Your best customer might be your worst margin. The small drop proves it.
In food and beverage, margin does not die in the plant. It dies on the route and on the deduction line. A half-pallet to a convenience store, a low-volume SKU that ties up a chilled slot, an out-of-code return credited back at full value, a promotion that quietly ate the whole margin: a gross-margin view never sees any of it. Time-driven costing prices the drop, the SKU and the customer as they actually behave, after trade spend and after spoilage.
Cost and Profitability Consulting · 150+ models since 2010 · TDABC
In food and beverage distribution, the cost that decides margin is cost-to-serve, not cost of goods. The same case of product can cost roughly two to three times more to deliver to a small convenience-store drop than to a full-truck account, and low-volume or perishable SKUs often lose money once handling, refrigeration, spoilage and shelf space are loaded. TDABC assigns warehouse, cold-chain, loading and delivery time to each drop, SKU and customer, so a flat per-visit cost stops hiding which accounts and products actually pay after deductions.
Margin dies after the product leaves the plant.
Most food and beverage businesses still manage margin from two numbers: cost of goods and a blended logistics rate. Both are averages, and averages are exactly what hide the problem. The cost of goods is broadly accurate and roughly the same for every customer who buys the same case. The margin difference between a profitable account and a loss-making one is almost never in the product. It is in everything that happens after the product leaves the plant: how often you deliver, how big the drop is, whether it needs the cold chain, how much comes back out of code, and how much the buyer deducts before paying.
Small drops cost a multiple of large ones
A flat delivery cost makes the half-pallet to a convenience store look as cheap as the full truckload. Cost-to-serve swings roughly two to three times between drops, and the tail of small, frequent drops is wider still.
Trade spend and deductions hide the real margin
Promotional allowances, slotting fees, off-invoice discounts and deductions sit between list and net. A report built on list price flatters every account that negotiates hard, which tends to be the largest ones.
Spoilage and out-of-code returns erase margin twice
Perishable product that ages past its sell-by date comes back, is credited at full value and is written off. The return handling, the lost sale and the credit never appear in cost of goods.
The cold chain is a cost nobody allocates
Refrigerated storage, chilled loading and temperature-controlled delivery cost real money, but a flat per-case rate charges the ambient and chilled SKU the same. Cold-chain SKUs are quietly subsidised by everything else.
THE SAME CASE, TWO DROPS
Illustrative. The same product on both trucks. Everything that makes the convenience-store drop expensive happens after the case leaves the plant, where the blended rate refuses to look.
Two parameters, no surveys.
A capacity cost rate per resource (warehouse, loader, chilled bay, driver), and time equations that describe how each drop, SKU and customer consumes time. The events are already in the data: a food and beverage business knows how many cases went on which truck, how many stops the route made, which products were chilled, and how much came back. What it does not do is connect those events to a cost and then to a customer and a SKU. As an illustrative scale, a regional dairy ran roughly 250 time equations across warehouse, loading and delivery.
Drop cost = travel time to the stop x driver capacity cost rate + 90 sec fixed park-and-unload per stop + 20 sec per chilled stack loaded + 5 sec per loose unit + return handling time x out-of-code units
Illustrative delivery time equation. The conditional terms, chilled stacks and out-of-code returns, are where two drops of the same case diverge.
A DROP IS BUILT, NOT AVERAGED
Illustrative. Each term is an event the business already records. The blended rate throws them away; the equation keeps them and lands the cost on the drop that caused it.
A steep whale curve, and the deduction line cuts deepest where volume looks best.
The whale curve in food distribution is steep: a top tier of full-truck accounts carries the business while a long tail of small, frequent, high-return drops erodes it. As an illustrative sector pattern, a regional dairy put TDABC behind its commercial decisions and acted on what it found. One account accepted a double-digit price increase, dropped two low-volume SKUs and moved to full trucks, worth a six-figure annual gain. A small inventory discount given to a self-managed convenience-store account cut out-of-code returns to almost none and saved six figures more. The business closed a sub-scale plant and removed a weekly shift once it could see which routes and products actually paid. A separate national food company with tens of thousands of customers found cost-per-visit ranging from very low to more than ten times that figure once fully costed, against a flat average under traditional ABC, an entire layer of margin inversion the averaged number had hidden.
The pattern repeats across the sector because the drivers are structural, not company-specific. A business that delivers perishable product, on a fixed route, to a mix of large and small accounts, against negotiated terms, will always find the same shape once it measures honestly: a profitable core, a thin middle, and a long tail of small, frequent, high-return, high-deduction relationships that the flat rate has been quietly underwriting. The value of the analysis is not the shock of the chart. It is that the chart turns into a list of specific, fundable actions: which accounts to renegotiate, which SKUs to retire, which routes to consolidate, which drops to reprice, which plants or shifts no longer carry their cost.
CUSTOMERS, RANKED BY TRUE MARGIN
Illustrative. A profitable core carries the business; a long tail gives much of it back once cost-to-serve and deductions are loaded. Some of the largest accounts sit in the inverted zone: low net price, high service.
Rich data, averaged away before it reaches a margin.
Food and beverage distributors are typically strong on Data & Technology, route, warehouse, order and temperature systems already capture the events, but weak on Cost Allocation and Profitability Visibility: the delivery, return and refrigeration events are logged, but they are never turned into a true cost per drop, SKU and customer. The data exists. It is averaged away before it reaches a margin. The sector is read qualitatively across the seven dimensions here; no numeric survey score is implied.
AI cuts the waste. Only if you priced the drop first.
AI is moving into the exact places where food and beverage margin leaks: demand forecasting on perishable stock to cut waste and out-of-code returns, cold-chain and route optimization to lower cost per drop, dynamic pricing on short-dated product, and yield optimization in production. None of it pays unless you already know the real cost of a drop, a SKU and a customer, because that is the baseline every model has to beat. A forecasting model that cuts spoilage saves nothing you can prove unless the spoilage already carried a cost on the SKU it belonged to. The cost picture is the scoreboard the AI plays against; without it, every claimed saving is a guess.
Six ways into the sector's cost.
Frequently asked questions
- How do you calculate the true cost to serve a food and beverage customer?
- Load warehouse, cold-chain, loading, delivery and return handling time onto each drop with time equations, not a flat per-visit average, and then subtract trade spend and deductions. The same case can cost roughly two to three times more to a small drop than to a full truck.
- Why are some food and beverage customers unprofitable?
- Small frequent drops, high out-of-code returns, heavy trade spend and aggressive pricing on large accounts. A flat cost-per-visit hides the inversion; TDABC exposes it.
- What is TDABC for food and beverage?
- Time-driven activity-based costing assigns cost via a capacity cost rate and time equations, for example pallet checks and chilled cooler loading, giving true per-drop, per-SKU and per-customer cost without timesheet surveys.
- How do low-volume or perishable SKUs hurt margin?
- They consume picking time, chilled slots and shelf space out of proportion to volume, and perishables add spoilage and out-of-code returns on top. Fully costed, many lose money.
See which drops, SKUs and accounts actually pay.
ProfitAudit 360 builds the per-drop, per-SKU, per-customer picture on a TDABC base. Or take the Profit Check first: five minutes, no data upload, and it points to where your blended rate is most likely wrong.