# How does landed cost change when orders split across warehouses? | LandedCostHQ

Source: https://landedcosthq.com/blog/2026-10-19-landed-cost-multi-warehouse-splits/

Cross-border operations guide · October 19, 2026


When one order ships from two warehouses, landed cost stops being a single number per unit and becomes a puzzle of split freight, duplicated handling, and inventory carried in the wrong place. Brands that model the split honestly price and place stock better.

## Why splits happen and why they cost more than double freight

Orders split when no single location holds everything the customer bought, which is an inventory placement outcome, not a warehouse mistake. The cost is more than two freight charges: two picks, two packs, two labels, often two carriers at worse rate tiers than the combined shipment would have earned, plus the service cost when halves arrive days apart.

Landed cost models built on average freight per order hide all of this. The average looks acceptable while a sizable minority of split orders quietly lose money, usually on multi item baskets, which are exactly the orders the business most wants to encourage.

## Assign split costs to the order that caused them

Cost split orders at their actuals: each leg freight, each location handling, and any expedite used to reunite the delivery experience. Then trace the cause distribution: which product combinations split, which regions split, and which location stockouts drive the pattern. The costing exercise usually pays for itself in the placement changes it provokes.

Resist spreading split cost across all orders as overhead. Spreading feels fair and teaches nothing. Attribution is what turns a costing complaint into a decision about where the next container of a fast line should land.

## Placement is the cheapest split prevention

Most splits come from a small set of products bought together that live in different buildings. Basket analysis finds those combinations. Co locating the top companions, or holding a deliberate complete basket buffer at the highest volume site, removes the majority of splits without a network redesign.

New product placement should inherit the lesson: assign a line a home location based on what it sells with, not just its own volume. A modest seller that completes a hundred baskets a day is a placement priority out of proportion to its own sales.

## When splitting is still the right answer

Sometimes the split is correct: the alternative is a stockout, a long backorder, or transferring stock between warehouses at greater cost and delay than the second parcel. The mature position is not zero splits. It is splits chosen knowingly, with the cost visible at the moment of the routing decision.

Routing systems can be taught the trade off: split automatically below a cost threshold, hold and consolidate above it when the delivery promise allows, and surface the expensive exceptions for human review. What finance needs is the threshold written down and the exceptions logged, not perfection.

## Report split rate as a landed cost KPI

Track split rate, average extra cost per split order, and split rate by product combination and region, reviewed beside freight cost per order. A rising split rate is an early warning that inventory placement and demand geography are drifting apart, months before it shows up as a freight budget crisis.

Fold the finding back into buying and allocation meetings. Split cost is one of the few numbers that connects warehouse geography, basket behavior, and margin in a single line, which makes it unusually good at starting the right argument.

Keep reading: [How should brands pass tariff increases into prices without losing the customer?](https://landedcosthq.com/blog/2026-10-19-pass-tariff-increases-prices), [How do freight surcharges and accessorial fees distort true landed cost?](https://landedcosthq.com/blog/2026-10-19-freight-surcharges-accessorial-landed-cost), and [Why should returns costs be counted inside true landed cost per order?](https://landedcosthq.com/blog/2026-10-17-returns-costs-true-landed-cost-order) cover the nearby cases in this series.

## Frequently asked questions

## Why does a split order cost more than two freights?

Because the order also pays two picks, two packs, two labels, worse rate tiers on each smaller parcel, and sometimes an expedite to close the delivery gap. Average freight per order models hide the loss, which concentrates in multi item baskets.

## How should split costs be recorded?

At actuals on the order that split, then analyzed by cause: product combinations, regions, and stockout locations. Spreading the cost as general overhead feels fair but produces no placement decisions.

## What prevents most splits?

Basket driven placement: co locating products that are bought together and holding complete basket buffers at high volume sites. Most splits trace to a small set of combinations, so the fix is targeted rather than a network redesign.

## Is zero splits the goal?

No. Sometimes splitting beats a stockout, a backorder, or an inter warehouse transfer. The goal is splits chosen with their cost visible, under a written threshold, with expensive exceptions reviewed.