Latest / Hiring & Firing with Fexingo: Recruitment, Performance, and Workforce Decisions / Why Apprenticeships Beat College Recruiting for Technical Roles
Transcript
- Lucas: So we talk a lot on this show about the flaws in traditional hiring — degree screens, personality tests, reference checks that reveal nothing. But there's one model that's quietly been outperforming all of them for technical roles, and it's been around for centuries. Luna: You're talking about apprenticeships, right? Not the trade-school kind — the white-collar kind. Lucas: Exactly. Apprenticeships for roles like software engineering, data analytics, cybersecurity. In the US, the Department of Labor's Registered Apprenticeship program has expanded into over a thousand new occupations since 2020, and a growing number of those are in tech and finance. Lucas: Siemens is probably the best-known example. They run an apprenticeship program in the US that's been around since 2011. They take high school graduates, put them through a three-year paid program combining classroom instruction at a community college with rotations through different departments. And their retention rate after three years is 94 percent. Luna: 94 percent — that's wild. What's their retention for conventional college hires? Lucas: They don't publish a direct comparison, but industry data from the Society for Human Resource Management puts the average three-year retention for early-career technical hires at about 68 percent. So you're looking at a roughly 26-point gap. Luna: And that's not even accounting for the cost of recruiting. College recruiting is expensive — career fairs, campus visits, signing bonuses. Apprentices you're basically growing yourself. Lucas: Right. IBM has been doing this at scale for years. They have a program called the IBM Apprenticeship Initiative that started in 2018. They've hired over a thousand apprentices across roles like cloud engineering and data science. And they found that the average cost to recruit and train an apprentice is roughly 30 percent lower than the cost to recruit and train a comparable new college graduate. Luna: Because you're not paying for the recruiting overhead, and the apprentice is productive earlier. They're learning on the job from day one. Lucas: Exactly. And there's a really specific number from a study by the Washington State Workforce Board. They looked at the total cost per hire for apprentices versus external hires across multiple companies. For a software engineering role, the five-year net benefit per apprentice was roughly one hundred and eighty thousand dollars — that's after accounting for training costs, wages during apprenticeship, and the fact that they ramp faster. Luna: One hundred and eighty thousand dollars per person. That's not a rounding error. Lucas: No, it's not. And the interesting thing is that these programs don't just produce better retention and lower cost — they also produce different kinds of employees. Salesforce, for instance, started a program called Futureforce Apprenticeships. They take people from non-traditional backgrounds — career changers, veterans, people who've been out of the workforce — and train them as Salesforce administrators and developers. And they found that their apprentices are promoted into management roles at a rate about 40 percent higher than their conventional hires. Luna: That makes intuitive sense. If you've been through an intensive apprenticeship, you've already been trained on the company's specific systems and culture. You're not starting from scratch on day one. Lucas: Right. And you've also built relationships across the company during the program. So when a management slot opens up, you're already a known quantity. Luna: But let's be honest — this is still pretty rare in the US. Most companies still default to the four-year degree filter. Why is that? Lucas: Part of it is inertia. The degree screen is easy — it's a binary filter. An apprenticeship program requires real infrastructure. You need a curriculum, you need mentors, you need to coordinate with a community college or a training provider. It's a lot more work upfront. Lucas: But there's also a perception problem. A lot of hiring managers still see an apprenticeship as second-best — like you're settling for someone who couldn't get into college. And that assumption is completely at odds with the data. Luna: I think that's changing, especially as the cost of college continues to go up and the ROI of a degree gets questioned more. I've seen some really interesting numbers from Zurich Insurance — they have a huge registered apprenticeship program. Lucas: Yeah, Zurich is a great case. They started their program in 2001, and it's grown to cover roles in underwriting, claims, IT, and actuarial science. They now have over 400 apprentices in the US alone. And their data shows that after five years, apprentices are actually outperforming college hires on performance reviews — by about 7 percent on average. Luna: Seven percent — that's statistically significant. And it's across multiple job families, not just one function. Lucas: Right. And Zurich also tracked something else — their apprentices are more likely to stay with the company long-term. The ten-year retention rate for their apprenticeship graduates is 71 percent, compared to 52 percent for their traditional college hires. Luna: That's a 19-point gap. Over a decade, that saves an enormous amount of replacement cost. Lucas: It does. And when you factor in the lower starting salary during the apprenticeship years — apprentices earn a wage but it's typically lower than a starting salary for a college grad — the total cost advantage becomes even starker. Some estimates from the Urban Institute suggest that for every dollar a company invests in a registered apprenticeship, they get back an average of one dollar forty-seven in increased productivity and reduced turnover. Luna: It's hard to argue with that kind of return. So what's holding more companies back? Lucas: One big hurdle is the regulatory structure. Registered Apprenticeship programs have to meet Department of Labor standards — there are requirements around on-the-job training hours, related instruction, progressive wage scales. Some companies see that as too much bureaucracy. Lucas: But there are also industry consortiums that make it easier. For example, the Apprenticeship in Computing program run by the state of Washington — multiple employers share the cost of developing curriculum and training infrastructure. It's a way to get the benefit without building everything from scratch. Luna: And I've noticed that some of the companies doing this well are also the ones that have been hiring through boot camps for years. They already have a culture of skills-based hiring. Lucas: That's a good point. Companies like Google and Amazon have been hiring from coding boot camps for a while, but they haven't really moved to formal apprenticeship programs at scale. Google had a program called Google Apprenticeships that launched in 2020 for IT support and data analytics — it's real, but it's still small relative to their total hiring. Luna: One thing I want to push back on — a lot of the data we're citing comes from companies that have been doing apprenticeships for over a decade. They've had time to optimize their programs. If a company starts one tomorrow, can they expect similar results? Lucas: Fair question. The data from newer programs is more mixed. The key variable seems to be the quality of the on-the-job training. If the mentors are well-trained and the rotations are structured, results are good. If it's basically just cheap labor with a few classes on the side, it doesn't work. Lucas: The Department of Labor actually has a set of best practices — they recommend a minimum of 2,000 hours of on-the-job training and at least 144 hours of related instruction per year. Programs that follow those guidelines tend to have much better outcomes. Luna: So the infrastructure matters. You can't just throw someone into a team and call it an apprenticeship. Lucas: Exactly. And that's where I think some companies trip up — they see it as a cost-cutting measure rather than a talent-development strategy. The companies that do it well treat it as a long-term investment in their pipeline. Luna: It's interesting — we've been talking about this purely from the employer's perspective, but there's a macro angle too. Apprenticeships are one of the few tools that actually increase the labor supply for technical roles. A lot of people who are perfectly capable of doing the work never get a four-year degree. Lucas: Right. And with the labor market tightening in certain technical fields — cybersecurity, cloud architecture, data engineering — companies that can tap into that pool have a real advantage. It's not charity, it's strategy. Luna: It's also worth mentioning that the demographic trend is on the side of apprenticeships. The number of 18-year-olds in the US is going to peak around 2026 and then start declining. Companies that rely solely on the college pipeline are going to face a shrinking pool. Lucas: That's a really important point. And it ties back to something we've talked about before — the companies that build alternative pipelines now will be the ones that don't have a talent crisis in five years. Lucas: If these conversations have moved your work forward in some small way — maybe you've changed how you think about your own hiring process — we'd love to keep doing them. The reason we can keep this show ad-free is that some listeners choose to support it directly. If that's something you'd like to do, the link is buy me a coffee dot com slash fexingo. Luna: Yeah, I think it's a nice way to keep the conversation focused on the substance rather than on ad reads. So if you're finding value, that's the place. Lucas: Alright — back to apprenticeships. One final data point I want to leave listeners with. A study from the National Bureau of Economic Research looked at Swiss apprenticeship programs — Switzerland has one of the most developed systems in the world — and found that apprentices who complete the program earn about 15 percent more over their careers than similar workers who didn't do an apprenticeship. And that's controlling for ability. Luna: So it's good for the worker, good for the employer. The question is why more US companies aren't doing it. Lucas: I think the answer is changing, slowly. Every time a company like Siemens or Zurich releases a positive retention number, it gets shared in HR circles. There's a growing body of evidence that makes it harder to dismiss the model as irrelevant to white-collar work. Luna: It's one of those rare things in hiring where the data and the intuition actually line up. You train people in your own systems, they stay longer, they perform better. It's almost too simple. Lucas: It is simple. But it's not easy. The upfront investment in curriculum, mentorship, and partnership with educational institutions is real. But the companies that have made that investment are seeing returns that most hiring initiatives can't touch. Luna: So if you're a hiring manager listening, what's the one thing you should do tomorrow? Lucas: Go to the Department of Labor's Office of Apprenticeship website and look up the Registered Apprenticeship standards for your industry. Then call a community college in your area and ask if they have an existing partnership with any employer for a program like this. Nine times out of ten, they'll have a template ready to go. Luna: And don't try to build it all yourself. That's the mistake. Lucas: Exactly. Use the infrastructure that's already there. The data is on your side.