The factory had everything.
ISO 9001 certification. A spotless quality manual. Calibrated machines. Documented procedures. Traceability systems. Every box a supplier audit could tick was ticked. The client, a Scandinavian furniture company, had done their due diligence. They were confident.
Then the defects started.
Not many—just a steady trickle. 2% here, 3% there. Nothing catastrophic, but enough to erode margins and frustrate customers. The factory's quality reports showed no pattern. The client's own inspections found nothing consistent. The problem was real, but it was invisible to every system designed to catch it.
I spent two weeks on the floor, watching, listening, asking questions no one had asked.
The answer emerged slowly. The machine operators—the people who actually ran the injection molding presses—had been with the factory for an average of eighteen months. Before that, they'd worked somewhere else. Before that, somewhere else. Turnover was high, training was minimal, and the experienced operators who'd run this job successfully for years had all moved on.
The new operators followed the procedures. They set the parameters as documented. But they didn't know the small things—the subtle adjustments that come from feel, the warning signs that aren't in the manual, the sound of a machine that's slightly off. They did everything right, but they didn't do everything well.
The systems were perfect. The people weren't. And the systems couldn't compensate.
A story that stayed with me: a factory that made complex optical components for a European client. The rejection rate was consistently low—under 1%—and the client was happy. Then the rejection rate doubled. Then tripled. The factory's engineers were baffled. Nothing had changed.
I interviewed the production manager, the quality team, the shift supervisors. Finally, someone mentioned that the senior technician on that line had retired three months earlier.
That technician had run that line for twelve years. He knew exactly how to tweak the parameters when the humidity changed, when the material batch varied slightly, when the machine started to drift. He'd never written any of it down—it was just what he knew, what he did, what worked.
When he left, the knowledge left with him. The procedures captured the "what" but not the "how." The new technician followed the manual, but the manual didn't include the adjustments that came from twelve years of experience.
Landmine #1: Knowledge lives in people, not manuals. When those people leave, the knowledge leaves too—and you may not notice until the defects start.
A US consumer goods client had a puzzling pattern: quality was consistently good during the day shift, but the night shift produced twice as many defects. Same machines, same material, same procedures. The factory's quality reports didn't distinguish by shift, so the pattern was invisible in the data.
I stayed late to watch the night crew.
The difference was subtle but consistent. The night shift operator was younger, less experienced, and unsupervised. When the machine made a strange sound, he ignored it—he didn't know what it meant. When the parts looked slightly off, he kept running—he assumed they'd be caught later. When he got tired at 3 AM, he let his attention drift.
The day shift operator, by contrast, had been there eight years. She knew every sound, every variation, every trick. And the production manager was on the floor during the day, watching, available, present.
The procedures were identical. The outcomes were not.
Landmine #2: What happens when no one is watching is what actually determines your quality. Audits and certifications capture the ideal. The night shift captures the real.
A factory in Guangdong had a simple incentive system: operators were paid a bonus based on the number of parts produced. More parts, more money. The system worked beautifully—output soared.
Then quality collapsed.
The operators, rationally, optimized for what they were measured on: quantity. They ran machines faster than recommended. They skipped maintenance to save time. They let borderline parts pass rather than stopping to adjust. The bonus system, designed to motivate, had instead created a culture of speed over quality.
The factory's quality team fought a losing battle. Inspectors flagged defects; operators kept producing them. The tension between production and quality became embedded in every shift, every decision, every part.
The client, reviewing reports from afar, saw only the rising defect rate. They blamed the factory. But the root cause was the incentive system—a system they'd never known existed.
Landmine #3: People do what they're paid to do. If your supplier's incentives reward speed over quality, you'll get speed—and the quality problems that come with it.
A French industrial client required all operators on their line to complete a specific training program. The factory confirmed, in writing, that the training was done. Certificates were provided. The box was ticked.
I asked to see the training materials. The factory produced a binder—a translation of a generic injection molding manual, printed five years ago, never updated. I asked to speak with an operator who'd completed the training. She showed me her certificate. She also showed me that she couldn't explain the basic principles of the process she ran every day.
The training had happened—if you defined "training" as sitting in a room while someone read from a binder. But learning hadn't happened. The operators had certificates. They didn't have understanding.
Landmine #4: Training is not the same as learning. A certificate proves attendance, not competence. And when operators don't understand the process, they can't adjust when things go wrong.
A Southeast Asian electronics assembler had a quality issue that kept recurring. Every few months, a batch would fail. The factory would investigate, find a cause, implement a fix. Then, a few months later, the same problem would reappear.
I dug into the history and found a pattern: the problems always correlated with the absence of a particular quality manager. When he was on leave, when he was sick, when he was busy with another client—those were the times defects slipped through.
The manager was the only person who truly understood the process. He'd built the quality system himself, trained the operators, knew every potential failure mode. But he'd never documented his knowledge, never trained a backup, never built a system that could function without him.
When he was there, quality held. When he wasn't, it drifted. The factory had a quality system that depended on one person—and one person is not a system.
Landmine #5: If quality depends on a person, it's fragile. People get sick, take vacations, leave for other jobs. Resilience requires that knowledge be distributed, not concentrated.
But there's another side to this story.
The technician who caught the loose screw before it destroyed a mold—the one I mentioned in an earlier chapter—he wasn't following a procedure. There was no checklist that said "listen for faint clicking sounds." He was simply paying attention, because he cared, because someone had taught him that his job wasn't just to run the machine but to protect it.
I've met many operators like him over the years. They're the ones who notice when a mold is running slightly hotter than usual, who catch a discoloration before it becomes a batch of scrap, who stop the line to ask questions when something doesn't feel right. They're not following a manual—they're using judgment, experience, care.
And they're the difference between a factory that delivers consistent quality and one that doesn't.
The question for any brand sourcing from China is not whether your supplier has certified operators. It's whether those operators are the kind who notice, who care, who act. And that question cannot be answered from a distance.
In a traditional market, a good buyer knows the people behind the product. Not just the owner or the manager—the ones who actually do the work. They know which butcher cuts meat carefully and which rushes. They know which farmer delivers consistently and which cuts corners. They build relationships not just with businesses, but with the people inside them.
This isn't sentiment—it's intelligence. Because systems don't make quality. People do. And when the people change, quality changes with them.
The question is not whether your supplier has good systems. The question is: who is watching the people who run them?
How We See the Human Layer
This is where the CPO model extends beyond systems to the human layer. Not to replace your supplier's HR function, but to add a dimension of observation—to notice when key people leave, when training is superficial, when incentives drive the wrong behaviors, when the night shift operates differently from the day shift.
Through Production Process Monitoring, we don't just audit systems. We watch the people who make them work. We identify the key operators, technicians, and managers whose knowledge and care are critical to your quality—and we monitor when they change. We visit during nights and weekends to see what happens when no one is watching. We ask about incentive systems and flag misalignments. And when critical people leave, we work with suppliers to ensure their knowledge is captured and transferred before quality drifts.
We don't just audit suppliers. We get to know the people inside them—because in the end, people matter more than procedures.
Learn more: Production Process Monitoring | Supplier Capability Audit
The Scandinavian furniture client whose 2% defect rate defied explanation? We traced it to operator turnover. The factory had lost its experienced crew and replaced them with untrained newcomers. The systems were intact. The knowledge was gone.
We worked with them to rebuild: structured training, mentorship from remaining veterans, better incentives to retain experienced operators. It took time. But over the following year, the defect rate dropped back to its historical level—and stayed there.
The client still audits the factory. But they also ask different questions now: Who's running the night shift? How long have the operators been here? What happens when a key person leaves? They've learned that the people matter as much as the paperwork.
In the next chapter of The CPO Chronicles, we'll explore what happens when all these layers—material, tooling, production, coordination, prevention, tier-two, human—come together. We'll look at how the CPO model creates a single point of integration in a fragmented world, and why having one person who sees the whole picture changes everything.
If you've ever felt like your supply chain is a collection of pieces that don't quite fit, the next installment is for you.
Alex Yi
Founder, WELL BEST | Your Chief Plastic Officer
Follow The CPO Chronicles: wellbestasia.com/info/cpo-chronicles
