Interview tips for hiring managers: run better interviews

Start with the job, not the conversation

Most interview tips for hiring managers start in the wrong place. They start with clever questions, confident body language, or the delicate theatre of making a candidate feel relaxed while trying to judge them. Those things matter, but they are not the point. The point of an interview is candidate assessment: collecting comparable evidence about whether someone can do the work.

The useful question is not "Do I like this person?" or even "Did that answer sound impressive?" It is "What capability did this answer help me assess, and what evidence did I hear?" The U.S. Office of Personnel Management defines a structured interview as a way to measure job-related competencies by asking about past behaviour or hypothetical job situations. That is a better starting point than the charming-but-vague coffee chat.

Levashina et al. (2013) summarise the same idea from the research side: structured interviews use predetermined rules for questions, observations and evaluation. McDaniel et al. (1994), in a meta-analysis of 245 coefficients from 86,311 people, found that structured interviews had higher validity than unstructured interviews. In plain English, managers make better predictions when the interview has a shape.

Before you write a single interview question, write down the capabilities the role actually requires. Three to six is usually enough. For each capability, decide what good evidence would look like, what weak evidence would look like and what would be irrelevant noise. Now the interview has a job to do. The conversation is just the delivery mechanism.

Build a structured interview before candidates arrive

A structured interview is not a script for managers who cannot think. It is a guardrail for managers who can think, but who would rather think about the evidence than improvise a fresh assessment method every 45 minutes.

Campion, Palmer and Campion (1997) identified 15 components of interview structure, split between the content of the interview and the way answers are evaluated. A hiring manager does not need to recite all 15 before breakfast. The practical core is simpler: base questions on the job, ask candidates the same core questions, limit random probing, take detailed notes, rate answers against anchored scales and keep discussion until after individual ratings are complete.

The OPM guidance makes the same point operationally: structured interviews ask candidates the same predetermined questions and evaluate responses using the same rating standards. That consistency is not bureaucracy for its own sake. It is what lets you compare Candidate A's evidence with Candidate B's evidence without pretending that two completely different conversations tested the same thing.

A simple interview guide for hiring managers should include five parts:

• The capabilities being assessed.
• The core question for each capability.
• Permitted follow-up prompts, used consistently.
• A rating scale with examples of weak, acceptable and strong evidence.
• A note field that asks for what the candidate said or did, not how the interviewer felt.

This is where hiring manager interview training often goes thin. It teaches people to ask questions, but not to build a repeatable assessment. The better test is whether another manager could run the same interview tomorrow and produce evidence that belongs in the same decision.

Ask fewer, better questions

Interview questions for hiring managers should earn their place. A question earns its place when it tests a named capability, produces evidence you can score and gives every candidate a fair chance to show the same thing. That rules out many favourites.

"What is your biggest weakness?" mostly tests interview coaching. "Where do you see yourself in five years?" tests tolerance for corporate fiction. "Sell me this pen" tests whether someone has watched enough online sales clips to perform mild hostage theatre. These questions may be entertaining, but entertainment is not selection.

Better questions usually fall into three families. Past-behaviour questions ask for evidence of what the person has actually done: "Tell me about a time you had to recover a delayed project. What did you do first, and what changed as a result?" Situational questions ask how the person would handle a realistic future problem: "A key customer is angry, the product team disagrees with the fix and you have two hours before a renewal call. What do you do?" Role-specific questions test the work directly: "Talk me through how you would diagnose this pipeline drop."

OPM's selecting-official guidance points managers towards job-related questions, and that is the discipline. Ask about specific responsibilities, relevant decisions, trade-offs and evidence. Avoid questions about protected characteristics, personal life, health or anything else that has nothing to do with the role.

The best interview tips from hiring managers tend to sound almost disappointingly plain: ask one thing at a time, listen more than you speak, interrupt less, and do not rescue a vague answer too quickly. If the candidate gives a general claim, ask for the example. If they give the example, ask for their role in it. If they give their role, ask what changed because of their action. That is enough. You are not trying to win the conversation. You are trying to get a scoreable answer.

Use an interview scorecard while the evidence is fresh

An interview scorecard is where good intentions become a decision tool. Without one, even a structured interview can decay into memory, mood and whoever speaks first in the debrief. With one, the hiring manager has to connect each rating to evidence while the evidence is still warm.

A useful scorecard has five columns: capability, question, evidence, rating anchor and recommendation. The evidence column matters most. A rating without evidence is just a vibe in office clothing. "Strong communicator" is a conclusion. "Explained a technical outage to a non-technical customer, named the trade-offs, and reduced escalation time from three days to one" is evidence.

Metaview's scorecard guidance says scorecards guide interviews by defining criteria and capturing specific responses. Greenhouse's product documentation shows the same habit in workflow form: after the interview, the interviewer can fill out their scorecard, rate attributes and submit an overall recommendation. The technology is not the main point. The sequence is.

Score as soon as possible, ideally during the interview for individual answers and immediately afterwards for the overall recommendation. If you wait until the debrief, the group will start editing everyone's memory. Someone will say "I loved her energy", someone else will say "I had concerns about polish", and suddenly the candidate assessment is being conducted by adjectives wearing lanyards.

Campion, Palmer and Campion (1997) include rating each answer, using anchored rating scales and taking detailed notes among the components of structure. The McGill primer makes the practical case: structure reduces the space for halo effects, similar-to-me bias and confirmation bias. None of this requires a complicated form. It requires a form that makes unsupported judgement slightly embarrassing.

Use the scorecard to separate three things managers often mush together: what the candidate said, what you inferred from it and what score you gave. Keep those separate and the debrief becomes sharper. People can disagree about the meaning of the evidence instead of trading impressions like weather reports.

Control the bias you can actually control

No hiring manager can remove bias by promising to be sensible. Sensible people are precisely the ones most at risk of trusting their own stories. Dana, Dawes and Peterson (2012) show why: in unstructured interviews, interviewers can form confident impressions from weak information. They called this sensemaking. The brain does not like loose ends, so it tidies them.

The same work describes dilution: extra non-diagnostic information can weaken the use of better evidence. Kausel, Culbertson and Madrid (2016) found a related pattern in personnel selection. Decision makers given unstructured interview information became more overconfident than those using standardised test information alone. The problem is not that managers are foolish. The problem is that unstructured conversations feel informative even when they are not.

That is why the practical answer is procedural. Ask candidates the same core questions. Give them broadly the same time. Take notes before discussing candidates. Submit individual scores before the group debrief. Do not let interviewers compare candidates between interviews, because the contrast effect is a quiet vandal. Define red flags before you meet candidates, not afterwards when a candidate has made you uneasy and you are shopping for a respectable reason.

Criteria Corp puts the broader problem bluntly: interviews are inherently subjective. Structure does not make them perfectly objective. It narrows the places subjectivity can hide, and it makes the remaining judgement easier to inspect.

The goal is not to become colder. In fact, candidates often experience a well-run structured interview as more respectful. They are asked relevant questions, given time to answer and assessed against criteria they can understand. The warmth is still there. It is just no longer driving the hiring decision from the passenger seat.

Install the process so managers actually use it

The awkward truth about hiring manager interview training is that most of it dies as a slide deck. People nod, agree that bias is bad, ask two better questions for a fortnight and then drift back to the old pattern when the next urgent vacancy appears. A process that depends on everyone remembering a workshop is not a process. It is a hope with catering.

This is where the work shifts from advice to installation. If managers are going to run better interviews consistently, they need the interview flow, capability model, question bank, scorecard and decision mechanics built into the way the company hires. That is the difference between "we believe in structured hiring" and "every interviewer knows exactly what they are testing on Thursday at 10:30".

HireSchool exists for that implementation problem. The Structured Hiring Method is a self-guided digital programme for small businesses and scale-ups that want consistent, evidence-based hiring without bringing in consultants for every role. It teaches teams how to define capabilities, design structured interviews, use scorecards and make decisions from evidence rather than debrief theatre. The format is deliberately practical: video content plus a learning management system, so managers can learn the method and apply it inside the actual hiring process.

That matters because the research is only useful if it survives contact with calendars, inboxes and managers who quite reasonably have other jobs. Levashina et al. (2013) and Campion, Palmer and Campion (1997) give the evidence base. A business still has to translate that into who writes the questions, who scores what, when feedback is submitted and how disagreements are resolved. Those operating details are where good hiring habits either become normal or quietly disappear.

The Structured Hiring Method is not a magic question bank, and it is not a promise that every hire will be perfect. No serious selection method can promise that. It is a way to codify the parts of hiring that should not change candidate by candidate: the capabilities being assessed, the evidence required, the scoring rules and the decision sequence. Managers still use judgement. They just use it inside a system that asks them to show their working.

For a small company, that is often the missing piece. You do not need an enterprise talent function to run a better interview. You need a shared method that a founder, hiring manager and interviewer can all follow without turning the process into HR origami. If that is the problem in front of you, the next step is to explore the Structured Hiring Method programme and decide whether the current interview process is good enough to keep defending.

Decide after all the interviews, not during the chat

The most expensive interview mistake is deciding early and collecting reasons later. It feels efficient. It is usually just confirmation bias with a calendar invite. Once a manager has privately decided "yes" or "no", the rest of the interview can become evidence decoration.

Use a decision sequence instead. First, every interviewer submits their scorecard independently. Second, the hiring manager compares ratings against the pre-agreed capabilities. Third, the group discusses the biggest evidence gaps and disagreements. Fourth, the decision is made against the role criteria, not against the candidate who happened to be most fluent, familiar or pleasingly low-maintenance.

This is where the common "red flag" question needs care. A red flag is not "something I did not like". It is a pre-defined risk that matters for performance in the role: repeated inability to explain ownership, no evidence of a required capability, or a pattern of blaming others without learning. If you cannot connect the concern to the role and to evidence, do not promote it to red-flag status. It can remain a note, quietly minding its own business.

Campion, Palmer and Campion (1997) recommend detailed notes, anchored ratings and limiting discussion between interviews because these habits protect the decision from drift. Kausel, Culbertson and Madrid (2016) give the warning label: unstructured interview information can inflate confidence without improving accuracy. The debrief should therefore test confidence, not perform it.

Better interviews are not more theatrical. They are more comparable. They ask candidates to show evidence against the same capabilities, help managers listen for the same signals and make the final decision after the evidence has been recorded. That is the quiet discipline behind most useful interview tips for hiring managers: decide what you are testing, test it consistently, score it promptly and only then let the room have an opinion.