Welcome to the show. I’m Claire Monroe with Edwin Carrington—and I want to start with something that’s… mildly uncomfortable. Most interviews? They’re basically confidence tests. Like—who sounds sharp, who tells a clean story, who makes you feel like, “yeah, this person gets it.” And somehow that feeling turns into a hiring decision… even though confidence and competence are not the same thing. They’re not even strongly correlated in many cases. What you’re describing is a very reliable pattern. Unstructured interviews tend to reward three things: charisma, similarity, and narrative fluency. We call it “gut feel” because that sounds respectable, but most of the time it’s just bias behaving politely. “Bias behaving politely” is… yeah, that’s going to stick. Because that’s confirmation bias, right? You like someone early—minute three, maybe—and then the rest of the interview turns into this quiet little mission to prove you were right in the first place. Exactly. The decision starts forming before the evaluation does. And from that point on, you’re not really assessing the candidate—you’re defending a conclusion you haven’t fully examined yet. Which is kind of unsettling, because everyone in that room would swear they’re being objective. So this is where structured interviews come in? Yes, and the purpose is very specific. Structure isn’t about making interviews rigid. It’s about making them comparable. Same questions, same criteria, same scoring logic. You decide what matters before the conversation begins instead of improvising your standards halfway through it. I think that’s where people get hung up. “Structured” sounds stiff, like you’re turning a human conversation into a checklist. But without it, you’re basically running three completely different interviews and pretending they’re comparable. That’s exactly what’s happening. If Candidate A is asked about conflict, Candidate B about strategy, and Candidate C about whatever came to mind after lunch, there is no shared basis for comparison. At that point, you’re not evaluating performance potential—you’re evaluating who gave the most compelling conversation. Which is a completely different skill. And not the job. Correct. The job is prediction. Can this person perform in this role, under these conditions? And the research trend is quite consistent here—structured interviews reduce noise, improve fairness, and are simply better predictors of performance than informal ones. That phrase—predicting performance—feels like the part most teams skip over. Because the default question is still, “did I like this person?” not “did I gather useful evidence?” And here’s the practical challenge: most interviewers believe they’re already good at interviewing. That’s the hardest audience to retrain—the confident amateur. If you suggest their instincts might be misleading them, it feels personal, even though it’s really just a cognitive limitation we all share. So it’s less about fixing the process and more about getting people to admit the process might be flawed in the first place. Yes, but once that shift happens, interviews usually become easier. Less wandering, fewer repeated questions, less debate based on vague impressions. A solid framework—something OAD has explored in depth—actually reduces effort because interviewers know what they’re looking for and how to evaluate it. So instead of trying to become some kind of human lie detector, you just build a system that doesn’t rely on that. Exactly. Better system, better evidence, better decisions. Okay, let’s make that concrete. OAD uses three core attributes: role-related knowledge, problem solving, and leadership. Sounds simple, but I feel like people interpret those very differently. They do. Role-related knowledge is the practical understanding of the work itself. Problem solving is how someone navigates ambiguity, constraints, and trade-offs. And leadership is where most teams get imprecise. It’s not limited to management roles. Right, because leadership for an individual contributor might look like influencing decisions or taking ownership, not managing people. Exactly. An individual contributor shows leadership through judgment, initiative, and influence. A manager shows it through coaching, delegation, and creating clarity. If you don’t define that properly, you risk filtering out strong candidates for the wrong reasons. Like expecting a senior engineer to sound like a VP, and then deciding they’re “not leadership material.” Precisely. And that’s where question design becomes critical. Behavioral questions ask for real evidence—what someone has actually done. Hypothetical questions explore judgment—what they would do. Both are useful, but neither works if you accept the first polished answer at face value. Because people rehearse. “I collaborate, I communicate, I prioritize…” That’s not insight, that’s just… corporate wallpaper. Exactly. Which is why structured follow-ups matter. Not random pressure, but intentional probing. What options did you consider? What data did you use? What trade-offs did you face? Now you’re evaluating thinking, not just storytelling. So the real signal shows up after the first answer, not in it. That’s where most of the useful information is, yes. Let me try to sharpen that. A weak question is “Are you good under pressure?” because the answer is obviously yes. A better one is “Tell me about a time you worked under pressure.” And then you go further—what happened, what did you choose not to do, what would you change now. That’s a strong direction. I’d make it even more role-specific. For a manager: tell me about a time your team was at risk of missing a critical deadline and how you responded. For an individual contributor: tell me about a time you influenced an outcome without formal authority. Specificity leads to better evidence. And probably a better experience for the candidate too. It feels more relevant, less random. Exactly. Candidates can tell when a process is structured. It feels more coherent, more deliberate. And the final layer is discipline after the interview—shared rubrics, independent scoring, and calibration before discussion. Without that, the loudest voice in the room still shapes the outcome. So the takeaway isn’t “trust your instincts more.” It’s define what matters, standardize the questions, score consistently, and stop equating a good conversation with a good hire. That’s a more accurate framing. And if someone’s listening to this thinking, “this sounds great, but implementing it feels like a lot…” That’s where tools can help. If you’re looking to operationalize structured hiring—things like consistent frameworks and behavioral assessments—OAD offers a practical starting point. So you’re not building the whole system from scratch and hoping people stick to it. Exactly. And if you’re wondering how to put this into practice, you can test OAD’s tools—like behavioral assessments—for free at o-a-d-dot-a-i. It’s a straightforward way to make your hiring process more consistent and more predictive. Which, at this point, feels like the real goal. Not just a good conversation—but a better decision. Thanks for listening to The Science of Leading.
Audio Coursesby Jellypod