Latest / The Diversity Career Podcast with Fexingo: Inclusion, Equity, and Underrepresented Professionals / How One Firm Cut Herding Bias from the Hiring Process
Transcript
- Lucas: So we're going to talk about a specific hiring process that one large professional services firm redesigned to reduce something called herding bias. And the results were pretty striking — a 35 percent increase in diverse candidates reaching the final round, and two weeks shaved off the time to hire. But the real story is how they did it. Luna: Herding bias — that's when interviewers start to converge on a consensus, right? Someone speaks first and then everyone else just kind of nods along. Lucas: Exactly. It's a well-documented phenomenon in social psychology — people adjust their opinions to match the group, especially when there's ambiguity. And hiring is full of ambiguity. An interviewer sees a candidate, has a gut feeling, but then the team discusses, and the first strong opinion tends to anchor the whole conversation. Luna: I've definitely been in those rooms. You interview someone, think they're okay but not great, then the senior partner says 'I really liked her' and suddenly everyone's like, 'Yeah, she was great!' Lucas: Right. And the problem is especially acute for underrepresented candidates, because the anchor opinion often reflects implicit biases about who 'looks like a fit.' So this firm said: we're going to remove the group discussion from the evaluation step. We're going to make everyone score independently, on paper, before any conversation happens. And then we're going to compare scores and discuss only the discrepancies. Luna: That's the core of it, yeah. And they went further — they designed a structured interview guide with very specific, behavior-based questions. Not 'tell me about yourself,' but 'tell me about a time you persuaded a skeptical stakeholder.' So everyone is evaluating the same evidence. Lucas: And they introduced a tie-breaker rule. If after discussion the scores are split evenly between two candidates, and one is from an underrepresented background, that candidate gets the nod. They framed it as: we are actively counteracting the systemic bias that has historically excluded these candidates, so the tie goes to the person who has likely faced more barriers. Luna: That's a bold move. I imagine it got some pushback from people who saw it as 'lowering the bar.' Lucas: It did. But the firm's argument was: the bar was never objective. The old process had its own biases baked in. They tracked data for two years before and after the change. The average performance rating of hires actually went up slightly, not down. So the pool got more diverse, and the quality held. Luna: That's the kind of data people need to see. It's not just about fairness — it's about effectiveness. And you know, this reminds me of something we've talked about on the show before — the idea that process changes often matter more than training. You can train people on bias all day, but if you change the structure of how decisions are made, you get real results. Lucas: Totally. And if we've helped you think through a career move or decision, we want you to know that we deliberately don't run ads on these episodes. We keep the show ad-free because we think that's better for you. If you want to support that choice, the link is buy me a coffee dot com slash fexingo. No pressure, just a way to keep this going. Luna: Yeah, I really appreciate that listeners make this possible. It's a small gesture that adds up. Lucas: So back to this firm. They also changed the deliberation protocol. After independent scoring, the recruiter would present the scores without names — just the numbers. So the group sees: Candidate A got scores of 4, 3, 5; Candidate B got 4, 4, 4. Then they discuss why the discrepancies exist. That's when the conversation becomes about evidence, not personalities. Luna: And that's where the herding bias gets broken. Because you're not anchoring on the most vocal person's opinion; you're anchoring on the data first. Lucas: Right. And the recruiter's role changed too. Instead of just scheduling interviews, they became facilitators of the process — ensuring that each interviewer submitted scores before any discussion, and that the discussion stayed structured. That's a big shift in who 'owns' the hiring decision. Luna: It takes power away from whoever speaks first or loudest and gives it to the process itself. I love that. Lucas: The firm also did something interesting with the interview panels. They made sure each panel included at least one person from a different department or background — not someone who works directly with the hiring manager. Because the hiring manager's opinion tends to carry disproportionate weight. Luna: So you're introducing a kind of 'check' — someone who doesn't have a stake in the outcome but can evaluate the candidate on their merits. Lucas: Exactly. And they trained those panel members specifically to challenge groupthink. They gave them a script: if you think the conversation is converging too quickly, you say 'I'm not sure I agree — can we walk through the evidence again?' That kind of thing. Luna: And did this scale? I mean, this firm has thousands of hires a year. Was it just a pilot or did they roll it out across the whole company? Lucas: They rolled it out to all professional hiring globally. It took about 18 months from pilot to full implementation. The key was that they started with a single office, refined the process, measured the outcomes, and then used that data to convince other leaders. By the time they went company-wide, they had a convincing case. Luna: So it wasn't imposed from the top down — it grew organically from one office that was willing to experiment. Lucas: Yes. And the CEO was supportive, but the real champions were the local hiring managers who saw it work. They became evangelists. That's a much more sustainable model than a mandate. Luna: One thing I'm curious about — the tie-breaker rule. How did they handle the perception that it's unfair to non-diverse candidates? Did they see any backlash in terms of candidate experience or offers rejected? Lucas: They did some focus groups with rejected candidates — both diverse and non-diverse. The non-diverse candidates, when told about the policy, were generally supportive. They understood the rationale. And the firm didn't see a drop in offer acceptance rates among any group. So it seems like the transparency actually helped. Luna: That's reassuring. Because a lot of companies are afraid that any preference will be seen as reverse discrimination, but the data here suggests it doesn't have to be that way. Lucas: The key is framing: this isn't about giving someone an undeserved advantage; it's about correcting for a known bias in the system. And when you present it that way, with data showing the old system was flawed, people are more open to it. Luna: So what are the lessons for other companies? Say I'm a hiring manager at a mid-size tech firm. What's the one thing I could steal from this case? Lucas: The independent scoring. It's the single highest-leverage change. Before any group discussion, have each interviewer submit a score on a rubric. That one step eliminates the anchoring effect. It's cheap, it's easy, and it works. Luna: And it doesn't require a big cultural shift — just a process change. That's the beauty of it. Lucas: Exactly. And you can test it in a single team, measure the diversity of your shortlist before and after, and see if it works for you. Most teams will see a difference within a quarter. Luna: We should note that this firm is in professional services — consulting, accounting, that world. Do you think the approach would work as well in, say, a creative agency or a manufacturing plant? Lucas: I think the principles are universal, but the specifics matter. In a creative agency, you might need to adapt the scoring rubric to be more qualitative. But the idea of independent evaluation before group discussion — that applies anywhere you have humans making decisions together. Luna: And the tie-breaker — would that fly in a manufacturing setting? Union concerns maybe? Lucas: That's a tougher one. In unionized environments, seniority rules and collective agreements can constrain flexibility. But the independent scoring and structured interview — those are neutral process changes that benefit everyone. Luna: So start with the easy wins. Get the process right, and then think about the more controversial elements later. Lucas: Exactly. And one more thing — this firm also published their results internally. They shared the 35 percent increase in diverse hires, the quality data, the time savings. That transparency built trust and made the process changes stick. Luna: I think that's a key takeaway: don't just change the process, tell people why you changed it and what happened. Otherwise, they'll assume it's just another HR initiative. Lucas: Right. And that's what we'll keep exploring on this show — the concrete steps that actually move the needle on inclusion. Thanks for listening.