Latest / The Diversity Career Podcast with Fexingo: Inclusion, Equity, and Underrepresented Professionals / How Unconscious Bias Training Actually Failed Companies
Transcript
- Lucas: If we've helped you think through a career move or decision this week, remember that small support keeps these conversations going. A couple of dollars a month is genuinely what keeps these shows ad-free and available. You can find us at buy me a coffee dot com slash fexingo if you've gotten something out of them. Luna: Thanks for the support. It really does help us dig into these harder topics without the noise. Now, let's get back to the real issue: why our standard diversity playbook isn't working. Lucas: Right. So, look, we've spent thirty years telling companies to run mandatory unconscious bias training. The logic was simple: people have biases, so tell them they have biases, and they'll stop being biased. It felt like common sense. But the data from September 2026 tells a very different story. Studies show that this type of training often has no long-term effect on hiring or promotion rates, and in some cases, it actually increases resentment among majority-group employees. Luna: That backlash effect is the part that always surprises people. They expect resistance to the of diversity, not to the about it. Lucas: Exactly. And the reason is psychological reactance. When you force someone to confront their own hidden prejudices in a high-stakes, mandatory setting, their brain goes into defense mode. They don't learn better interviewing techniques; they just learn how to hide their bias better, or they disengage entirely. I looked at a report from a major consulting firm last week showing that over sixty percent of Fortune five hundred companies still require this annual training, despite evidence that it's ineffective. That’s a massive misallocation of resources. Luna: So if awareness doesn't work, what actually changes behavior? Is it just about better processes? Lucas: It’s about structural constraints. The most effective interventions aren't about changing hearts and minds directly; they're about removing the human discretion where bias usually creeps in. Let's talk about structured interviews. This sounds boring, but it’s the single highest-impact tool we have. In an unstructured interview, two candidates might be asked completely different questions based on the interviewer's mood or interest. One gets asked about their favorite college memory; the other gets grilled on technical specifics. Structured interviews use identical questions and standardized scoring rubrics for every candidate. It removes the 'gut feeling' which is usually just bias in disguise. Luna: It feels robotic, but robots are fairer than humans when it comes to consistency. Do companies actually do this well? Lucas: Most don't, which is why it's such a competitive advantage. Take a mid-sized tech firm we’ve been tracking. They switched from free-form chats to a rigid four-question framework for all senior hires. Within eighteen months, their promotion rate for women and minority candidates went up by twenty-two percent. Not because those candidates were suddenly more qualified, but because the previous system was so inconsistent that it favored people who were charismatic or shared the interviewer's background. Structure levels the playing field. Luna: I love that number. Twenty-two percent is huge. But what about after they’re hired? Promotion bias is often worse than hiring bias. Lucas: You're hitting on the second big failure point: mentorship versus sponsorship. For decades, companies pushed mentorship programs as the solution for underrepresented professionals. Mentors give advice. Sponsors use their political capital to advocate for promotions behind closed doors. The data shows that while everyone needs mentors, only sponsored individuals get promoted at equal rates. The problem is that senior leaders, who are mostly white men, tend to mentor people who look and think like them. It’s natural affinity bias. So you need a formal sponsorship program that ties executive bonuses to the promotion rates of their direct reports from underrepresented groups. Luna: Tying it to compensation makes it real. Otherwise, it’s just another nice to have initiative that falls apart when budgets get tight. Lucas: Precisely. And there’s a third lever that’s gaining traction in 2026: algorithmic audit tools for performance reviews. Yes, algorithms have their own biases, but they are auditable. Human managers are not. Some forward-thinking firms are using software to scan performance review text for gendered language. They found that women were described with communal words like 'supportive' or 'collaborative,' while men were described with agentic words like 'strategic' or 'visionary.' These subtle linguistic differences heavily influence promotion decisions. By flagging this language before it goes to HR, companies can force managers to justify their ratings with concrete metrics rather than personality assessments. Luna: That’s fascinating. It turns a subjective judgment into a data point. Does that create pushback from managers who feel micromanaged? Lucas: Initially, yes. There’s always friction when you remove autonomy. But the key is framing it as quality control, not policing. One global bank told me their managers initially hated the review scanner. Then they realized it saved them time. Instead of writing vague paragraphs that got rejected by HR, they had to fill in specific boxes tied to business outcomes. It streamlined the process. And once they saw that their top performers were getting recognized more fairly, they bought in. It wasn’t about diversity for diversity’s sake; it was about retaining talent that was previously slipping through the cracks. Luna: I think that’s the shift we’re seeing now. It’s moving from moral imperative to operational efficiency. People respond better to efficiency arguments sometimes. Lucas: Absolutely. And let’s talk about the blind resume screening piece again, because it’s surprisingly hard to implement correctly. Many companies tried this ten years ago and abandoned it because it seemed too cumbersome. But with modern applicant tracking systems, it’s easier than ever. You strip names, universities, and graduation years from resumes before they reach the hiring manager. Just the skills and experience remain. A study published earlier this year showed that blind screening increased the callback rate for candidates with african american sounding names by nearly forty percent. Forty percent. That’s not a rounding error. That’s a fundamental flaw in the old system. Luna: Forty percent is staggering. It suggests that the talent pool was always there; we just weren't looking at the right part of the page. Lucas: Right. And here’s the thing about blind screening: it doesn’t fix everything. It only fixes the initial filter. Once the person is in the door, you still need those structured interviews and sponsorship programs I mentioned. If you blind screen but then hire based on 'culture fit,' you’re just swapping one bias for another. Culture fit is often code for 'people I’d want to have a beer with,' which is inherently exclusionary. We need to replace culture fit with culture add. Who brings a perspective, skill set, or background that the current team lacks? Luna: Culture add is such a simple reframe, but it changes the entire dynamic of the interview. It makes inclusion a strategic goal rather than a compliance checkbox. Lucas: Exactly. And it forces the team to be intentional about who they bring in. If your team is homogenous, adding someone who looks exactly like them doesn’t add anything. Adding someone who challenges the status quo or brings a different customer insight adds value. It aligns diversity with innovation. Companies that focus on culture add tend to have higher engagement scores because employees feel their unique identities are assets, not liabilities. Luna: But measuring culture add is hard. How do you quantify 'adding a perspective' during a hiring cycle? Lucas: You start by mapping the team’s existing skills and backgrounds. Before posting a job, ask: what are the three biggest gaps in our collective viewpoint? Maybe we have four engineers who went to the same three schools. Maybe we have no one who understands rural markets. Define those gaps explicitly. Then, write the job description to target those gaps. If you’re looking for someone to expand your market reach, state that clearly. It attracts candidates who see themselves as problem-solvers, not just task-completers. It shifts the narrative from 'fitting in' to 'filling a hole.' Luna: I like that. It makes the hiring decision objective. You have a gap, you need to fill it. It’s less emotional, less biased. Lucas: And it reduces the anxiety for candidates too. Underrepresented candidates often spend energy trying to decode whether they’ll 'fit in' to a new environment. If the role is framed around solving a specific, high-value problem, they can focus on their competence. It’s a more professional, respectful way to engage talent. Luna: Do you see any industries leading this charge, or is it still mostly tech and finance? Lucas: Healthcare is surprisingly ahead of the curve. Because patient outcomes depend on diverse communication styles, hospitals have been implementing structured intake protocols and bias audits for years. They have to. The liability of a misdiagnosis due to cultural misunderstanding is too high. Manufacturing is also catching up because supply chain pressures demand broader talent pools. They’re finding that blind screening and structured interviews reduce turnover significantly. Turnover costs money, and in low-margin industries, that matters. Luna: So cost avoidance is becoming the primary driver. That’s a powerful motivator for CFOs who usually ignore diversity initiatives. Lucas: Yes. And we’re seeing CFOs get involved in the design of these programs. They want to know the ROI. And the ROI is clear: lower recruitment costs, higher retention, better decision-making. A McKinsey report from earlier this year linked diverse executive teams to profitability in the top quartile by twenty-five percent. That’s not a side benefit. That’s a core business metric. When you tie diversity to profit, it stops being a HR project and starts being a strategy session. Luna: It’s interesting how the conversation has evolved from 'doing the right thing' to 'doing the smart thing.' Does that alienate the original advocates for equity? Lucas: It shouldn’t, but it can if not handled carefully. The best leaders frame it as both. Ethics and economics are not mutually exclusive. In fact, they reinforce each other. If you only talk about ethics, you lose the skeptics. If you only talk about economics, you lose the heart. You need both. The narrative needs to be: we are building a fairer system because it is right, and we are building a more efficient system because it works. They are two sides of the same coin. Luna: That balance is crucial. Otherwise, it feels transactional. People want to believe their work has moral weight. Lucas: Agreed. And finally, let’s touch on the role of external benchmarks. Companies used to compare themselves to industry averages. Now, the best ones are comparing themselves to top deciles. They realize that average is bad. Average includes the failures. They want to be in the top tier of inclusion, which requires radical transparency. Publishing their own diversity data, including the gaps, builds trust. It shows they aren’t hiding. Hiding data destroys credibility. Sharing it invites accountability. Luna: Transparency is scary for most CEOs. But silence is scarier in the age of social media. Every mistake is documented. Lucas: Exactly. And the consumers, especially younger workers, are watching. They will vote with their feet. If a company claims to be inclusive but has no data to back it up, or worse, admits to failing and having a plan to fix it, they win loyalty. The lack of a plan is what kills reputation. So, the takeaway for listeners today is clear: dump the mandatory lecture. Build the structure. Measure the outcome. Tie it to performance. And be honest about where you stand. Luna: Simple steps, but hard to execute. It requires discipline over comfort. Lucas: It does. And that’s the difference between a trend and a transformation. Trends are easy. Transformations are work. But the work pays off in ways that go beyond compliance. It creates a workplace where people can actually do their best work. Luna: I hope more leaders listen to that distinction. Comfort is a trap. Lucas: For sure. What do you think is the next frontier? Is it AI in hiring, or is it something else entirely? Luna: I think it’s post-hire development. We’ve talked about getting people in the door, but keeping them and advancing them is the real bottleneck. Sponsorship programs seem like the logical next step. Lucas: Sponsorship is huge. And tying it to executive pay is the only way to make it stick. Without skin in the game, it’s just talk. Luna: Well said. I think we’ll see more of that in the next few quarters. Executives are realizing they can’t outsource diversity anymore. Lucas: No, they can’t. It has to come from the top down, but it has to be built from the bottom up with structure. It’s a delicate balance. Luna: It is. And it requires constant vigilance. You can’t set it and forget it. Lucas: Definitely. So, for anyone listening who is in a position to change hiring or promotion processes, what’s one thing you could do tomorrow? Luna: Audit your job descriptions. Remove any language that implies a specific lifestyle or background. Focus purely on the skills needed to solve the problem. Lucas: That’s a great starting point. Small changes in language can open up huge pools of talent. It’s low effort, high reward. Luna: Exactly. And it’s a signal to candidates that you care about substance over style. Lucas: Right. Substance first. Style follows. Thanks for thinking through this with me, Luna. Luna: My pleasure. Let’s see if we can get more companies to choose substance. Lucas: Here’s hoping. Until next time, keep questioning the defaults.