-record only if the new take will be materially better and you still have takes left in your budget. Done when you accept the take or exhaust your allotment.
  7. Reach the final review screen and submit
    On platforms that allow it, review all responses before final submission; on platforms that lock a question when you advance, treat each submit as final. Done when the confirmation screen or email arrives.
  8. Exercise legal rights if applicable
    If the role uses AI scoring and you are covered by a regime such as Illinois AIVIA, you may decline analysis or request deletion within 30 days after the interview. Done when your request is acknowledged.

Budgeting takes: the step people skip

Step four is the one that separates this playbook from a listicle, so it deserves detail. You are allocating a fixed, non-renewable resource across a fixed set of questions. Do it on paper before you touch record.

The logic is simple: spend takes where a better answer changes the score, and hoard them where it does not. A "tell me about yourself" opener rarely needs three attempts. A behavioral question tied directly to the role's core competency is where a second, tighter take earns its keep.

Per-question take budget
Q1  Intro / warm-up          | takes: 1 of 2 | notes: accept unless audio fails
Q2  Motivation for role      | takes: 1 of 2 | notes: one clean pass
Q3  Core competency (STAR)   | takes: 2 of 2 | notes: highest value, re-record allowed
Q4  Conflict / failure       | takes: 2 of 2 | notes: re-record if I ramble past limit
Q5  Situational / role-fit   | takes: 1 of 2 | notes: reserve; think-time may force single take
Rule: if think-time is on for any question, plan for exactly one take.

Fill this in during setup, before you record anything. Adjust take counts to whatever the invite actually allows.

Delivering to the lens

Structure every answer so the machine and the reviewer can both follow it. STAR - Situation, Task, Action, Result - is the reliable frame because it front-loads the point and lands on an outcome. Say the result out loud; do not let it get cut by the time limit.

Look at the camera lens, not your own face in the preview window. Most firms allow 30 seconds for preparation and 1 to 3 minutes to record each response, so use the prep seconds to pick one example and its result, not to draft every word.

How this goes wrong: failure modes and false positives

Most one-way interviews are lost to a handful of predictable mistakes, not to a weak answer. Each one below has a tell and a check you can run before it costs you.

Failure modeWhat it looks likeThe check
Trusting a "standard" numberYou assume 30 sec / 3 min / 2 takes always appliesRead the invite and practice screen; Spark Hire and Willo allow one take to unlimited
Burning takes earlyThree tries on Q1, then a single-take question laterNote the take count per question and reserve for high-value ones
Prep timer auto-recordsYou plan to click record when readyHireVue records automatically when prep ends; confirm on practice
Advancing locks the questionYou move on, expecting to returnSpark Hire locks on advance; Willo allows final review
Optimizing facial expressionYou rehearse smiles and eye contact drillsFacial data scored below 0.25% and was dropped; focus on spoken content
Assuming legal noticeYou expect to be told AI is usedCompliance is patchy; ask the recruiter directly

The two that catch experienced candidates most often are burning takes early and the timer counting retakes against you.

Burning takes early is a resource error. You re-record question one three times chasing polish, then a later question with think-time enabled gives you one take and you fumble it with nothing in reserve. On Willo you can see remaining retakes, so check the counter before you spend.

The timer-counts-retakes problem is subtler. On Willo the clock includes retake time and keeps running if you refresh the page. A candidate who thinks a refresh resets the situation instead loses the difference. Treat a re-record as spending from the same budget as the original answer, not as a fresh start.

Your rights when a machine scores the recording

Three legal regimes govern AI video scoring, and each gives you different levers, but enforcement is uneven enough that you should treat notice as something to request, not receive. Here is what each requires.

RegimeNotice to candidateConsent/opt-outDeletion right
Illinois AIVIABefore interviewYes, written consentYes, 30 days
NYC LL144>=10 business daysOpt-out referencedNot specified
EU AI ActYes (transparency)Explanation rightVia GDPR

Illinois AIVIA requires employers to inform applicants when AI is used during a video interview, explain the AI's mechanics, obtain prior consent, and, on request, delete the interview and instruct other parties to delete it within 30 days of receipt. NYC Local Law 144 requires bias audits and at least 10 business days' notice before the automated tool is used in your evaluation. The EU AI Act classes recruitment and candidate-evaluation AI as high-risk, with mandatory risk assessments, bias testing, human oversight, and transparency disclosures, and core obligations dated 2 August 2026.

Two caveats keep this honest. First, the NYC deletion right is not specified in the law as written, so do not assume it. Second, the EU AI Act's core obligations are dated to a future point, and a proposed deferral was not yet adopted at the time of the dossier's sources, so re-check the effective date against the current statute rather than trusting a fixed date here. If your rights matter to a decision, the reliable move is to email the recruiter and ask directly.

Finding help for the format, and why it is scarce

Dedicated one-way-interview coaching is a genuinely thin specialty, so if you want expert help you are fishing in a small pool and should search precisely. In Refolk's index of professional profiles, there are 46 US professionals with the current title "Interview Coach" against 4,695 with "Career Coach" - roughly a 102x gap.

Title (US)CountRatio to Interview Coach
Interview Coach461.0x
Career Coach4,695102x

That scarcity has two consequences. Most people advertising as career coaches are generalists who may never have coached this exact format, and the specialists who exist are hard to surface by keyword. In Refolk's index, the "Interview Coach" profiles sit mostly at individual practices and firms rather than a dominant platform, and a search for US or UK recruiters naming "HireVue" in their headline keywords returned zero results - platform-specific screening expertise is rarely stated publicly.

So a keyword search on a jobs site will mostly miss the people you want. Describing the exact expertise in plain language works better than matching a title. Refolk writes your resume from your own history, tailors it to each posting, and finds people by description rather than by the label they happen to carry.

Why keyword search misses the specialists

  1. Career Coaches (US)
    4,695

    generalists, most never coached this format

  2. Interview Coaches (US)
    46

    the specialist niche you actually want

  3. Recruiters naming HireVue in headline
    0

    platform expertise not surfaced by keyword

The pool of interview-specific coaches is a fraction of the career-coach population, and platform expertise is rarely stated in a headline.

If you do want a specialist, search by what they do, not what they are called.

Before you hit submit

Run this checklist on the final review before you close the interview. It is the last cheap moment to catch a problem the reviewer cannot un-see.

Pre-submission check

  • The deadline in the invite is still in the future with margin to spare
  • Every question has a recorded response, none left blank
  • Audio is clear on playback for each answer, not just video
  • Each answer fits inside the platform's time limit without being cut off
  • My eyes are on the lens in the takes I kept, not on the preview
  • I used a re-record only where it made a materially better answer
  • I have not advanced past any question I still wanted to revisit
  • The confirmation screen or email has arrived

Keeping this current

The mechanics in this guide are set by employers and change per role, and the legal dates are moving targets, so re-verify two things on every interview. First, confirm the takes, prep window, and answer limit from the actual invite and the practice screen rather than from any remembered "standard" - that single habit prevents most of the failure modes above. Second, if AI scoring and your rights matter to the decision, check the current effective date of the regime that covers you and ask the recruiter directly, because published notice is unreliable in practice. The procedure itself is stable; the numbers it operates on are not, which is precisely why the first step is to read the invite and log the constraints.

Questions job seekers ask

how many takes do you get in a one-way video interview

It depends on the employer, not the platform. HireVue commonly gives two total tries, meaning one re-record. Spark Hire lets the employer set 1 to 5 takes or unlimited, and Willo ranges from zero to unlimited. Critically, Spark Hire forces a single take whenever think-time is enabled. Read the invite and the practice screen to confirm your actual count before you start, because assuming a standard number is the most common way candidates get burned.

does the prep timer start recording automatically

On several platforms, yes. HireVue starts recording automatically once the prep time is up, whether you are ready or not. Treat the prep window as a countdown to speaking, not a decision point where you click record when comfortable. Confirm the behaviour on the practice question so a live question does not catch you mid-thought with the camera already rolling.

should i optimize my facial expressions for AI scoring

No. HireVue's own internal research put facial expression data below 0.25% of its job-performance prediction, and the company dropped facial analysis, shifting to natural language processing that scores your verbal responses against the role. Older prep pages still repeat the facial and body-language claims, but for the leading vendor they are outdated. Spend your effort on clear, structured spoken answers instead.

how long do i have to complete a one-way video interview

The invite-to-deadline window is commonly 24 to 48 hours, but it is set per role, so check the deadline stated in your invite. Do not wait until the last hour: you want time to set up, run the practice question, and re-record deliberately rather than racing the clock. If a platform's timer counts retakes against you, as Willo's does, a rushed session costs you takes as well as time.

can i re-record after i move to the next question

It depends on the platform. Spark Hire locks a question once you advance and submits your most recent take, and you cannot record more takes after moving on. Willo, by contrast, offers a final review screen that lets you review and re-record before final submission. Verify which behaviour applies on your practice run so you do not advance expecting to come back.

what are my legal rights if AI scores my video interview

They vary by jurisdiction. Illinois AIVIA requires notice, an explanation of the AI, prior consent, and deletion within 30 days on request. NYC Local Law 144 requires a bias audit and at least 10 business days' notice. The EU AI Act treats hiring tools as high-risk with transparency duties. In practice, compliance is patchy, so if it matters to you, ask the recruiter directly whether AI is used and how to opt out.

Put this to work

Reading about the job search is not the job search.

Paste your career in once. I write the resume, then every week I rank the live openings against your history, tailor a resume and a cover letter to the best of them, fill in the forms if you ask me to, and keep going until you land. Your part is deciding what goes out.

  • 140+ curated roles a week, found, written, and scored for you.
  • Every bullet stays inside what your history actually supports.
  • Queued, submitted, interviewing, offer, all in one place instead of a spreadsheet.

500 free credits on sign-up. No card.

Read next