A structured interview asks every candidate the same job-relevant questions, in the same order, and scores each answer against a rubric defined before anyone is interviewed. An unstructured interview is a free-flowing conversation: questions vary by interviewer, order changes with the discussion, and evaluation rests on overall impression.

The difference is not formality. It is comparability: whether two candidates can be judged on the same evidence. For a deeper look at how the structured format works when AI delivers it, see our blog on structured AI interviews.

In practice, the two formats differ on every dimension that matters. Structured interviews use identical questions for every candidate, score criterion by criterion against a predefined rubric, and leave an audit trail where every score traces to a specific answer. Unstructured interviews vary by interviewer and moment, rest on overall impression, stay open to affinity and first-impression bias, and leave little to reconstruct after the fact.

Three failure modes show up again and again.

Prediction. An unstructured interview measures how the conversation felt, which is heavily shaped by charisma, confidence, and similarity to the interviewer. None of those reliably predict job performance, especially in frontline roles where the work looks nothing like the interview.

Consistency. The same candidate can interview on Tuesday and Thursday with two different people and get two different outcomes. When results depend on who conducted the interview and when, the process is measuring the process, not the person.

Defensibility. If a rejected candidate, a stakeholder, or a regulator asks why a decision went the way it did, an unstructured process has little to show. There is no rubric, no comparable record, and often no notes beyond a gut verdict.

This is one of the most studied questions in selection science. Schmidt and Hunter's landmark meta-analysis, covering 85 years of hiring research, found structured interviews to be among the strongest predictors of job performance available, well ahead of unstructured interviews. Later work on mechanical versus judgment-based scoring points the same way: applying a consistent scoring formula to responses produces more accurate and fairer outcomes than in-the-moment human judgment.

The pattern is intuitive once you see it. Structure removes noise. What remains in the score is the signal: how well the candidate's answers match the demands of the job.

Yes, and it is worth being precise about where. A late-stage conversation about team fit, expectations, or selling the role to a strong candidate benefits from being human and unscripted. The mistake is not that unstructured conversations exist; it is using them as the screening instrument that decides who advances. Screen with structure, then converse freely with the shortlist.

If the evidence is this one-sided, why does unstructured screening survive? Because structure is expensive to run by hand. A rigorous structured interview takes 30 to 45 minutes per candidate plus scoring time, which no team can afford across hundreds of applicants per role. So the structured format gets pushed late into the funnel, and the top of the funnel falls back to CV skims and quick calls: the least predictive methods applied to the largest number of people.

This is where AI changes the equation. Structured AI interviews deliver the same competency-based interview to every applicant, scored by deterministic models: same input, same output, full explainability. Screening accuracy runs 2–5x higher than traditional methods, and every score remains auditable.

OKQ8, the Scandinavian fuel and convenience retailer handling 23,000+ applications a year, reaches an average time-to-shortlist of 57 minutes with a 90% completion rate. As Alma Lindqvist of OKQ8's People, Process & Development team puts it: "Candidates are assessed against the same structured criteria, which reduces bias and improves the quality of the shortlist by capturing practical details often missing from a CV."

Want to see how structured screening works for your roles? Book a demo with Hubert.

Schmidt & Hunter, 1998. The Validity and Utility of Selection Methods in Personnel Psychology: Practical and Theoretical Implications of 85 Years of Research Findings. Psychological Bulletin, Vol. 124, No. 2, 262–274.

Kuncel et al., 2013. Mechanical Versus Clinical Data Combination in Selection and Admissions Decisions: A Meta-Analysis. Journal of Applied Psychology, Vol. 98, No. 6, 1060–1072.