Written pre-screening or one-way video: what each one measures

Both formats promise the same thing: a look at the candidate before anyone spends an hour on a call. They do not deliver the same thing, because they do not ask for the same thing. A one-way video asks a person to perform an answer under a timer. A written questionnaire asks a person to produce one.

That difference decides what you learn, who drops out, and what you can defend later if the candidate asks why they were rejected.

What a one-way video actually measures

A recorded answer carries everything at once: the reasoning, the delivery, the accent, the room behind the person, the webcam they could afford, and how they look on camera at seven in the evening after a full working day.

The reasoning is the part you wanted. The rest arrives whether you want it or not, and it does not sit quietly in the background. It shapes the impression before the content does, and it does so in a way that is very hard to notice from the inside. Reviewers are not lying when they say the strong candidate just came across better.

There is a second thing being measured that nobody chose to measure: comfort on camera. For a sales role that is close to the job. For a backend engineer, a data analyst, or a technical writer, it is noise with a confident face on it.

What written answers measure

Written answers strip the channel down to the thing most knowledge work actually runs on. The candidate has time to think, and they have to put the thinking into words that survive being read by someone who was not in the room.

That is not a lesser signal. In most roles, it is closer to the work than a recorded monologue. A person who can explain a tradeoff clearly in three paragraphs will explain it clearly in a design doc, a pull request comment, and a message to a client.

Written answers also travel. A recorded video has to be watched, in order, at the speed the candidate spoke. A written answer can be read by two people in five minutes, quoted in a hiring discussion, and compared side by side with another candidate's answer to the same question.

Where each format misleads you

Neither is clean, and pretending otherwise is how teams end up trusting a format more than it deserves.

Video misleads through charisma. Fluency reads as competence. The candidate who pauses, restarts, and then gives the better answer scores worse than the one who gives a polished non-answer without breaking eye contact. Non-native speakers pay this tax twice.

Written misleads through generated text. This is the obvious objection now, and it is a real one. A candidate can paste a question into a model and send back four confident paragraphs that say nothing they could defend.

The answer is not to ban the tool, which is unenforceable, and not to try to detect it, which does not work: a candidate writing with an assistant from their own notes produces true answers in their own words. The answer is to look at what the text gives you to check. A generated answer in the abstract names nothing; a real one names a company, a product, a decision with a price. Eneya reports that next to the score, High or Medium, with the fragments that gave the reading, and it is a reason to ask one more question on the call, not a verdict.

Who drops out, and why that matters more than it sounds

Every screening step loses candidates. The question is which ones.

A one-way video loses people who are employed and cannot record during the day, people who share a flat, people who are uncomfortable on camera, and people who have simply decided that recording a monologue for a company that has not yet spoken to them is not a fair trade. Some of those are exactly the senior, currently-employed candidates you wanted.

A written questionnaire loses people who will not write. Sometimes that is a signal about the role fit. Sometimes it is just length: a questionnaire that takes an hour will be abandoned by good candidates who have three other processes running.

Keeping the question count low is not a courtesy. It is what keeps the top of your funnel from filtering by patience.

What to do with the choice

Some practical lines that hold up in most hiring processes:

• Match the format to the medium of the job. If the work is written, screen in writing. If the job is talking to people all day, hearing the person talk is a real signal, not a vanity one.
• Ask questions that have a wrong answer. "Tell us about yourself" measures preparation. "Here is a situation from this role, what would you do first and why" measures judgement, in either format.
• Decide what the step is for. Screening is not interviewing. Its job is to tell you who is worth an hour, and to give the interviewer something specific to dig into.
• Write down the reason before you look at any score. Whatever tooling you use, a judgement formed after seeing a number is not really your judgement.
• Keep the evidence. A rejected candidate may ask why. An answer you can point at is a better position than a recollection of a video.

The short version

One-way video buys you presence and costs you fairness and time. Written pre-screening buys you comparable evidence and costs you vigilance about how the text was produced.

For most roles where the job is thinking and explaining, written is the straighter line to what you actually need, provided you keep it short, ask questions worth answering, and treat any automated reading of the answers as something that points you at a question rather than something that answers it for you.

That last part is not a limitation of the tooling. It is the job.