By Sean Perry, Chief Information Officer at Kelly
Most interviews tell you how well someone interviews. They don't tell you how well someone performs on the job.
After 20 years in recruiting—the last seven at Kelly, where I manage a national team of 15 recruiters across large accounts—I can say with confidence that the single biggest shift I've made in my approach to hiring is moving from hypothetical questions to behavioral ones.
A behavioral interview question is an open-ended question that prompts a concrete response based on past experience. Instead of asking a candidate what they would do in a given situation, you're asking them what they did. Research indicates behavioral interview questions are 55% predictive of future on-the-job behavior, compared to just 10% for traditional interviewing.
While traditional interviews give you opinion-based answers (a candidate tells you what they think you want to hear) behavioral questions require them to pull from memory, to give you a specific example, to describe what actually happened. That's where you learn who someone really is at work.
One of the most common mistakes I see hiring managers make is pulling a list of behavioral interview questions off the internet and using the same set for every role. If you're hiring a project manager and you ask them to describe a time they had to learn a new technical skill, that's fine, but it's not getting at what actually makes or breaks that role. You'd learn more by asking how they kept a project on track when the scope changed midway through.
Instead, start with the job description. Identify the top three to five responsibilities for the role, then figure out the soft skills required for each. From there, you develop questions that prompt responses tied directly to those skills.
Target willingness to learn and adaptability. Candidates at this level may not have deep professional experience to draw from, so focus on situations that reveal how they respond to new challenges, take direction, and handle pressure for the first time. Something like, "Tell me about a time you had to meet a deadline you weren't sure you could hit" works well here.
Lean toward problem-solving under pressure. The best questions for technical candidates surface how they work through complex problems when the path forward isn't obvious—how they diagnose, troubleshoot, and iterate. Ask about situations where they had to solve a problem with limited information, or where their first approach didn't work and they had to regroup.
Focus on communication and conflict resolution. These roles require people who can manage expectations, de-escalate tension, and maintain the relationship even when things go sideways. Questions about handling upset customers, navigating miscommunication, or delivering difficult news will be revealing.
Focus on decision-making, strategic thinking, and measurable impact. At this level, you're looking for evidence that someone can make hard calls, manage competing priorities across teams, and drive results that go beyond their individual output. A question like, "Describe a time you had to make a difficult decision as a leader—how did you handle it, and what happened?" gets at competencies that matter for that level.
Here's a starting point, organized by the competency each question targets. Pick the ones that match the responsibilities of the role you're hiring for, and build from there.
A single bad hire can cost an average of $17,000 in hiring costs, training, and lost productivity. Seventy-five percent of employers have admitted to hiring the wrong person for a role at some point, and more than half pointed to a negative attitude or behavioral issue as the primary reason the hire didn't work out. These are the kinds of red flags behavioral questions are designed to surface.
Of all the behavioral interview questions I use, one consistently surfaces the most valuable information: Tell me about a time you made a mistake at work. How did you resolve it, and what did you learn?
You're evaluating someone's ability to handle a challenging situation, take accountability, and benefit from what happened. Candidates who can speak openly about a real mistake tend to be the ones who grow, adapt, and earn trust from their teams.
The biggest mistake interviewers make with behavioral questions is they don't ask follow-up questions. They have their standardized list in front of them, they ask the question, they get an answer, and they move on. The interview ends up feeling robotic instead of conversational, and the interviewer walks away with maybe half the information they could have gotten.
Behavioral questions are not an endpoint. Think of them as a guide. When a candidate gives you their initial answer, that's usually about 75% of the story. It's the interviewer's job to dig into the remaining details. Ask things like, "What was the outcome?" or "What would you do differently?" or "What specifically was your role in that?"
If a candidate keeps saying "we did this" and "we achieved that," it might mean they're a great collaborator. But it might also mean they're telling a story that isn't really theirs. When I hear a lot of "we," I redirect: "I understand what the team accomplished: can you walk me through your specific contribution?"
Beyond the follow-up, pay attention to red flags in how candidates respond — a consistently negative tone when describing past roles, body language that doesn't match the story they're telling, or an inability to provide any specific examples. These signals are just as telling as the content of the answer itself.
On the flip side, some candidates are strong performers who struggle to articulate their experience in an interview setting. For those candidates, I use their resume as a reference point and tailor my follow-up questions around specific roles or projects they've listed. It gives them a concrete anchor, and it usually opens up the conversation.
Most people know STAR—Situation, Task, Action, Result—as a tool candidates use to structure their answers. But it's just as useful on the interviewer's side. When you're listening to a candidate's response, STAR gives you a way to assess whether you're actually getting a complete answer or just a partial one.
A strong response covers all four parts: the context of the situation, what the candidate was specifically responsible for, the actions they took, and what happened as a result. If any piece is missing, that's your cue to follow up. A candidate who describes a great outcome but skips over what they personally did hasn't given you enough to score on. Someone who walks you through their actions but never mentions the result may be avoiding a story that didn't end well.
This connects directly to the follow-up problem. When interviewers don't have a framework for what a complete answer sounds like, they're more likely to accept surface-level responses and move on to the next question. STAR gives you a simple checklist: Did I get the situation? Do I know what their role was? Do I know what they did? Do I know what happened? If any of those are missing, keep asking.
If I could recommend one immediate improvement for any employer looking to raise the quality of their interviews, it would be this: build a structured interview process. That means a standardized set of behavioral interview questions for each role, a consistent scoring system across all candidates, and clear job requirements that inform both the questions and the evaluation.
Unstructured interviews have a predictive validity of roughly 0.21, which is barely better than chance. Structured interviews using behavioral and situational questions perform significantly better. When every candidate for a role answers the same set of job-relevant questions and gets scored against the same criteria, you're comparing apples to apples instead of going on gut feel.
This kind of consistency also matters for fairness. When you standardize your process, you reduce the influence of unconscious bias. Each candidate gets an equal opportunity to demonstrate how their experience aligns with the job's requirements. Candidates who had a positive, in-depth interview experience were 50% more likely to stay with the company after three years. And 53% of U.S. candidates say being asked inappropriate or irrelevant questions would be enough for them to reject an offer. A structured behavioral interview helps you avoid both problems.
Yet most interviewers have never been trained on how to do this well. Only about 3% of U.S. hiring managers opt into formal interviewer training, and fewer than 10% of employers mandate it. Even a short session on active listening, the STAR framework (Situation, Task, Action, Result), and how to use a scoring rubric can make a noticeable difference in interview quality and hiring outcomes.
Behavioral interview questions work because they're grounded in evidence. They ask candidates to show, not tell. They give interviewers a clearer, more reliable picture of how someone actually operates: how they handle pressure, own their mistakes, work with others, and solve problems.
But the questions themselves are only as good as the process around them. Tailor them to the role. Train your interviewers. Score consistently. And always, always follow up. That's where the real insight lives.