Shipper Freight Spending Rose 28.1% While Volume Fell: Fix the Cost-to-Volume Disconnect

Freight volume can fall while the transportation budget climbs sharply. That apparent contradiction is now showing up in shipper data: second-quarter spending increased 28.1% year over year even as shipment counts declined. The useful question is not whether freight became more expensive. It is which combination of rates, shipment characteristics, service decisions, and exceptions produced the increase.

Averages alone will not answer it. Cost per shipment can rise because carriers charge more, but it can also rise because freight moved on longer lanes, used faster services, required more special handling, or accumulated preventable accessorials. A transportation management system should separate those effects before procurement renegotiates rates or operations cuts service.

Lower volume no longer guarantees lower spend

FreightWaves reported that tighter truck capacity pushed shipper spending 28.1% higher year over year in Q2 2026 despite falling volume. That gap matters because it signals more than routine inflation. If the number of shipments falls but total spending rises by more than a quarter, the cost attached to the remaining freight has changed materially.

Market conditions support part of that explanation. Carriers have regained pricing leverage as available capacity contracts. Logistics Management noted that the pricing pendulum shifted toward truckers during 2026, leaving shippers with tougher conditions than they faced during the previous three years. Yet market rates are only one layer of the variance.

Even mode-level benchmarks require care. A separate FreightWaves analysis found that LTL cost per shipment rose 0.7% sequentially in Q2 while weight per shipment fell 4.8%, with fuel costs contributing to the increase. That example shows why a single cost-per-shipment KPI can conceal a changing shipment profile.

Build a freight-cost bridge

Instead of comparing two total-spend figures, build a bridge from the prior period to the current one. Assign every dollar of variance to a defined driver:

• Volume: the spend that would have changed if shipment count moved but every other input stayed constant.
• Base rate: changes in contract, spot, minimum, and deficit-weight charges on comparable lanes and services.
• Mode and service mix: shifts among parcel, LTL, truckload, intermodal, expedited, and guaranteed products.
• Lane and distance mix: a greater share of long-haul, imbalanced, rural, or capacity-constrained movements.
• Shipment profile: changes in weight, cube, pallet count, freight class, density, or hazardous-material requirements.
• Fuel and accessorials: fuel surcharge, detention, layover, redelivery, liftgate, limited-access, storage, and other add-ons.
• Service failure: premium recovery moves, missed appointments, reconsignment, and duplicate transportation caused by an exception.

The categories must be mutually exclusive enough to prevent double counting. For example, an expedited replacement shipment triggered by a missed pickup belongs under service failure, not ordinary mode mix. A fuel surcharge belongs in fuel even when it appears on an LTL invoice with the linehaul charge.

Once the bridge is built, teams can act on the responsible layer. Procurement owns an unexplained base-rate increase. Network design owns a costly lane shift. Shipping sites own avoidable detention and reweigh fees. Customer-service policy may own excessive expedites. The bridge turns a budget argument into a set of operational decisions.

Why common averages can mislead

Cost per shipment is useful only when the shipment population is stable. If average weight falls 10% while cost per shipment remains flat, the business is paying more for each pound moved. If the network adds short-haul deliveries, cost per mile may rise because fixed pickup and delivery costs are spread across fewer miles—even if carrier pricing did not deteriorate.

The reverse can also happen. Cost per mile may improve because the lane mix shifted toward longer trips, while total cost and customer lead time both worsened. A blended metric across modes is even less reliable: parcel, LTL, truckload, and intermodal have different pricing units and cost structures.

Use matched comparisons whenever possible. Compare the same origin-destination pair, mode, service level, equipment, weight band, and customer requirement. Then report both the matched-lane result and the effect of mix changes. Useful normalized measures include cost per hundredweight for LTL, cost per loaded mile for truckload, cost per package by zone and service for parcel, and total landed transportation cost per order.

The TMS fields that explain budget variance

The analysis depends on clean shipment, tender, event, and invoice data. At minimum, capture:

• shipment ID, order ID, origin, destination, lane, planned distance, and actual distance;
• mode, service level, equipment, carrier, contract ID, and spot or contract designation;
• pieces, pallets, weight, cube, freight class, and billed-versus-tendered measurements;
• planned pickup and delivery windows plus actual event timestamps;
• tender attempts, acceptance response, rejected carrier, and recovery carrier;
• base charge, fuel, each accessorial code, tax, discount, and final approved amount;
• reason codes for expedites, reconsignments, detention, layovers, redeliveries, and claims;
• invoice version, audit adjustment, approval owner, and general-ledger cost center.

Standard reason codes matter as much as timestamps. Free-text labels such as “customer issue” or “carrier problem” cannot support reliable trend analysis. Require a controlled cause, responsible party, and resolution for every material exception.

Alert weekly, before the monthly close

A month-end scorecard identifies damage after it is booked. Weekly alerts should flag leading indicators: cost per unit more than a chosen percentage above matched-lane baseline; spot usage above target; first-tender acceptance deterioration; accessorial dollars above plan; expedite frequency by site or customer; billed weight outside tolerance; and invoice charges without a matching contract rule.

Set thresholds by lane and mode rather than applying one network-wide limit. A $150 variance can be immaterial on an urgent cross-country truckload but severe on a regional LTL shipment. Pair every alert with an owner and deadline, then track whether the corrective action actually removes the cost in subsequent weeks.

The 28.1% spending increase is a warning against managing transportation through volume and blended averages alone. Shippers need a cost bridge that distinguishes market pressure from mix, operational failure, and invoice leakage. With that view, teams can protect service while targeting the dollars they can genuinely control.

Ready to make transportation variance visible before it reaches the budget review? Request a CXTMS demo and see how unified shipment, event, and freight-audit data can sharpen every cost decision.