Latest / Hiring & Firing with Fexingo: Recruitment, Performance, and Workforce Decisions / The One Interview Question That Exposes Overconfident Hires
Transcript
- Lucas: Luna, I want to start with a number that made me genuinely reconsider a question I've asked in almost every interview I've ever conducted. Luna: Okay, you've got my attention. Which question? Lucas: The 'biggest weakness' question. You know, 'What's your biggest weakness and what are you doing about it?' Classic, right? Almost everyone uses some version of it. Luna: Right, and most candidates have a scripted answer. 'I work too hard,' or 'I'm too detail-oriented.' The safe stuff. Lucas: Exactly. And a 2025 study from Harvard Business School suggests that those polished, self-improvement-type answers are actually a red flag. The study tracked about 2,000 new hires across a dozen companies and found that people who gave a generic, positive-sounding weakness — like 'I sometimes struggle with delegation because I care so much' — had a 34 percent higher turnover rate in their first eighteen months. Luna: Thirty-four percent? That's not marginal. That's a massive signal. Lucas: It is. The researchers argue that the question, as typically asked, doesn't measure self-awareness. It measures a candidate's ability to manufacture a humblebrag. And the people who are best at that are often the most overconfident — they know how to spin a flaw into a strength. But overconfidence, it turns out, correlates with lower learning agility and a higher likelihood of quitting when things get hard. Luna: So the smooth answer isn't just useless — it's actively misleading. Lucas: Worse than useless. Because if you hire the person who delivered the perfect weakness answer, you might be getting someone who lacks the humility to improve. The study found that candidates who gave genuinely specific, sometimes uncomfortable answers — like 'I'm not great at public speaking, and I've been taking a course to get better' — had significantly lower turnover and faster ramp-up times. Luna: Faster ramp-up? That's counterintuitive. You'd think the self-aware candidate who admits a real weakness would take longer to get up to speed. Lucas: Right, but the data says the opposite. The researchers think it's because those candidates are already in a learning mindset. They know their gaps, so they actively seek feedback and training. The overconfident candidates — the ones who gave the polished answer — tend to assume they've got it figured out, so they don't ask for help. And when they inevitably hit a wall, they're more likely to blame the company and leave. Luna: That matches something I've seen. I remember interviewing a product manager candidate who said her biggest weakness was that she sometimes over-analyzed data and delayed decisions. She was very specific — she gave an example from her previous job, and she told me she was working with a coach on time-boxing analysis. I hired her, and she was one of the best hires I ever made. Lucas: She told you she was working with a coach. That's the key detail. The best answers aren't just honest — they show active remediation. The study coded answers into three buckets: 'humblebrag', 'vague but honest', and 'specific with action plan'. The specific with action plan group had the lowest turnover by a wide margin. Luna: So if I'm an interviewer, how do I get more of those honest answers? Because candidates have been trained to give the safe version. Lucas: The researchers suggest reframing the question entirely. Instead of 'What's your biggest weakness?' they recommend asking something like: 'Tell me about a time in the past year when you realized you were wrong about something important at work. What did you do next?' Luna: Oh, that's a much harder question to fake. Because you need a real example. Lucas: Exactly. It shifts the focus from a static trait — 'I am bad at X' — to a dynamic behavior: 'I was wrong, and here's how I responded.' Overconfident people struggle with that question because they don't like admitting they were wrong. The humble but competent candidates have a story ready. Luna: Does this connect to that concept of 'learning agility' I've been hearing about? The ability to learn from experience and apply it to new situations? Lucas: Directly. Learning agility is actually the meta-skill the study was trying to measure. The researchers argue that the 'biggest weakness' question, when asked in the traditional way, is a poor proxy for learning agility. But the 'tell me about a time you were wrong' question is a much better one. In fact, one of the companies in the study — a large Fortune 500 tech firm — replaced the weakness question with a 'learning agility' prompt entirely. They ask: 'Describe a situation where you had to learn a new skill quickly to solve a problem. How did you approach it?' Luna: And what happened to their hiring outcomes? Lucas: After two years, they saw a 20 percent reduction in early attrition — people leaving within the first year. And their hiring managers reported that the new question produced much richer, more honest conversations. Candidates stopped giving scripted answers and started telling real stories. Luna: Twenty percent drop in attrition is huge. That's worth a lot of money in recruiting costs alone. Lucas: Absolutely. The study estimates that for a company hiring 500 people a year at an average cost per hire of $4,000, that 20 percent reduction in turnover saves about $400,000 annually just in replacement costs. Not to mention the productivity gains from having more engaged, learning-oriented employees. Luna: So the practical takeaway for someone listening who does hiring: don't ask the weakness question in the old way. Replace it with a question that forces a real story about learning or being wrong. Lucas: That's the core recommendation. But the researchers also caution against overcorrecting. They found that candidates who gave a 'vague but honest' answer — like 'I'm not great at public speaking' without any example or action plan — actually had slightly higher turnover than the humblebrag group. So the key is specificity plus remediation. Luna: Interesting. So if a candidate says 'I'm not great at public speaking' and leaves it there, that's actually worse than someone who says 'I work too hard'? Lucas: In the data, yes. The researchers think it's because a vague honest answer signals that the candidate is aware of a weakness but hasn't taken steps to address it. That can indicate a lack of initiative or a fixed mindset. The humblebrag, while dishonest, at least shows an attempt to frame things positively. But the best scenario is specific plus active improvement. Luna: So as an interviewer, you're listening for that pattern: a concrete example, a specific action, and ideally a measurable outcome. Lucas: Exactly. And there's another angle here: the study also looked at how interviewers rated candidates on confidence. They found that interviewers consistently rated overconfident candidates higher on 'leadership potential' and 'cultural fit', even though those same candidates had worse performance reviews later. So there's a bias we need to be aware of — we like confident people, and we mistake confidence for competence. Luna: That's the classic 'confidence heuristic' in action. We assume confident people are more capable, even when the data says otherwise. Lucas: Right. And the antidote is to structure interviews around specific behavioral examples rather than general self-assessments. The 'tell me about a time you were wrong' question is one example. Another one that performed well in the study was: 'Tell me about a project where you failed to meet a goal. How did you handle it?' Luna: Both of those questions probe for humility and learning orientation. They're harder to bluff through. Lucas: Much harder. And interestingly, the study found that candidates who gave strong answers to those questions also tended to have higher scores on a standard measure of learning agility — the 'l a i' or Learning Agility Inventory. So the interview itself becomes a valid assessment tool, not just a conversation. Luna: Let's talk about the practical implementation. If I'm a hiring manager and I want to adopt this, how do I train my team to evaluate the answers consistently? Lucas: The companies in the study used a simple rubric. They scored answers on three dimensions: specificity — did the candidate name a concrete situation?; action — did they describe what they did to learn or improve?; and outcome — did they mention the result, even if it was a lesson learned? Each dimension got a score from one to three, and the total predicted turnover with surprising accuracy. Luna: So it's not just about changing the question — it's about systematizing how you evaluate the responses. Lucas: Exactly. And that's what the best performing companies in the study did. They didn't just swap one question for another. They trained interviewers to listen for those three dimensions and to avoid the halo effect of a confident delivery. Luna: One concern I have: Could candidates start gaming this new question too? If it becomes widely known, won't people just prepare a story about being wrong? Lucas: It's possible, but the researchers argue it's harder to fake a genuine story of learning. Because if you're not actually a learning-oriented person, you'll struggle to come up with a specific, detailed example that rings true. The humblebrag weakness answer is easy to script — everyone knows the template. But a story about being wrong requires a level of self-reflection that overconfident people genuinely lack. Luna: So the new question acts as a sort of honesty filter. If you're not the real deal, you can't produce a convincing answer on the spot. Lucas: That's the theory, and the data supports it. The study also did a follow-up experiment where they told some candidates the new question in advance. Even with preparation, the overconfident candidates still gave less specific, less action-oriented answers than the humble ones. So the question seems to be resistant to coaching. Luna: That's really powerful. It means you can actually design interviews to surface the traits you want, rather than just testing interview skills. Lucas: And that's the bigger takeaway from this research. The traditional interview is a terrible predictor of job performance — most studies put its validity at around 0.2 or 0.25 correlation. But structured interviews focused on learning agility and humility can push that to 0.5 or higher. That's a huge improvement. Luna: I want to push back slightly though. Isn't there a risk that we over-index on humility and miss out on people who are genuinely confident and competent? Some roles really do require bold decision-making. Lucas: That's a fair point. The study didn't suggest that confidence is bad — it's overconfidence that's the problem. And the researchers distinguished between the two by looking at calibration. They asked candidates to predict how they would perform on a job-relevant test, and then compared that prediction to their actual score. Overconfident candidates overestimated their performance by an average of 30 percent. Well-calibrated candidates were much closer. Luna: So the key is to hire people who are confident but accurate in their self-assessment. Not people who are modest but unsure, and not people who are arrogant but deluded. Lucas: Exactly. The sweet spot is what psychologists call 'accurate self-awareness'. And the interview questions we've been talking about — the 'time you were wrong' or 'project you failed at' — are designed to probe for that. A well-calibrated person can tell you about a failure without being defensive, and they can describe what they learned without making excuses. Luna: That reminds me of a concept from the book 'The No Asshole Rule' — the idea that a single toxic hire can cost more than multiple good hires bring in. Overconfidence, if it leads to poor learning and eventual turnover, could be a form of toxic behavior. Lucas: That's a good connection. And the cost is real. The study calculated that for a mid-level hire with a salary of $80,000, the cost of a mis-hire — including recruiting, training, and lost productivity — is about $240,000. So improving your interview process to avoid overconfident hires has a direct financial impact. Luna: We should talk about one more thing before we wrap: how does this apply to remote hiring? Because a lot of interviews are now conducted over video, and it's harder to read subtle cues. Lucas: Great point. The study included both in-person and video interviews, and interestingly, the video interviews actually produced more honest answers to the 'time you were wrong' question. The researchers speculate that the slight distance of a screen reduces the pressure to perform, so candidates are more natural. But the downside is that it's harder for interviewers to detect nervousness or discomfort, which can be a signal of a faked answer. Luna: So the rubric becomes even more important in a remote setting. You need to rely on the content of the answer, not the body language. Lucas: Exactly. And that's why the three-dimension scoring system — specificity, action, outcome — works well regardless of medium. It focuses on what the candidate says, not how they say it. Luna: Alright, so let's land this. If a listener takes one thing away from this episode, what should it be? Lucas: I'd say: stop asking the 'biggest weakness' question in its old form. Replace it with a question that asks for a specific story about learning from a mistake or being wrong. Then score the answer on specificity, action, and outcome. That simple change could meaningfully reduce turnover and improve the quality of your hires. Luna: And if today's conversation helped you think differently about your interview process, there's a small way you can support the show. If it was worth the price of a coffee to you, you can find us at buy me a coffee dot com slash fexingo. It's that simple, and it helps us keep these episodes ad-free. Lucas: We appreciate that. And honestly, even just sharing an episode with a colleague who does hiring helps us reach more people. So thank you for listening, and for being part of this community. Luna: Alright, let's get back to the research. Lucas, you mentioned that the study also looked at how long the benefits of this hiring change lasted. Did they do a long-term follow-up? Lucas: They did — an eighteen-month follow-up. And the results held. The hires who scored high on the learning-agility interview question were not only less likely to quit, but they also received higher performance ratings from their managers. So the effect wasn't just about retention — it was about actual job performance. Luna: That's the kind of data that makes you want to change your interview process tomorrow. Lucas: Absolutely. And the best part is that it doesn't require any fancy tools or expensive consultants. Just a change in the questions you ask and a simple rubric to evaluate them. Luna: We'll put links to the study and the rubric in the show notes. For now, thanks for listening to Hiring & Firing with Fexingo. I'm Luna. Lucas: And I'm Lucas. See you next time.