How to order catch-weight inventory before you know the actual weight

A poultry and seafood distributor running four vans out of a single depot took a standing Thursday order from a chain of twelve restaurants: 180 kilos of whole chicken, split across the week's deliveries. The buying team did what buying teams do with a catch-weight product — they ordered "units," not kilos, because that's what the vendor's order form asked for. Forty birds, going by the vendor's quoted average of 4.5 kilos each, should land close enough to 180. Most weeks it did. The week it didn't, the depot was short fourteen kilos on a Thursday afternoon with no second delivery coming and twelve restaurants expecting their order.

The order form asks for a number that doesn't exist yet

Catch weight breaks the normal purchasing math at the exact point most procurement systems assume it works: you order a quantity, the system multiplies by a unit cost, and you get a total. But a catch-weight item doesn't have a fixed unit weight — a "4.5 kilo average" bird is a planning figure, not a promise. Order forty and you might get anywhere from 160 to 200 kilos depending on the week's flock, the supplier's grading, even the season. The buyer isn't wrong to order in units; that's how the vendor sells and how the order form is built. The problem is treating the average as if it were the actual, and only finding out it wasn't on the day the trucks are already loaded.

For a while the gap just got absorbed. A short week meant someone called around for a top-up order at a worse price, or a restaurant got told their order would be ten kilos light and could they take it as credit next time. A heavy week meant kilos of perishable protein sitting in the cold room with no order to put them against, aging toward a markdown. Nobody treated either as a real problem because neither showed up as a clean number anywhere — it was just "this week ran a bit short" or "we've got extra stock again," said often enough that it stopped sounding like something worth fixing.

The cost compounds because it's never the same number twice

What made it expensive wasn't any single short or heavy week. It was that the business couldn't plan around a figure it could trust. Collections and restaurant relationships took the hit on short weeks — a chain customer who gets told "we're light again" twice in a month starts asking a competitor for a quote. Heavy weeks ate into margin directly: protein that doesn't move in its sell-by window gets marked down or written off, and high-value seafood and poultry carry that cost hard. The owner's actual exposure was never "we don't know the weight." It was "we don't know the weight, and we're ordering as if we do."

The fix wasn't to stop ordering in units — the vendor's system doesn't support that, and renegotiating the order form for one supplier wasn't realistic. The fix was tracking, per vendor and per SKU, what "40 units" had actually weighed the last dozen times it was ordered, and ordering against that measured average instead of the vendor's quoted one. Over three months of orders, this distributor's 4.5-kilo quoted bird was running a real average closer to 4.65 — not a huge gap, but enough that forty units reliably overshot the 180-kilo target, which meant the real fix for most weeks was ordering 38 or 39 units, not 40, and letting the system flag it when a week's incoming weight looked like it was drifting outside the usual range before the truck even left the vendor's dock.

What ordering against a measured average actually requires

• A weight history per vendor and SKU, not a single quoted average. The number that matters is what this specific vendor's "average" bird has actually weighed on your last several orders, which can run meaningfully different from the number printed on their price sheet.
• A target weight, not just a target unit count, on any order where the final invoice will be weight-based. The unit count is how you place the order; the target weight is the number the business actually needs to hit.
• An agreed tolerance band with the vendor — how far the delivered weight is allowed to run from the expected weight before it's a conversation rather than a normal variance. Without one, every week's gap looks the same whether it's routine noise or a vendor quietly shifting their grading.
• A flag before the truck leaves, not after it arrives. If a vendor's confirmed shipment is already trending light or heavy against history, that's the point to adjust a delivery commitment or place a top-up order — not the point to discover it when the scale at the dock gives you the real number.

iotoms tracks actual received weight against every catch-weight purchase order by vendor and SKU, so a buyer ordering "40 units" sees what 40 units from that specific vendor has actually weighed historically — not just the quoted average — and can size the next order against a real number before the truck is loaded.