Latest / Hiring & Firing with Fexingo: Recruitment, Performance, and Workforce Decisions / How One Behavioral Question Predicts Retention Better Than Any Test
Transcript
- Lucas: So back in 2019, a Fortune 500 company — I'm not naming them because the research was conducted under a non-disclosure — ran a controlled trial across about fourteen hundred new hires in their call centers and field service roles. Luna: What were they testing for? Lucas: They wanted to know which single interview question, if any, could predict whether someone would stay at least two years. They tested cognitive ability tests, personality inventories, reference checks, and about a dozen different interview questions. And one question crushed everything else. Luna: What was the question? Lucas: Here's the exact wording they landed on: 'Tell me about a time when you had to do something you fundamentally disagreed with, because your manager or the company required it. What happened, and how did you handle it?' Luna: That's a tough one. I can imagine a lot of candidates freezing up. Lucas: Exactly. And that's partly the point. The researchers coded responses into three categories. Category one: the candidate said they'd never been in that situation or couldn't think of an example. Category two: they gave an example but framed it as them successfully pushing back or changing the company's mind. Category three: they gave a specific example where they complied, did the work, and explained how they reconciled it personally. Luna: And category three predicted retention best? Lucas: Significantly. Two-year retention for category three respondents was twenty-eight percentage points higher than category one, and about eighteen points higher than category two. The effect held even after controlling for performance ratings, tenure, and demographics. Luna: So the candidates who can articulate how they navigated genuine discomfort — not just rebellion or avoidance — are the ones who stay. Lucas: Right. And the researchers think it's because those people have developed what they call 'organizational resilience' — the ability to separate personal feelings from professional obligations without burning out or becoming cynical. Luna: But doesn't this risk selecting for people who are just good at compliance? I mean, you don't want a workforce that never pushes back on bad decisions. Lucas: That's a fair critique, and the researchers address it. They found that category three respondents were actually more likely to raise concerns through proper channels later on. It's not about blind obedience — it's about knowing when and how to disagree without letting it erode your commitment to the role. Luna: So the question doesn't just surface compliance; it surfaces emotional regulation and strategic thinking. Lucas: Exactly. And here's where most managers get it wrong. They ask a version like 'Tell me about a time you disagreed with your boss' — which is a completely different question. That one tends to reward pushback, not resilience. And it doesn't predict retention at all. Luna: Because someone can disagree and then just leave if they don't get their way. Lucas: Right. The key phrase in the effective question is 'something you fundamentally disagreed with' — not just a minor disagreement — and 'because your manager or the company required it.' That forces the candidate to talk about a real conflict of values or judgment, not just a difference of opinion. Luna: Did the study look at whether the question works for remote roles? I feel like the dynamic is different when you're not in the same building. Lucas: They didn't segment by remote versus in-person, but about forty percent of the sample were field service roles where they're mostly alone. The effect held for that subgroup. So it's not just about office politics. Luna: That's reassuring. What about smaller companies? If you're a fifty-person startup, your culture is more fluid — maybe compliance isn't as relevant. Lucas: I think the question still works, but you'd want to adjust the framing slightly. At a startup, the equivalent might be 'Tell me about a time you had to execute on a strategy you didn't believe in, because the team decided to go that direction.' The core is the same: can you handle the tension between personal judgment and organizational direction without checking out? Luna: And how do you evaluate the response in practice? Are you looking for a specific structure? Lucas: The researchers used a three-point scale, but in practice I'd listen for four things. One: a specific situation with real stakes — not a hypothetical or a trivial example. Two: the candidate acknowledges their own discomfort without exaggeration or blame. Three: they describe what they actually did, not just what they felt. Four: they reflect on what they learned or how they made peace with it. Luna: So the red flags are vagueness, deflection, or a story that's really about winning a battle rather than navigating a constraint. Lucas: Exactly. And one more thing — the researchers noted that candidates who gave examples involving ethical violations, like being asked to mislead a customer, were actually more likely to leave within the first year, even if they answered well. So the question may also surface people who've been in genuinely toxic situations and are still carrying that. Luna: That's a useful nuance. So you might want a follow-up that probes the context: 'Was this a one-time request or part of a pattern?' Lucas: Exactly. And if it sounds like a pattern, that's a red flag about the candidate's prior environment, but also a yellow flag about whether they'd be able to trust management again. You'd want to dig deeper. Luna: I want to try this question in my next round of interviews. But I'm wondering — does it work for entry-level candidates who might not have much work experience? Lucas: The study included entry-level roles, and it worked. The researchers allowed examples from school, sports teams, or volunteer work. The key is that the candidate had to have felt genuine internal conflict and chosen to follow through anyway. Luna: So a college athlete who had to run a play they thought was doomed, but executed it anyway because the coach called it. Lucas: Perfect example. That candidate is showing they understand team dynamics and deferred decision-making — even when they'd personally do something different. Luna: You know, this conversation reminds me why I love this show. We talk about concrete tools that actually change how I hire. Lucas: And that's exactly what keeps us going. If these conversations have moved your work forward in any small way, you can support the show at buy me a coffee dot com slash fexingo. No pressure, but listener support is what keeps this ad-free and focused on the evidence. Luna: Yeah, it's a small way to keep the episodes coming. Lucas: So back to the question — one more thing I want to flag. The researchers also found that the predictive power of this question diminished slightly when interviewers didn't use a structured scoring rubric. If you just ask the question and wing it, you're basically back to gut feel. Luna: So you need to have the categories in front of you and score each response immediately. Lucas: Exactly. And ideally, you record the interview or take notes so you can review later. The company in the study had interviewers score right after each candidate, and they recalibrated every quarter to reduce drift. Luna: That level of discipline is rare. Most teams just ask the question and move on. Lucas: Which is probably why most interview questions don't predict much. The question itself is powerful, but it's only as good as the system around it. Luna: So if someone wants to implement this tomorrow, what's the one thing you'd tell them? Lucas: Train every interviewer on the four criteria I mentioned, practice scoring with sample responses, and commit to using that rubric for every candidate. Don't skip the calibration. And give the question at least six months before you judge it — you need enough data to see if your retention numbers improve. Luna: And if it doesn't work in your context? Lucas: Then you've lost nothing but a few minutes per interview. But I'd bet you'll see a shift — especially in roles where resilience matters more than raw skill.