Is asking candidates to do a video and an assessment too much before you've even talked to them?

Here’s a fear worth taking seriously instead of talking someone out of: you post a role, you build a one-way interview, and then you’re staring at the option to add an assessment on top of it, and something in you says that’s one ask too many before this person has even talked to a real human at your company.

That instinct isn’t wrong on its face. Every extra step is a place a candidate can quit. But the assumption underneath it, that adding a second step is what costs you the candidate, doesn’t hold up against the one piece of real data on how candidates actually feel about assessments. Criteria Corp’s 2025 Candidate Experience Report, based on responses from more than 2,000 job seekers worldwide, found that 86% either like or feel neutral about taking assessments when applying for jobs. That’s not the reaction you’d expect if the ask itself were the thing driving people away.

So if the second step isn’t the problem, something else is. This is about what that something else actually is, when stacking a one-way interview and an assessment is worth it, when it isn’t, and how to ask for both without losing the candidates you actually want.

What the completion-rate fear gets right, and what it gets wrong

The fear gets one thing right: a screening process with more steps than the role justifies does bleed candidates. Nobody’s disputing that. Where it goes wrong is treating step count as the whole explanation, because that leads to the wrong fix. The wrong fix is cutting the assessment to protect completion, on every role, regardless of what the role actually needs to know.

We’ve written before about why the completion-rate conversation gets stuck at the wrong stage of the funnel. Most of the drop-off happens at the application form itself, long before anyone reaches a screening interview or an assessment. That’s a different leak, and fixing it doesn’t tell you anything about whether your second screening step is the right call.

The 86% figure matters because it isolates the one variable people assume is toxic and shows it isn’t, on its own. Candidates aren’t reflexively allergic to being asked to demonstrate something. What they’re reacting badly to, when they do react badly, is something more specific than “there were two steps instead of one.”

What actually drives a candidate to quit partway through

Picture the two most common ways a one-way interview plus an assessment gets sent. Version one: a form email, no context, a link to “complete your next step,” with no indication of how long it takes or what happens after. Version two: a short note that says this role gets 40 applicants a week, the video takes 8 minutes, the assessment takes 12, and the hiring manager personally reviews every completed one within 48 hours.

Same two steps. Wildly different completion. What changed between the two versions is whether the candidate can tell, at the moment they open the link, that finishing it is worth their time. That’s the same pattern behind most candidate experience problems in an automated process: the automation runs fine. The silence around it is what candidates actually notice.

Be honest about the drop-off trade-off

We’ve heard this from owners running screening themselves more than once: the honest answer to “will candidates actually do this” is that some won’t, and that’s a real trade-off worth naming rather than hiding from. A one-way interview earns its place on a role that pulls a big, loosely-qualified crowd (the kind of posting that turns into hundreds of applicants inside a week), where a short recorded answer tells you more than a resume ever could. For a role with a thin candidate pool, where you want to talk to almost everyone who applies, you can skip the video and lean on ranked resumes and an assessment instead. The process is yours to design, and being upfront about the drop-off trade-off beats pretending it doesn’t exist.

There’s also a cost most owners assume exists that doesn’t. On Truffle, you’re never charged for a candidate who starts an assessment or interview and doesn’t finish it. A credit is spent only when a screening produces a result you can act on: a completed assessment, a submitted one-way interview, a resume scored by AI Match. That changes the actual risk calculation. Testing whether a second step is worth it for a specific role doesn’t cost you anything extra when candidates decide it isn’t for them. It only costs you something when it works.

The honest case for keeping it to one step

None of this means stack methods on every role. There’s a real version of the fear, and it shows up on a specific kind of pool: a thin, hard-to-fill role where you can’t afford to lose anyone, and where you’d rather talk to a mediocre-on-paper candidate than risk losing a strong one to a process that felt like too much before a human ever entered the picture.

If you’re hiring for a specialized role where five qualified people exist in your market and you need one of them, adding a second screening step before any human contact is a real risk, not a framing problem. The fix there isn’t better copy on the invitation email. It’s using fewer methods, earlier human contact, or skipping the assessment on this specific role even though it might tell you something useful. The same logic that explains why candidates go quiet on a slow process applies here in reverse: on a thin pool, you can’t afford to be the reason a scarce candidate looks elsewhere. Acknowledging that limit is part of using the tool honestly instead of treating more evidence as always better.

The mistake isn’t stacking a video and an assessment. The mistake is stacking them as a default on every role, including the ones where the pool can’t absorb it.

The real filter is pool depth, not step count

So the actual decision rule has almost nothing to do with how many steps feel like “a lot.” It has to do with what kind of pool you’re screening.

A high-volume, loosely-qualified pool (think a coordinator, sales, or front-desk role that pulls 150 to 300 applicants from a single posting) is exactly where a second step earns its keep. Each screening method answers a different question, and a resume alone can’t tell you who can actually communicate under a live-ish prompt or whether their judgment matches how your team actually handles a real scenario. A one-way interview tells you the first. An assessment tells you the second, on things a resume and even a video interview can’t surface on their own. Talent assessments exist for exactly this gap: a personality read based on the validated IPIP/Big Five model, a situational judgment test built around how your team actually handles real scenarios, or an environment fit check on whether the day-to-day reality of the role matches what the candidate actually wants. None of these claim to predict who’ll succeed. They show you alignment on something a resume hides, which is a different, more honest claim.

A thin, specialized pool is the opposite case. There, every additional step is pure downside, because you don’t have the volume to absorb the candidates you’d lose. Same tools, opposite call, and the variable that decides it is how many real candidates you can afford to lose, not how many steps feel reasonable in the abstract.

How to frame the ask so candidates actually finish it

Once you’ve decided the pool justifies a second step, the framing is what determines whether people finish it. Three things do most of the work, and none of them require rebuilding your process.

The three things that determine whether people finish it

Say how long it takes, in the invitation itself, not buried in the assessment’s own instructions. “8 minutes for the video, 12 for the assessment” beats “complete your next step” because it lets the candidate decide with real information instead of guessing and assuming the worst.

Say who looks at it and when. A candidate who knows a person, not a queue, reviews their submission within a defined window treats the ask differently than one who suspects it disappears into a black hole. This is also where being specific about what the AI actually does helps, not hurts: the honest version is that AI transcribes and scores the response against your stated criteria, and you’re the one who reviews it. Nobody’s evaluation gets auto-rejected by an algorithm they can’t see.

Connect the ask to the actual job. A one-way interview question that mirrors a real situation on the role reads as relevant. A generic “tell us about yourself” prompt reads as busywork, and busywork is what people actually quit on, not the format.

This is the version of the workflow that holds up: resumes get ranked the moment they arrive, so you already know who’s worth asking for more from. The invitation to the one-way interview and the assessment goes out together, with the time and the reviewer named up front. Completed responses come back scored against your criteria, with Candidate Shorts surfacing the most relevant moments so you’re reviewing in minutes, not watching every recording start to finish. The steps didn’t shrink. The candidate just always knew what they were walking into.

Screening depth is a decision, not a default

The number of steps was never the real lever. Candidates don’t quit because you asked for two things. They quit because you asked for two things and never told them why.

That reframes the actual question you’re answering every time you build a screen. Not “how many steps can I get away with before candidates bail,” but “does this specific role, with this specific pool, justify a second layer of evidence, and can I explain the ask in one honest sentence.” Roles that pull a flood of loosely-qualified applicants usually answer yes on both counts. Roles pulling from a pool of five people in your market usually don’t, and that’s fine too.

Treat the decision as something you make per role instead of a policy you apply everywhere, and the completion-rate fear mostly resolves itself. You’re not fighting the ask. You’re being honest about it.

Ready to see what a resume-plus-interview-plus-assessment flow looks like for one of your own roles? Try Truffle’s talent assessments with a 7-day free trial, 30 credits, no card required.

Frequently asked questions about combining a video interview and an assessment

Does adding an assessment after a one-way interview always lower completion rates?

Not automatically. Criteria Corp’s 2025 Candidate Experience Report found 86% of job seekers like or feel neutral about taking assessments, so the ask itself isn’t the default problem. What lowers completion is an unexplained ask on a pool too thin to absorb any drop-off, or a request with no stated time investment or reviewer.

How do I know if my role’s candidate pool can handle a two-step screen?

Look at your applicant volume and qualification spread. A role pulling 100+ loosely-qualified applicants from one posting can usually absorb a second step, because you have enough volume that some drop-off still leaves a strong shortlist. A role with a handful of realistic candidates in your market usually can’t.

What should I say when inviting a candidate to do both a video interview and an assessment?

State the time each part takes, who reviews the results and by when, and how the tasks connect to the actual work in the role. Vague invitations read as busywork. Specific ones read as a real step in getting hired.

Does it cost more to test whether a second screening step works for a role?

On Truffle, no. You’re only charged when a screening produces a completed result, resume score, finished assessment, or submitted interview. A candidate who starts and doesn’t finish doesn’t cost you anything, so testing a second step on one role before rolling it out further carries no extra downside.