Handled, or Just Assumed? A Question for Supply Chain Leaders

At Multimodal this year, I watched the same thing happen more than once. A supply chain director who had told us for years that customs was handled, thank you, sat in front of CAT360, our customs analytics platform, watching their own import data laid out line by line, and went quiet. Not because a number was alarming. Because they had never seen it before. 

There is a name for what was happening in that silence. In 2002 two Yale psychologists, Leonid Rozenblit and Frank Keil, handed people ordinary objects, a zip, a bicycle, a flushing toilet, and asked them to rate how well they understood how each one worked. Confidence was high. Then they asked the same people to write out, step by step, how the thing worked. Confidence collapsed. People discovered mid-sentence that they did not understand the object they had just rated themselves an expert on. Rozenblit and Keil called it the illusion of explanatory depth, and it has a specific habitat. It does not attach to facts you can recite or stories you can tell. It attaches to complex systems with hidden, interlocking parts, the kind you touch at the surface and assume you understand underneath. Customs is close to a laboratory example. Classification, valuation, origin and procedure feed into one another through layers most people never see, while the surface, goods arrive, duty is paid, the shipment clears, keeps working. 

For a supply chain leader, that gap is expensive. The distance between the control you assume and the control you can evidence is exactly where landed cost leaks, where clearance delays incubate, and where one misclassified line repeats across thousands of entries before anyone notices. You cannot manage what you have never looked at, and the illusion guarantees that looking feels unnecessary right up until the moment you do. 

The useful part of the Yale finding is what breaks the illusion. Not a warning, not a statistic, not being told you might have a problem. The one thing that deflates false confidence is being made to explain the mechanism in detail. Philip Fernbach and Steven Sloman later showed the same effect well beyond the lab: ask people to walk through, step by step, how something works, and overstated certainty falls away on contact with the detail. This is why the reaction at Multimodal was so consistent, and why the strongest reactions came from the most experienced people in the room. Seeing your own customs data at the line-item level, across years rather than a single snapshot, is that explanation test made real. It is the difference between believing the machinery works and watching each part move. The experienced leaders were not caught out because they knew least. They were caught out because they had the most confidence to lose. 

Then there is the sentence we hear most often: we have someone in-house who handles all this. It sounds like control. Under the research it points the other way. Expertise does not switch the illusion off. Decades of calibration studies, pooled across more than two hundred experiments, find that when experts say they are 80% sure, they are right closer to 65% of the time, and the effect is most stubborn among specialists. A capable in-house customs person is necessary. They are not, on their own, visibility, and they are often the reason nobody feels the need to look. 

None of this says supply chain leaders are careless. Quite the reverse. The illusion of explanatory depth is not a lapse in diligence, it is a standard feature of how confident, capable people relate to complex systems, and the professionals most exposed to it are the ones who have kept things running for years. The only move that changes anything is an unglamorous one: stop assuming the machinery and look at it, line by line, in your own data. Assumed control and evidenced control feel identical from the inside. Only one of them holds up when you read it back. 

 

Would you like to see your own customs data presented this way? CAT360 lays it out line by line, at the level where cost and risk actually sit. That is where assumed control becomes evidenced control.