Latest / Hiring & Firing with Fexingo: Recruitment, Performance, and Workforce Decisions / Why You Should Hire People Who Failed Before
Transcript
- Lucas: Let me start with a résumé I saw recently that would make most recruiters flinch. Guy in his early thirties, solid university, then three startups in six years. First one—acqui-hired for peanuts after burning through twelve million dollars. Second one—shut down in eighteen months. Third one—he left just before it ran out of runway. On paper, it reads like a string of failures. Luna: And I'm guessing you didn't flinch. Lucas: I didn't. Because I've been watching a really interesting data set from Harvard Business School—a 2024 study that tracked about 2,100 early-career engineers over a four-year window. The researchers found something that cuts against every instinct in traditional hiring. Engineers who had experienced a significant professional failure—a project that cratered, a product launch that flopped, a startup that folded—were 34 percent more likely to be rated a top performer in their next role than engineers with no such failure on their record. Luna: 34 percent—that's not a rounding error. That's a meaningful edge. Lucas: It is. And the effect held even after controlling for years of experience, school quality, and industry. The failure itself—not the recovery narrative, not some hidden privilege—was the variable that predicted outperformance. Luna: If these conversations have moved your work forward in some small way—maybe changed how you think about a hiring decision, or gave you a concrete stat to take to your team—there's a simple way to keep them coming. A couple of dollars a month is genuinely what keeps these going, ad-free and independent. Buy me a coffee dot com slash fexingo, if you've gotten something out of them. Lucas: Yeah, it's a small thing that makes a big difference for us. Back to that 34 percent—what's going on inside that number? The HBS team did something clever. They interviewed a subset of the engineers and coded their responses for specific cognitive patterns. The engineers who had failed were significantly more likely to talk about 'pattern recognition'—they could describe early warning signs they'd missed before. They were also more likely to mention 'experimentation'—trying multiple approaches instead of committing to one plan early. Luna: So failure teaches you how to spot the next failure coming. That's a concrete skill, not just a character trait. Lucas: Exactly. The study's lead author called it 'diagnostic humility'—the willingness to say 'I might be wrong about this assumption' early enough to change course. And that's almost impossible to teach in a training program. You have to live it. Luna: But there's an obvious tension here. Every recruiter has seen the candidate who fails upward, who has a great story about what they learned but then goes and makes the same mistakes at the next company. How do you distinguish genuine reflection from a polished redemption arc? Lucas: That's the million-dollar question. The HBS researchers actually gave a framework for that. They said the key isn't whether the candidate can describe what went wrong—it's whether they can describe what they would do differently in a specific, concrete scenario that's structurally similar to the failure. If they say 'I would communicate better,' that's useless. If they say 'I would set up a weekly cross-functional checkpoint with engineering and product, and I'd flag any deviation from the roadmap within 48 hours'—that's pattern recognition. Luna: So you're testing for transfer, not narrative. You want them to apply the lesson to a new context, not just recite the old one. Lucas: Right. And there's a second signal that correlates with outperformance: the candidate's willingness to talk about what they personally contributed to the failure. Not just 'the market shifted' or 'the CEO made bad calls.' The HBS data showed that candidates who used first-person pronouns when describing the failure—'I misjudged the timeline,' 'I didn't push back on the feature scope'—had higher subsequent performance ratings than candidates who used third-person or passive constructions. Luna: That's fascinating but tricky. Because in an interview, the candidate who says 'I was responsible for everything' might just be playing the humility card. How do you know it's genuine? Lucas: You push on specifics. Ask for the one decision they made that, in hindsight, had the biggest negative impact. Then ask what data they had at the time. Then ask what data they wish they'd had. The candidates who can answer all three without hedging—those are the ones who have really processed the failure. The ones who get vague or defensive on the second or third question—those are the ones who have a story, not a lesson. Luna: That's a good framework. I want to bring in another angle from venture capital. VCs are famous for betting on failed founders. Some firms explicitly prefer second-time founders because the data shows they have higher success rates. Is that the same mechanism? Lucas: I think it is. There's a well-known study from Harvard and Stanford that looked at venture-backed startups and found that founders who had previously failed had a 20 percent higher chance of succeeding in their next venture than first-time founders. The failure itself was a better predictor than the founder's education, prior industry experience, or even the quality of the business plan. The VCs figured this out years ago. But in corporate hiring, we still treat failure as a stain. Luna: Why do you think that gap exists? Why are VCs more comfortable with failure than corporate hiring managers? Lucas: Part of it is asymmetry of consequences. A VC who bets on a failed founder and wins looks like a genius. A VC who bets on a failed founder and loses—well, the founder already had a failure on their record, so the VC has cover. But a hiring manager who brings in someone with a failed startup on their résumé, and that person underperforms? The manager gets blamed for a 'bad hire' and the failure becomes evidence that the manager made a poor judgment. The incentives are misaligned. Luna: So the hiring manager is penalized for taking a risk that the data actually supports. That's a structural problem, not a judgment problem. Lucas: Exactly. And that's why I think the best companies are the ones that build failure tolerance into their hiring process explicitly. They train recruiters and hiring managers to distinguish between 'productive failure'—where the candidate gained diagnostic humility—and 'unproductive failure'—where the candidate just made the same mistake repeatedly. And they give the hiring team permission to make the bet. Luna: Let's get practical. If I'm a hiring manager listening to this, and I want to start looking for failure-signal in candidates, what's the one change I should make to my interview process? Lucas: Add one question to every interview. Ask: 'Tell me about a time you were wrong about something important at work. Not a small mistake—something that had real consequences. Walk me through what you thought, what happened, and what you did about it.' Then listen for whether they can name a specific belief they held, not just a fact they missed. The best answers sound like: 'I believed that our user base would tolerate a slower onboarding if we had more features. I was wrong. We lost 40 percent of new users in the first week.' That's diagnostic humility. Luna: And what do you do with the candidate who says 'I can't think of a time I was seriously wrong'? Lucas: Then you have a different problem. That candidate either lacks self-awareness or hasn't been in situations where they could fail. Either way, the HBS data suggests they're less likely to be a top performer. The engineers who couldn't recall a significant failure had lower performance ratings across the board. It's not that they were bad—it's that they hadn't developed the pattern-recognition muscle. Luna: So the absence of failure on a résumé isn't a positive signal. It's maybe even a negative one. Lucas: I think that's right for certain roles, especially roles that require judgment, adaptability, or creativity. For highly procedural roles where consistency is everything—maybe failure tolerance matters less. But for the kinds of roles that most of our listeners are hiring for—the ones where the job changes every quarter, where you have to make decisions with incomplete information—past failure is one of the best predictors of future success that we have. Luna: I want to push back on one thing. The HBS studied engineers. Does this translate to other functions? Sales? Marketing? Operations? Lucas: The HBS study was engineering-specific, but there's supporting evidence from other fields. A meta-analysis published in the Journal of Applied Psychology in 2023 looked at 47 studies across industries and found a consistent positive correlation between prior failure and subsequent performance in complex, dynamic roles. The effect was strongest for roles with high autonomy and high ambiguity—think product managers, strategy roles, startup generalists. For sales, the data is mixed because failure in sales can sometimes indicate a mismatch with the specific market rather than a generalizable lesson. Luna: So the type of failure matters. The failure of a product launch gives you different lessons than the failure of a sales quarter. Lucas: Exactly. And that's where the interview becomes critical. You're not just looking for any failure—you're looking for a failure that's structurally similar to the challenges of the role you're hiring for. If you're hiring a product manager, you want to hear about a product failure and what they learned about user behavior. If you're hiring a finance director, you want to hear about a forecasting error and what they learned about model assumptions. Luna: One last thought. We're recording this in June 2026, and the labor market is still tight in a lot of sectors. Companies are competing for talent. If you're willing to hire people with failure on their résumé, you're accessing a pool of candidates that most of your competitors are filtering out. That's a real advantage. Lucas: It is. And it's an advantage that's rooted in data, not just intuition. The HBS study gives us a framework: failure teaches diagnostic humility, pattern recognition, and a willingness to experiment. Those are exactly the qualities that matter in a fast-changing environment. So the next time you see a résumé with a failed startup or a derailed project, don't flinch. Dig into it. You might find your next top performer. Luna: And if you want to dig deeper into that HBS study, we'll link to it in the show notes. Thanks for listening. Lucas: Thanks, Luna. Talk to you next time.