Latest / The Productivity Podcast with Fexingo: Output, Efficiency, and Long-Term Economic Growth / How the Fifties-Era Factory Still Haunts Productivity
Transcript
- Lucas: So there is a number that has haunted American productivity thinking for seventy years, and it's ninety-two. Ninety-two widgets per hour. Luna: Ninety-two widgets per hour — that sounds like a specific time and motion study. Lucas: Exactly. 1956, General Electric's Erie, Pennsylvania plant. Industrial engineers spent six months with stopwatches and clipboards, timing every movement of assembly-line workers. Their conclusion: a properly trained operator, given optimal conditions, could produce ninety-two units in a sixty-minute hour. That became the standard. Luna: And that standard didn't just stay in Erie. Lucas: It seeded an entire management philosophy. The idea that productivity is about measuring discrete, repeatable actions against a benchmark. If you're below ninety-two, you're inefficient. If you're above, you're a star. That framework worked beautifully for mid-century manufacturing. Luna: But we're not in mid-century manufacturing anymore. Most of us are doing cognitive, collaborative, often non-repetitive work. Lucas: And yet that ninety-two-widget-per-hour logic still dictates how we evaluate performance, allocate resources, and structure teams. A 2023 study from the National Bureau of Economic Research found that knowledge workers spend roughly thirty-eight percent of their day on tasks that could be automated or eliminated. Things like data entry, scheduling, routine reporting. Luna: Thirty-eight percent — that's almost half the day. Lucas: Right. And those are exactly the kinds of tasks that fit the ninety-two-widget framework. They're measurable, repeatable, and easy to benchmark. But the sixty-two percent of the day that's left — creative problem-solving, strategic thinking, cross-functional collaboration — that's where real value compounds. And we have almost no tools to measure it. Luna: So we default to measuring what's easy. Lines of code written per developer. Calls handled per agent. Emails sent per hour. Lucas: And those metrics are not just useless for complex work — they're actively harmful. A developer optimizing for lines of code writes verbose, brittle code. A call-center agent optimizing for handle time rushes customers. The incentive structure is misaligned with the actual goal. Luna: There's a famous example from the 1990s at Microsoft. They tried to measure developer productivity by lines of code. The result was a massive increase in code volume and a decrease in quality. They scrapped the metric within a year. Lucas: Right. And that's the same trap. The thing is, the industrial model isn't just about measurement. It's about control. The assumption that management can observe, standardize, and optimize work from the outside. That breaks down completely when the work is inside someone's head. Luna: But companies still try. I see job postings asking for 'ten years of experience' in technologies that have only existed for five. That's a productivity heuristic from an era where experience linearly predicted output. Lucas: Exactly. And the cost of these mismatches is enormous. A 2021 study by the Society for Human Resource Management estimated that U.S. companies waste over $550 billion per year on performance-management systems that don't actually improve output. Most of those systems are built on the ninety-two-widget assumption. Luna: So what's the alternative? If we can't count widgets, what do we measure? Lucas: That's the billion-dollar question. Some firms are experimenting with outcome-based evaluation. Instead of tracking hours or tasks, they assess the impact of a completed project. Valve Corporation, the game developer, famously had no fixed job titles and evaluated employees through peer reviews focused on shipped products, not hours worked. Luna: But that requires a lot of trust and a strong culture. Not every company can pull that off. Lucas: No, and it's not a panacea. But the first step is recognizing that the industrial measurement framework is a historical artifact, not a law of nature. The longer we cling to it, the more we misallocate talent and effort. Luna: It's almost like we need a new productivity metric for the knowledge economy. Something that accounts for compounding value over time. Lucas: Some economists have proposed measuring 'idea productivity' — the rate at which a worker generates insights that lead to improvements. But it's fuzzy. And that fuzziness makes managers uncomfortable. They want numbers. Luna: Which brings us back to the ghost of ninety-two. Lucas: Right. The ghost is still there, sitting in the corner of every quarterly review. And I think if we're serious about raising productivity in services and knowledge work, we have to exorcise it. Luna: I wonder if the listener, like us, has felt that tension — the gap between the measurable and the meaningful. If walking through the economy with us has made something click, or helped you articulate a frustration you couldn't name, that's exactly the kind of value we try to create here. And listener support is what keeps this show ad-free and independent. If you'd like to chip in, you can at buy me a coffee dot com slash fexingo. Lucas: Yeah, it's a small way to help us keep digging into stories like this one. And we genuinely appreciate it. Luna: So back to that ghost — one of the most interesting attempts to banish it comes from a company called Basecamp. They've publicly argued that the best productivity metric is whether you shipped something that moved the business forward in the last six weeks. Lucas: Right, and that's a very different rhythm. Instead of daily or hourly measurement, they use a six-week cycle with a two-week cooldown. No mid-cycle check-ins on progress. Just: did the team deliver a meaningful outcome by the deadline? Luna: And they report that this reduces busywork and increases focus. But it took years to build that culture. Lucas: It did. And that's the hard part. Changing measurement frameworks is not a process change. It's a philosophical change. It requires leaders to trust that unmeasured cognitive work is still productive. That's a leap many aren't willing to take. Luna: So maybe the first step is just acknowledging the ninety-two-widget legacy. Naming it. Lucas: Exactly. Once you name the ghost, you can start designing a measurement system that actually fits the work. And that might be the most productive thing any organization can do this decade.