A structured interview asks every candidate for a role the same job-related questions, in the same order, and scores each answer against a predefined rating scale. An unstructured interview is a free-flowing conversation that varies from candidate to candidate and ends in an overall impression. Decades of research, including a major 2022 re-analysis of hiring methods, rank the structured interview as the most accurate single predictor of job performance, about twice as predictive as the unstructured version. Structured interviews are also fairer and easier to defend, yet most organizations still rely on looser formats.

Most job interviews still look like conversations. A hiring manager reviews a resume, asks whatever questions come to mind, follows the discussion where it leads and comes away with a feeling about the candidate. In a structured interview, every candidate for the role answers the same questions, drawn from an analysis of what the job requires, and interviewers rate each answer against the same defined standards.

In the most comprehensive recent review of hiring research, published in 2022 by Paul Sackett and colleagues, structured interviews had the highest predictive validity of any selection method studied, at .42, compared with .19 for unstructured interviews. "Structured interviewing is two times more predictive than unstructured interviewing, but it's always been pretty hard for companies to execute well," Josh Millet, Criteria's co-founder and CEO, said in 2025.

The gap matters more now that resumes carry less weight. With candidates using AI to polish applications, nearly a quarter of hiring professionals in Criteria's 2026-2027 Hiring Benchmark Report said they no longer trust the resume as a reliable signal, which puts more of the hiring decision on the interview itself.

What the research shows

Structure consistently improves accuracy

"One of the most consistent findings in the history of research on the employment interview is that structured interviews are much more reliable and valid than unstructured interviews," researchers Julia Levashina, Christopher Hartwell, Frederick Morgeson and Michael Campion wrote in a 2014 review in Personnel Psychology, noting that twelve separate meta-analyses had examined the question.

Earlier studies point the same way. A 1994 meta-analysis by Michael McDaniel and colleagues, covering more than 86,000 people, found validity of .44 for structured interviews and .33 for unstructured ones. Allen Huffcutt and Winfred Arthur found the same year that validity rose steadily with each level of structure, from .20 for the least structured interviews to .56 and .57 for the two most structured levels. The 2022 re-analysis by Sackett and colleagues lowered the estimates for nearly every hiring method but widened the structured interview's lead. "Structured interviews emerged as the predictor with the highest mean validity," the authors wrote in a 2023 follow-up.

Unstructured interviews create false confidence

Unstructured interviews persist partly because they feel informative. A 2013 study by Jason Dana, Robyn Dawes and Nathanial Peterson tested that feeling directly. Participants predicted students' grades using background information, and some also interviewed the students. The interviews made predictions less accurate, and participants formed confident impressions even when the students' answers had been randomized without the interviewers' knowledge. "Our simple recommendation for those who make screening decisions is not to use them," the authors concluded.

Daniel Kahneman described a similar experience in Thinking, Fast and Slow. As a young officer in the Israeli army in 1955, he was asked to fix an interview process whose predictions of soldiers' success were essentially worthless. He replaced it with ratings of six specific traits scored one at a time, and the new system predicted performance far better than the old conversations.

Structure also makes interviews fairer

Structured interviews narrow the room for bias by tying each rating to job-related evidence. Levashina and colleagues found minimal group differences in structured interview scores in more recent studies, and the U.S. Office of Personnel Management describes the structured interview as "among the most valid assessment tools available," with "relatively low adverse impact on minority groups compared to other assessment tools." A consistent, documented process is also far easier to defend if a hiring decision is ever challenged.

What makes an interview structured

In a widely cited 1997 review, Michael Campion, David Palmer and James Campion identified 15 components of interview structure, split between what interviewers ask and how they evaluate the answers. Most organizations do not need all 15 to see results. Huffcutt and Arthur's research suggests that validity levels off once questions and scoring are both standardized, so the elements below do most of the work.

Questions based on a job analysis

The questions come from a study of what the role actually requires, such as the skills, behaviors and situations that separate strong performers from weak ones. That link to the job gives the interview its predictive power and its legal footing.

The same questions for every candidate

Every candidate for the role hears the same questions in the same order. Follow-up prompts are limited and planned in advance, so no candidate gets an easier or harder interview because of who happens to be asking.

Behavioral and situational questions

Behavioral questions ask candidates to describe how they handled a real situation in the past, such as a time they resolved a conflict with a customer. Situational questions present a hypothetical scenario and ask what the candidate would do. Research by Paul Taylor and Bruce Small found that behavioral questions predict performance slightly better, and reviewers generally recommend using both.

Rating scales with defined anchors

Each answer is scored on its own, using a behaviorally anchored rating scale (BARS) that describes what a weak, average and strong answer looks like. Anchored scales improve both accuracy and agreement between interviewers, according to Levashina and colleagues.

Trained interviewers who rate independently

Interviewers learn the questions, the scale and common rating errors before they start. When several people interview a candidate, each records ratings before the group discusses, which keeps one strong opinion from swaying the rest. The overall score is then built from those individual ratings.

The U.S. Office of Personnel Management shares an example of a structured interview question and rating scale, and Google's re:Work program offers a public guide to building structured interviews.

Why structured interviews remain underused

Most hiring managers accept that structure helps, but far fewer practice it. In Criteria's 2022 Hiring Benchmark Report, 66% of hiring professionals said structured interviews lead to better hiring decisions, yet only 24% said they used tightly structured interviews. Fifty-seven percent described their interviews as loosely structured, and 17% said they relied on open conversation with minimal structure.

Effort and habit account for most of the gap. Writing job-specific questions and anchored rating scales takes time, interviewers need training to use them consistently, and many managers prefer the flexibility of a conversation. Some candidates also find tightly scripted interviews less personal, a reaction Levashina and colleagues documented in their review. Well-run structured interviews address that by leaving time at the end for the candidate's own questions and by training interviewers to deliver the questions naturally.

How technology is lowering the barrier

Newer tools take on much of the work that made structure hard to sustain. One-way video interviews let employers send the same structured questions to every applicant, and candidates record their answers on their own schedule. Live video platforms can build the rating scale into the interview itself so interviewers score answers as they go.

AI scoring adds another layer. Criteria's interviewing products generate behaviorally anchored rating scales for each question and score recorded interviews against them, using only the transcript of what the candidate said, which leaves appearance, accent and background out of the evaluation. When the company launched the scoring tool in 2025, Millet said it addressed a long-standing trade-off in the industry. "For a long time, video interviewing providers have been content to sacrifice quality for automation," he said.

Human judgment still makes the final call. "AI doesn't make hiring decisions at Criteria, people do," Chris Daden, the company's chief technology and AI officer, has written. Millet framed the broader goal when Criteria expanded its interviewing products in June 2026. "The future of hiring won't be built on resumes or gut feel," he said. "It will be built on trusted talent signals that are science-backed, explainable and guided by human judgement."

Structured interviews work best alongside other evidence. Pairing them with pre-employment assessments of cognitive ability, personality or job skills gives employers a more complete picture than any single method, and research going back to the 1990s supports that combination.

FAQ

What is the difference between a structured and an unstructured interview?

In a structured interview, every candidate for a role answers the same job-related questions in the same order, and interviewers score each answer against a predefined rating scale. An unstructured interview is an open conversation that varies by candidate and usually ends in an overall impression instead of a scored evaluation.

Are structured interviews more effective?

Yes. A 2022 re-analysis of hiring research by Paul Sackett and colleagues found that structured interviews had the highest predictive validity of any selection method, at .42, compared with .19 for unstructured interviews. Structured interviews are also more consistent across interviewers and tend to produce smaller differences between demographic groups.

What are examples of structured interview questions?

Structured interviews typically use behavioral questions, such as "Tell me about a time you had to resolve a disagreement with a coworker," and situational questions, such as "What would you do if a customer demanded a refund your policy doesn't allow?" Each question is tied to a skill the job requires and scored against a defined scale.

What is a behaviorally anchored rating scale?

A behaviorally anchored rating scale (BARS) describes what weak, acceptable and strong answers to an interview question look like, so every interviewer scores responses against the same standard. Research shows anchored scales improve both accuracy and agreement between interviewers.

Can AI conduct or score structured interviews?

AI tools can deliver structured questions, generate rating scales and score recorded answers against them. The more careful tools evaluate only the transcript of a candidate's answers and leave the hiring decision to people.