Why Your Business Stays Overstaffed Even When You're Losing Money
I spent about six years managing a mid-sized retail operation before I started looking into what Zeynep Ton was writing about. We were doing the textbook thing — squeezing every operational mile out of lean staffing, tracking every minute, running tight schedules. And yet we kept losing money while our turnover sat around 80 percent annually. That's when I actually read through The Good Jobs Strategy How Smartest Companies Invest In Employees To Lower Costs And Boost Profits Zeynep Ton and tried applying some of it. Took me another two years before the numbers started moving in the right direction. Ton's central argument is straightforward but goes against everything most operations people are trained to do. Instead of minimizing labor cost as a line item, treat labor investment as a cost-reduction lever. The companies she profiles — Costco, Southwest, Harrah's, REI, and a few others — all have one thing in common. They deliberately pay above market, cross-train heavily, reduce managerial layers, and give frontline workers meaningful decision-making authority. The result isn't just better morale. Turnover drops, productivity per hour climbs, shrinkage decreases, and customer spend goes up. The math works out because the alternative — constant recruiting, constant retraining, constant mistakes — is quietly more expensive than anyone wants to put on a spreadsheet. Here's the part people miss. The strategy isn't about spending more on people. It's about removing the operational complexity that makes cheap labor expensive in the first place. Most companies are running broken systems and expecting broken workers to compensate. That's backwards.
The Three Levers You Actually Need To Pull
Ton identifies three interconnected elements. Simplify operations. Optimize the use of labor. Invest in employees. They're not optional add-ons. They're sequential dependencies. Start with simplification. Look at your standard operating procedures and figure out which steps exist only because someone left the company three months ago without documenting what they were doing. I found about 40 percent of our task variety came from workarounds that had hardened into policy. Removing that variety is where the first real savings live. You don't get there by buying new software. You get there by standing on the floor and watching what people actually do versus what the handbook says they do. Then comes labor optimization. This is where most people trip up. Labor optimization in the Good Jobs sense doesn't mean fewer hours. It means fewer skill gaps, fewer handoffs, fewer decision bottlenecks. Cross-train people so the person checking out a customer can also handle a return without paging a manager. Design work so that problems get solved where they happen instead of routed up a chain of command. At my old place we eliminated about seven approval levels for routine customer issues. A $25 adjustment that used to take 45 minutes of back-and-forth between three departments now takes one person fourteen seconds. That's the difference the strategy measures.
The third lever is investment. Better pay attracts people who stay longer. Longer tenure builds institutional knowledge that replaces supervision. Better hiring practices mean you're not constantly onboarding people who will quit in ninety days. The ROI here is real but it's lumpy. You'll see the cost hit immediately. The savings come gradually through reduced turnover, reduced errors, and higher throughput. Budget accordingly. If you cut the investment before the optimization proves itself, you're just spending more on the same broken system.
Get the Full Details
What Actually Happens When You Try This
I'll be honest about the friction. The first six months feel terrible. You're raising wages and the P&L gets worse before it gets better. Your finance team will ask questions. Your investors or board might push back. That's normal. The model Ton describes requires a longer measurement window than most quarterly reporting cycles support. I learned this the hard way after presenting a six-month results report to our CFO who wanted to pull the plug. I rebuilt the dashboard to track three things instead of the usual twelve: turnover rate, average handling time per transaction, and customer transaction value. Those three metrics started improving within month four. The financials caught up by month nine. There's a specific edge case I keep running into. People try to implement this in environments where task variety is genuinely high and unpredictable. Healthcare, emergency services, custom fabrication. The Good Jobs framework assumes a certain baseline of repeatable work where simplification and cross-training are possible. I saw a clinic director try to apply it wholesale and nearly break scheduling because the clinical workflow couldn't be simplified the way retail checkout can. The workaround was to apply it only to the administrative and customer-facing layers while keeping clinical roles separately optimized. Don't force the model onto work that doesn't fit it. Pick the parts of your operation that are actually repetitive and start there.
The Counterintuitive Part Nobody Talks About
The biggest insight from Ton's work that beginners consistently overlook is that reducing employee autonomy actually increases your operating costs. It sounds backwards. You'd think more control means more efficiency. But when frontline workers can't make decisions, every minor exception becomes a manager intervention. Manager time is expensive. That's why Harrah's gave dealers enough authority to comp meals and drinks without calling a supervisor. The individual bets were small. The aggregate savings from removing the escalation step were enormous. Another thing: this strategy doesn't scale by copying the playbook. It scales by copying the logic. Costco's model won't transfer directly to a twenty-person restaurant. But the logic — simplify the workflow first, then cross-train, then pay well enough that people want to stay and learn — applies at any size. The mistake I see most often is companies trying to replicate the symptoms (higher wages, better benefits) without doing the prerequisite work (simplifying operations). That just makes you expensive without making you efficient.
Where This Strategy Breaks Down
It doesn't work everywhere. Highly commoditized service work with thin margins and no pricing power — think some call center models — struggles to absorb the wage premium even with productivity gains. Pure volume-driven operations where the work is intentionally simplified to the point that skill doesn't matter much also have limited upside. And if your core product or service quality depends on specialized expertise that can't be cross-trained, the labor optimization lever has less leverage. In those cases, the simplification piece might still help, but don't expect the full profitability story. Also, this requires leadership that won't flip-flop every time quarterly earnings look soft. I've watched two separate companies try this, commit for four months, then abandon it when a bad quarter hit. That's the most common failure mode. Not the strategy itself. The willingness to stick with it through the lumpy transition period. If you're starting from scratch and your operation is already complex with high task variety, I'd recommend reading Ton's work and then picking one location or one department to test the model on before rolling it out company-wide. The proof matters more than the pitch. The numbers from companies that actually did this speak for themselves. The question is whether your organization can tolerate the timeline required to see them.
