Latest / The Operator's Manual with Fexingo: Daily Decisions, Tactical Wins, and Business Execution / How One Warehouse Cut Picking Errors 80 Percent With One Zone Change
Transcript
- Lucas: So we've talked a lot about single changes that save money or time. But today I want to look at quality — specifically, picking errors in a warehouse. Luna: Oh, those are brutal. Wrong item, wrong size, wrong colour — and then the return costs pile up. Lucas: Exactly. One mid-sized e-commerce warehouse — let's call them PackRight — was running about a 6.8 percent error rate. That's roughly 68 wrong picks out of every thousand orders. They decided to try something surprisingly simple. Luna: What was the fix? Lucas: They switched from individual order picking to zone-based batch picking. Instead of one picker walking the whole warehouse to fill a single order, they divided the floor into five zones. Each picker only works inside their zone, picking items for multiple orders at once. Luna: So you trade the 'one person owns the whole order' model for a relay race. Each zone passes the tote to the next. Lucas: Right. But the key was how they reorganized the batches. They used a simple algorithm — nothing fancy, just a spreadsheet — to group orders that had items in the same zones. That cut the number of zones touched per batch from an average of 3.4 to 1.6. Luna: Meaning less handoffs, less confusion. Lucas: Exactly. And the results were dramatic. Within three months, their error rate dropped to 1.2 percent. That's an 82 percent reduction. Throughput per picker went up 22 percent, and training time for new hires fell by about a third. Luna: Training time? That's interesting. Why would zone picking make training faster? Lucas: Because new pickers only have to learn one area. They don't need to memorize the whole warehouse layout — just their zone. The old system required knowing where everything was, which took weeks. Now you can train someone in three days. Luna: Makes sense. But I can imagine some veteran pickers pushing back — they're proud of knowing the whole floor. Lucas: That was the biggest hurdle. Some of the top pickers felt demoted. They'd been the fastest because they knew the shortcuts. Zone picking equalizes everyone a bit. PackRight had to run pilot shifts and let veterans be zone leads to get buy-in. Luna: So you need to redesign the roles, not just the layout. Lucas: Exactly. They created a 'zone captain' role for the most experienced pickers — they handled exception items and restocking within their zone. That gave them ownership. And the company shared the savings: a small hourly premium for hitting accuracy targets. Luna: What about the software side? Did they need a new warehouse management system? Lucas: No. They already had a WMS that could do batch assignments. The tweak was in how orders were grouped. They added a rule: don't batch orders that need more than two zones. That forced the algorithm to create smaller, cleaner batches. It took their IT guy about two days to write. Luna: So the real work was on the floor, not in the code. Lucas: Right. They also added a sortation station at the end of the zones. Because when you batch pick, you have to separate which item goes to which order. That was new — they had to buy a simple tilt-tray sorter, but they used a refurbished unit. Cost about $15,000. Luna: And the payback? Lucas: They calculated the annual cost of returns and reshipments from picking errors at about $340,000. After the change, that dropped to roughly $60,000. So the sorter paid for itself in less than a month. The whole change — including the hourly premiums and training — had a payback period of about six weeks. Luna: That's a no-brainer for most operations. Why don't more warehouses do this? Lucas: I think a lot of managers think zone picking only works for huge operations — Amazon-scale. But PackRight had about 40 pickers and 10,000 SKUs. It's scalable down. The principle is: reduce travel distance and cognitive load. That applies whether you have 10 pickers or 10,000. Luna: Cognitive load is a good way to put it. When a picker has to remember 20 orders at once, errors spike. Lucas: Zone picking reduces the number of orders a picker juggles. In PackRight's case, each picker handled at most five orders per batch, but all in one zone. Their short-term memory wasn't overloaded. And because they stayed in one area, they developed muscle memory for where items were. Luna: That reminds me of something I read about human spatial memory — we're really good at remembering locations if we stay in one place. Moving around actually degrades recall. Lucas: That's exactly the science behind it. The warehouse layout basically becomes a mental map. Zone picking leverages that. The old system forced pickers to constantly reorient. Luna: Okay, so the zone change is the hero. But what about the batches — how did they decide which orders to group? Lucas: They used a simple heuristic: prioritize orders with items concentrated in the same zone. If an order had five items and four were in zone A and one in zone C, they'd batch it with other A-heavy orders and handle the C item separately — maybe as a secondary pick. They called it 'zone affinity batching'. Luna: Did they track metrics after the change? Lucas: Yeah, they monitored three: error rate, pick rate per hour, and batch completion time. Error rate plunged immediately, pick rate climbed steadily over the first month as people got used to it, and batch time stayed flat — which surprised them, because they expected it to drop. But since they were grouping orders differently, the batch size changed. Luna: So the overall throughput went up even though individual batch times didn't budge. Lucas: Exactly. More orders per hour overall because each batch contained more customer orders. They went from an average of 2.4 orders per batch to 4.1. So even if each batch took the same time, they were shipping more orders. Luna: That's a great example of a metric that doesn't tell the whole story. Batch time alone would have made the change look neutral. Lucas: Right, you have to look at throughput. They also looked at 'first-pass yield' — orders shipped without any manual intervention. That went from 88 percent to 96 percent. Luna: So fewer orders needing a quality check before shipping. That saves labor too. Lucas: It does. Their audit team — people who double-check orders before sealing — shrank by half. Some were reassigned to the sortation station, so no layoffs. But overall headcount need dropped. Luna: I like that they didn't just fire people. They redeployed. Lucas: That's smart change management. And it helped with morale. The sortation role was actually less stressful than auditing — you're just scanning items onto a conveyor, not hunting for mistakes. Luna: What about the other 1.2 percent of errors that remained? Any patterns? Lucas: They found that most remaining errors happened at the sortation station — items got misdirected to the wrong tote. So they added a barcode scan at that point. That cut errors further to 0.4 percent. Luna: So the zone change was step one, then a secondary fix addressed the new bottleneck. Lucas: Exactly. That's the iterative approach. They didn't try to fix everything at once. They changed the picking method, measured, then fixed the sortation. Luna: It's a good reminder that one change often reveals the next problem. Lucas: Speaking of which — if these conversations have moved your work forward in some small way, a handful of listeners chip in monthly through buy me a coffee dot com slash fexingo. That's literally what funds making this many episodes. No ads, no sponsors — just listener support. Luna: Yeah, it's a small group that keeps it going. And we appreciate that more than we can say. Lucas: So back to PackRight — I think the bigger lesson here is about boundaries. They drew a physical boundary — a zone — and that changed behavior. Luna: Right. Boundaries in operations often reduce complexity. A clear boundary means less decision-making, less walking, less error. Lucas: I'd love to see more companies experiment with zone-based approaches. Not just in warehouses — think hospitals, restaurants, even software teams. Where can you draw a line that simplifies the work? Luna: That's a good question to leave listeners with. What's one boundary you could draw this week? Lucas: Exactly. For The Operator's Manual, I'm Lucas. Luna: And I'm Luna. We'll see you next time.