Are You Hiring the Best Candidate or the Best Interviewer?

An interview is a short sample of how someone presents themselves under pressure. It is not a complete picture of how they will perform over the next year.

That does not make interviews a poor hiring tool. It means employers need to be careful about the conclusions they draw from them. A candidate may give a convincing account of past achievements, but the interviewer still has to establish what the person actually did, which skills they used and how closely that experience matches the vacancy.

The quality of that judgment depends partly on the questions being asked and the way the answers are assessed.

A 2025 meta-analysis published in the International Journal of Selection and Assessment examined how well interview assessments predicted different aspects of job performance.

The researchers reviewed 37 studies involving more than 30,000 participants. They found that interview ratings were associated with later task performance and contextual performance, which includes behaviour that helps the wider organisation function, such as supporting colleagues.

The study also found that interviews did not always distinguish cleanly between these different performance dimensions. In other words, an interviewer assessing one quality could also be picking up on another.

The authors identified structured scoring as particularly useful when assessing contextual performance. They also noted that more research is needed to establish how specific interview design choices affect results.

The findings support a measured conclusion: interviews can provide useful evidence, but employers need to be precise about the qualities they are trying to assess.

Interview questions should come from the responsibilities listed in the job description, not from a standard list used for every vacancy.

For a remote operations role, an employer might ask how the candidate handled a missed deadline when several people depended on the same deliverable. For a customer support position, the question could focus on resolving a complaint when the customer was unhappy with the available options.

These questions give candidates an opportunity to describe relevant experience. Follow-up questions can establish what they personally contributed, what decisions they made, and what happened afterwards.

Questions such as “Tell me about yourself” or “What is your greatest weakness?” may help start a conversation, but they are unlikely to provide enough evidence on their own to compare candidates for a particular job.

The interviewer should know what each question is intended to reveal. If the answer will not help the hiring team assess a necessary skill, it may not deserve time in the interview.

When every applicant is asked a different set of questions, it becomes difficult to know whether one performed better or simply had a more favourable conversation.

Structured interviews address this by giving candidates the same core questions and using agreed criteria to assess their answers. Interviewers can still ask follow-up questions, but the main areas of assessment remain consistent.

Consider a team hiring for a role that requires careful decision-making. One candidate might be asked about a difficult decision they made, while another is asked how they handle disagreements. Both answers may be interesting, but they do not provide directly comparable evidence of the same skill.

A common question, followed by a consistent set of assessment criteria, makes the comparison more useful.

This does not mean every interview must sound scripted. Interviewers can have a natural conversation while ensuring that each candidate has a chance to address the same essential requirements.

Hiring discussions often include comments such as “I liked them,” “They seemed confident” or “I am not sure they would fit in.”

Those impressions may influence a decision, but they do not explain whether a candidate can do the job. Confidence, for example, is not the same as competence. A quiet candidate may have strong technical knowledge, while a fluent speaker may struggle to provide specific evidence for their claims.

Interviewers should record examples from the answers and assess them against the role's requirements. If communication is essential, the criteria might cover clarity and the ability to explain complex information. If the role involves problem-solving, the interviewer can look for evidence that the candidate identified the issue, considered options, and justified a decision.

This gives the hiring team something concrete to discuss after the interview. It also makes it easier to explain why one candidate scored more highly than another.

When several people interview the same applicant, they may influence one another before they have formed their own judgments.

A senior manager's opinion can carry particular weight. Once that person expresses enthusiasm or concern, other interviewers may become less willing to disagree, even if they noticed something different.

One practical safeguard is to have each interviewer complete their notes and ratings before the group discusses the candidate. The team can then compare the evidence and identify genuine differences in assessment.

Disagreement is not necessarily a sign that the process has failed. It can reveal that interviewers are using different standards or interpreting an answer in different ways.

The important part is to resolve those differences by returning to the candidate's answers and the agreed criteria, rather than allowing the most confident voice in the room to settle the decision.

Some abilities are difficult to establish through conversation alone.

A candidate for a writing role can describe their editing experience, but a short editing exercise gives the employer a chance to examine the work itself. An analyst can explain their approach to data, then demonstrate it on a small, relevant example.

The task should be proportionate to the position and should not become an unpaid project that takes hours to complete. Employers should also use consistent criteria to evaluate the results.

Practical exercises are not necessary for every role. They are most useful when they answer a specific question that the interview cannot settle.

The same care applies to AI tools. If the job requires candidates to use AI, the employer should decide whether the task should allow it. If the aim is to assess unaided ability, the rules should be stated clearly in advance.

AI can also conduct the first interview, but employers need to understand what the technology can and cannot assess. Our guide, Can AI Really Handle Job Interviews?, looks at where automated interviews can help and why human judgment still matters when evaluating candidates.

Employers can learn from their own recruitment decisions by reviewing interview scores alongside later job performance.

That review should be handled carefully. A new employee's results may depend on training, workload, management and other conditions, not just their interview performance. Even so, patterns can be useful.

If candidates who receive high scores regularly struggle with a particular responsibility, the interview may not be testing that skill effectively. If employees who perform well were repeatedly rated poorly during recruitment, the team should examine what its interviewers are rewarding.

The 2025 meta-analysis offers a reason to be specific about interview criteria rather than treating a general impression as a reliable prediction of every aspect of performance.

A sound interview process gives the hiring team a clear record of what it assessed and why a candidate was selected.

That record should include the core questions, the criteria used to score answers and the evidence supporting the final decision. It need not be complicated. A short scorecard can be enough if it reflects the actual requirements of the role.

The aim is not to remove human judgment from hiring. It is to make sure that judgment is based on relevant evidence and applied consistently.