The sentence we hear most often at the start of an engagement is some version of: “It’s pretty simple, a few SKUs, one supplier, nothing too complicated.”
Half a day into the actual discovery, the picture is different. Not because the client was misleading us. They genuinely believed what they said. They just don’t have a complete map of their own operation.
The Our Place version
The Our Place COO described an e-commerce and DTC business with premium cookware, multiple sales channels, and some international distribution. That’s accurate as far as it goes.
What became clear over the first weeks of the engagement: seven warehouses spanning California, Texas, Pennsylvania, Illinois, Ontario, Colton, UK, and Australia. Approximately 1,000 containers per year. FBA supply limits that cap sends at 5,000 units and behave differently across marketplaces. An oversized-items classification with its own storage, routing, and limit dynamics. A UK marketplace that ran out of standard-size storage capacity. Australia that couldn’t create FBA shipments properly at Prime Day.
And then, in the first weeks: a 170% tariff on Chinese-origin appliances that made the existing supply chain for the Wonder Oven and Dream Cooker lines economically unviable overnight.
None of that came up in the initial “tell me about your supply chain” conversation. It wasn’t hidden. It was simply more than anyone had tried to explain in a single sentence.
The Realsy version
“Food brand, date-stuffed snacks, a few retail accounts, Amazon, DTC.”
That sentence contains a supply chain that took months to fully map. Manufactured in Mexico, which means customs clearance on the date side. Two separate nut butter suppliers (peanut, chocolate peanut, and almond butter from Camel Foods) each with their own lead times. Multiple pack sizes creating a SKU matrix that doesn’t simplify the planning. 3PLs at RAD 3PL and Golden Prep. Wholesale EDI relationships with UNFI, Jewel Osco, Target, DoorDash, Sunset Foods, and independent accounts through AirGoods, each with different PO acknowledgment requirements, payment terms, and freight scheduling contacts.
The tech stack: Cin7, Extensiv, CartRover, Crystal for wholesale, AirGoods, Bill.com, Shopify. Six platforms that need to talk to each other, with an integration layer that was configured wrong in a way that was blocking all POs created against backordered inventory.
The production dependency on almond butter, where a delay from a single supplier blocks an entire product line’s manufacturing run, was not something the client flagged as a risk. It was a structural vulnerability that only became visible when we built out the full supply picture.
The French fashion brand version
A smaller brand, sourcing from China via the 1688 platform, described their supply chain as: “We have a warehouse, we ship from there, lead time is about 7 to 10 days from China.”
What the discovery call surfaced: the 7 to 10 day lead time was the best case. The effective lead time, accounting for supplier reliability issues on 1688, customs clearance variance, and carrier delays, was considerably longer. More significantly, their order management system was configured to block an entire multi-item order if any single SKU in that order was out of stock. Not split the order. Not flag it for partial fulfillment. Block the whole order.
A customer ordering three products, where one SKU was on backorder, got nothing. The stockout impact on that one SKU cascaded across every multi-item order that included it. That’s not an inventory problem. It’s a systems configuration problem, and it wouldn’t have surfaced in a standard inventory conversation.
Why clients describe their supply chain incorrectly
They’re not lying. They’re describing the ideal-state process, the version where everything works as designed. The actual operational reality includes the workarounds, the exceptions, the manual interventions that have become so routine that nobody thinks to mention them.
The gap between “how it’s supposed to work” and “how it actually works” is where almost all the complexity lives. That’s also where almost all the risk is.
The questions that surface actual complexity are different. Instead of “tell me about your supply chain,” you ask: when did you last have a stockout, and why? Which supplier has the most unreliable lead time? What happens in your system when you create a PO against insufficient inventory? What’s the worst thing that’s happened operationally in the last six months?
Those questions produce different answers. Those answers are the real scoping input.
The practical consequence
Engagements that are scoped against the client’s self-description are always under-resourced relative to actual complexity. The client says simple, the scope is sized for simple, and then the engagement reveals a seven-warehouse global operation with a tariff crisis on day three.
The right approach is to assume the stated complexity is the floor, not the ceiling. Build discovery questions that are specifically designed to surface the gap. Don’t quote an engagement until you’ve asked about recent operational failures, multi-system integrations, international suppliers, and unusual channel structures.
The supply chain is almost always more complicated than the sentence it gets summarized in. The job is to find out how much more complicated before you’ve committed to a scope that doesn’t account for it.
