Latest / Hiring & Firing with Fexingo: Recruitment, Performance, and Workforce Decisions / Why Skills Based Hiring Is Beating Degree Screens
Transcript
- Lucas: So last week I was looking at a LinkedIn report from January 2026 — their Global Talent Trends — and one number jumped out at me: roles that used skills-based filters instead of a degree screen saw a 23 percent larger applicant pool and a 14 percent lower first-year turnover rate. Luna: Twenty-three percent more applicants and you keep them longer? That's the kind of number that makes HR people sit up. Lucas: It does. And it's not just tech startups anymore. IBM, Bank of America, Walmart — they've all removed bachelor's degree requirements for a significant chunk of roles. I think IBM's number is something like half their US job postings no longer require a four-year degree. Luna: Half? That's a lot. And are they actually seeing the same performance from non-degree hires? Lucas: That's the key question. The LinkedIn data suggests yes — at least on retention. But there's a specific case I want to drill into. There's a mid-size logistics firm out of Columbus, Ohio — I'll call them MetroLogistics because they're not a household name. They did a controlled experiment in 2025 on their warehouse supervisor roles. Luna: What kind of control? Lucas: They split hiring into two tracks. Track A kept the old system — required an associate degree or higher, plus two years of experience. Track B removed the degree requirement entirely and replaced it with a structured work sample test: candidates had to plan a shift schedule given a set of constraints — delivery volumes, staffing levels, peak hours. It took about 45 minutes. Luna: So they swapped the credential for a task. That's a pretty high-fidelity way to test the actual job. Lucas: Exactly. And the results after nine months: Track B hires performed at or above Track A hires on every metric — on-time shipments, team satisfaction scores, even overtime costs were slightly lower. But here's the kicker: time to hire dropped by 11 days. Luna: Eleven days? That's massive for a warehouse role where every empty seat costs you throughput. Lucas: Right. And the applicant pool was not just larger — it was more diverse. They saw a 32 percent increase in applicants who identified as non-white and a 19 percent increase in women candidates. But I want to be careful here. Not everyone is convinced this scales to every role. Luna: Yeah, I was about to say — a work sample for a warehouse supervisor is one thing. What about a senior financial analyst or a product manager? Can you really test for that in forty-five minutes? Lucas: That's the pushback I hear most often. And it's fair. But the trend is moving toward what's called 'task-based interviewing' — not necessarily a single test, but a structured set of job-relevant exercises. For a financial analyst, you might give them a messy data set and ask them to build a three-statement model in two hours. For a product manager, it's a product critique exercise followed by a mock prioritization session. Luna: So it's less about a one-size-fits-all test and more about designing a simulation that actually reflects the job. Lucas: Precisely. And there's a whole ecosystem of tools that have grown up around this. Platforms like TestGorilla and Codility let companies administer skills assessments at scale. TestGorilla does everything from cognitive ability tests to role-specific scenarios. Codility is more tech-focused — coding challenges, obviously — but they've expanded into data science and even some product management tasks. Luna: I've used Codility on the candidate side. It's a double-edged sword — some people love the clarity of a concrete task, but others find the time pressure unrealistic. Lucas: That's a legitimate critique. And there's research showing that timed tests can disadvantage candidates from non-traditional backgrounds — people who haven't been trained to perform under that kind of pressure. So the design of the assessment matters a lot. A well-designed work sample gives you a realistic preview of the job without the artificial stress. Luna: Let's talk about the equity angle more directly. Companies like Bank of America made headlines when they dropped degree requirements for many roles — they framed it as an inclusion move. But is the actual outcome better performance, or is it mainly optics? Lucas: I think it's both, but the data is still early. Bank of America hasn't published detailed performance metrics yet — they've mostly talked about pipeline changes. But the LinkedIn study I mentioned tracked over 5,000 roles across 200 companies, and the retention improvement held even after controlling for industry, role seniority, and company size. That's a pretty robust signal. Luna: Still, retention is one thing. What about actual productivity? Did the LinkedIn study look at that? Lucas: They did — they used manager ratings and promotion rates as proxies. Workers hired through skills-based assessments were promoted at roughly the same rate as degree-holders within the first two years. Slightly higher, actually, in some categories. But the sample sizes get smaller when you slice that finely, so I'd take it with a grain of salt. Luna: Fair. So the case for skills-based hiring seems strong for entry to mid-level roles. What about executive positions? Lucas: That's where the degree screen is stickiest. A lot of C-suite roles still list an MBA or equivalent as preferred if not required. But even there, some companies are experimenting. I know of a mid-cap tech firm that hired a VP of Engineering without a college degree — he'd built his career at a series of startups and had a reputation as a phenomenal team builder. And his performance has been strong. But those are still outliers. Luna: It feels like the biggest barrier isn't the assessment design — it's habit. Recruiters and hiring managers are used to using the degree as a quick filter. It's cognitively easy. Lucas: That's exactly right. And there's a status-quo bias at play. If you've always hired from certain schools and gotten good results, why change? The counterargument is that you're missing a huge pool of talent that didn't have the opportunity or the desire to get a degree but has the actual skills. And in a tight labor market — which we've had for the last few years — that's a competitive disadvantage. Luna: But what happens if we enter a recession? When there are more applicants per role, do companies revert to the degree filter because it's faster? Lucas: That's the million-dollar question. Historically, during downturns, employers do become more credential-heavy. It's a buyer's market, so you can afford to be picky. But I think there's a counter-trend now: the companies that have invested in skills-based infrastructure — the assessments, the training for interviewers, the data tracking — they're less likely to abandon it because they've seen the results. And the cost of a bad hire is even higher when you're trying to protect margins. Luna: So the ones who've built the muscle will keep using it. The ones who haven't might fall back on old habits. Lucas: That's my guess. And the companies that do it well are also the ones that are most transparent with candidates about how the assessment works. They tell you upfront: 'You'll be given a case study, here's what we're looking for, here's how long it takes.' That builds trust and reduces anxiety. Luna: We've had a few listeners ask about how to prepare for these kinds of assessments. Do you have any practical advice? Lucas: Yeah, a couple things. First, if you're applying to a company that uses skills-based screening, the biggest mistake is not taking the practice tests seriously. Many platforms offer sample assessments — do them under timed conditions. Second, read the job description carefully. The assessment is usually designed to mirror the tasks listed there. If the role says 'analyze sales data to identify trends', expect a data analysis exercise, not a general logic puzzle. Luna: And for hiring managers listening: what's the single biggest thing they should get right if they want to adopt this approach? Lucas: Start with one role family — don't try to redesign your entire hiring process at once. Pick a role where you have a clear performance baseline, design the work sample based on input from your top performers, pilot it with a small group of candidates alongside your traditional process, and measure the outcomes. Once you have data that convinces your stakeholders, then expand. Luna: Smart. Prove it works before you sell it to the org. Lucas: Exactly. And that's how the best companies are doing it. It's not about a grand policy announcement — it's about one team showing that a better filter exists. Luna: If today's episode was useful to you and you want to keep it ad-free, buy me a coffee dot com slash fexingo helps. Lucas: Yeah — a handful of listeners chipping in is literally what keeps these conversations going. No pressure, just if it's worth that to you. Luna: Alright — back to the data. One thing I've been wondering: do skills-based assessments actually predict performance better than interviews? Lucas: There's a meta-analysis from 2024 that suggests they do — work samples have a predictive validity of about 0.54, versus unstructured interviews around 0.20. Structured interviews are better, around 0.35, but still below work samples. The gap is significant. Luna: Point-five-four is pretty high for a single selection method. What about cognitive ability tests? Those are usually around 0.5 too. Lucas: They are. But cognitive ability tests can have adverse impact depending on the population. Work samples, when designed carefully, tend to have smaller subgroup differences. That's part of why companies like them — they're both valid and more equitable. Luna: So the ideal might be a combination: a short cognitive screen plus a work sample plus a structured interview? Lucas: That's the gold standard many firms are moving toward. But it's resource-intensive. The work sample alone can take an hour to administer and grade. So you need to decide where to invest that time. For high-volume roles, you might automate the grading using AI — though that introduces its own set of biases to watch for. Luna: Right — algorithmic bias is a whole other episode. But for now, I think the takeaway is: skills-based hiring isn't a fad, but it's not a silver bullet either. Lucas: Exactly. It's a tool that works well when applied thoughtfully. And the companies that are doing it well are the ones that treat it as an ongoing experiment, not a one-time policy change. Luna: I'd love to check in a year from now and see if the recession question changed anything. Lucas: Me too. I suspect the companies that have invested in the infrastructure will stick with it, and the ones that haven't will revert. But that's a guess, not a prediction. Luna: Fair enough. Thanks, Lucas. Lucas: Thanks, Luna. That's all for this episode of Hiring & Firing with Fexingo.