Latest / Hiring & Firing with Fexingo: Recruitment, Performance, and Workforce Decisions / How a Hiring Scorecard Reduced Attrition by 30 Percent
Transcript
- Lucas: So a few years ago I was talking to the head of talent at a mid-sized SaaS company — about 400 employees, growing fast — and she told me their first-year attrition rate was pushing 35 percent. Luna: Oof. That's basically one out of three new hires gone within twelve months. Lucas: Right. And the cost of that — recruiting fees, onboarding, lost productivity — was easily north of a million dollars annually. So they decided to try something they'd read about in some academic paper: a structured hiring scorecard. Luna: I've seen those. Some teams use them religiously, others think they're too bureaucratic. What did they actually include? Lucas: Good question. They started by identifying the five competencies that actually predicted success in their top-performing sales and engineering roles. Things like 'problem-solving under ambiguity' and 'collaborative communication.' For each competency, they wrote three behavioral anchors — specific, observable examples of what a weak, moderate, or strong candidate looks like. Luna: So instead of a vague 'rate from one to five,' you have concrete descriptions. That forces interviewers to calibrate. Lucas: Exactly. They also added a section for 'red flags' and 'green flags' — non-scored but noted. And crucially, every interviewer scored the candidate independently before any discussion. That removed the halo effect from the first person who speaks in a debrief. Luna: Did they train interviewers to use it? I feel like a scorecard only works if people actually use it consistently. Lucas: They did a half-day calibration workshop where everyone scored the same mock interview video, then discussed their ratings. It took two rounds to get inter-rater reliability above 0.8. But the payoff was real. Within eighteen months, first-year attrition dropped from 35 percent to 24 percent. Luna: That's a 11-point drop — more than a 30 percent reduction. Just from changing how they evaluated candidates? Lucas: That's what the data showed. But there was a second-order effect I found even more interesting. Their time to hire actually decreased by eleven days on average. Luna: Wait — I would have expected a scorecard to slow things down, since you're adding structure. Lucas: That's what the team assumed too. But what happened was they stopped wasting time on candidates who looked great on paper but scored poorly on the competencies that mattered. The scorecard made them say 'no' faster. And when they did find someone who scored well across the board, the team felt confident enough to move quickly. Luna: So the scorecard acted as a forcing function for discipline. Instead of debating subjective impressions for days, they had data to point to. Lucas: Exactly. The head of talent told me their debrief meetings went from hour-long circular arguments to twenty-minute data reviews. 'Candidate A scores 4.2 on problem-solving, 3.8 on collaboration. Candidate B scores 3.1 and 4.5. Which gap is easier to close?' That clarity is powerful. Luna: I want to push back a little though. I've seen scorecards that are so rigid they filter out people who think differently, who might bring fresh perspective. Doesn't this risk homogenizing the workforce? Lucas: That's a legitimate concern, and I think it depends on how you define your competencies. If you define 'collaboration' as 'always agrees with the team,' then yeah, you'll get groupthink. But if you define it as 'constructively challenges ideas while maintaining relationships,' you're actually inviting diversity of thought. The key is writing behavioral anchors that reward cognitive diversity, not conformity. Luna: So the scorecard itself isn't the enemy — it's how you design it. Lucas: Right. The company I mentioned actually revised their scorecard twice in the first year. They found that one of their original competencies — 'speed of execution' — was weeding out people who were thorough but slower. So they split it into 'pace' and 'quality' and adjusted the weights. Luna: That's smart. It treats the scorecard as a living document, not a decree from HR. Lucas: Absolutely. And I think that's the meta-lesson here: the tool is only as good as the team's willingness to iterate on it. The company's CEO told me the scorecard forced them to get really clear about what they actually valued, which turned out to be a useful conversation in itself. Luna: So if a listener wanted to build something like this for their own team, where would they start? Lucas: I'd say start small. Pick two or three roles that have the highest turnover or the longest time to hire. Involve the top performers in those roles to help define the competencies. Then test the scorecard for a month with just those roles, collect feedback, and iterate before rolling it out company-wide. Luna: And don't skip the calibration workshop. I think that's where most implementations fail — they print the scorecard and expect everyone to use it the same way without practice. Lucas: One hundred percent. The calibration is where you catch the biggest biases. I remember one interviewer in that workshop gave a candidate a 5 on 'resilience' because the candidate mentioned running marathons. The group had to gently point out that running marathons isn't the same as handling a project crisis. Luna: Yeah, that's classic halo effect — letting one positive trait color everything. A good scorecard with good training can catch that. Lucas: If today was actually useful to you — maybe you're thinking about your own hiring process — the way these conversations stay ad-free is listener support. You can find us at buy me a coffee dot com slash fexingo. Luna: Yeah, it's a small way to keep this resource available to anyone who needs it. We appreciate everyone who's chipped in. Lucas: Alright, back to the scorecard story. One more data point I thought was striking: after eighteen months, the company also saw a 15 percent increase in manager satisfaction with new hires. Managers felt they were getting people who actually fit the role, not just people who interviewed well. Luna: That's the ultimate test, isn't it? The people doing the day-to-day work with the new hire should feel the difference. Lucas: Exactly. And it wasn't just a feel-good metric — they tracked it through a simple quarterly survey. I think that's a good reminder that process improvements should be measured, not just assumed. Luna: So what's the one takeaway you'd leave listeners with? Lucas: I'd say this: if you're hiring based on gut feel and resumes alone, you're leaving a lot of money on the table. A structured scorecard doesn't eliminate judgment — it focuses it on the things that actually predict performance. And the data shows it works. Luna: Well said. Thanks for digging into this one. Lucas: Always a pleasure. Catch you next time.