Where Does the Money in Transport Really Leak? Rarely Where Companies Look

In this episode of Rozmowy Logistyków, Piotr Skobało, drawing on experience buying around 3 billion złoty worth of transport services, discusses where the money in transport costs really leaks. He shows that the real costs often hide not in the carrier's rate, but in poor planning and empty running.

THE FULL COST OF DELIVERY, NOT JUST THE CARRIER'S INVOICE

Most companies look at transport cost through the lens of rates and invoices, overlooking failed delivery attempts, the hidden cost of complaints, the cost of delays that never gets translated into a monetary value, and the cost of poor planning. Without calculating the full, end-to-end "cost of delivery," it's hard to even start optimizing — looking only at the price per kilometer in isolation from context (period averages, historical comparisons) shows very little.

DATA YOU CAN'T AFFORD TO SKIP

The minimum checklist for meaningful transport analytics: number of shipments, number of load carriers (pallets, boxes, sometimes full tankers), weight and volume tracked separately (because different goods skew more toward weight or volume — an LTL price list differentiates a 400 kg pallet from a 200 kg one at the same cubic volume), number of delivery points, kilometers, and a regional coverage map. On top of that, a financial layer: full understanding of multi-tier price lists (freight, fuel surcharges, road tolls, seasonal charges), and a quality layer: order-to-delivery, service level, OTIF, and the number and value of damages. A practical trick: discard the 2-3% most extreme, distorting outliers — at large volumes, this doesn't hurt the quality of the statistical analysis.

SHARED RESPONSIBILITY FOR QUALITY

If a carrier has a 95% service level, the key question is: who's losing that 5%? Is it the company loading trucks too late, or loading damaged goods, or is the carrier not delivering its part of the deal? Instead of purely negotiating the rate down, it's worth splitting responsibility and working to improve on both sides — a partnership approach that's rarely practiced, because it's easier to squeeze the carrier for another percent of discount than to analyze your own contribution to the problem.

FILL RATE AS AN UNDERUSED LEVER

Many transport optimization projects focus solely on rate negotiation, ignoring fill rate — even though it's often a bigger cost lever. The unit of measure for calculating delivery cost depends on the product category: kilogram, pallet, square meter of surface (tiles), cubic meter (upholstered furniture) — a single measure rarely fits the whole portfolio. Building a pallet above a certain weight can also be constrained by the carrier's distribution network, so there's a ceiling to this optimization.

THE STRATEGIC METRIC: TRANSPORT COST TO REVENUE

The most important, and often overlooked, metric is the share of transport cost in the company's total revenue (freight cost to revenue) — it lets you benchmark against industry competitors (e.g. 7.5% at others versus your own 6-8%) and talk to the board in strategic, not just operational, language. An important caveat: good quality metrics (e.g. high order-to-delivery) don't on their own mean a good operation — a certain level of quality can already be unprofitable if the customer isn't paying for the extra speed, and research consistently shows customers prefer predictability of timing over raw delivery speed.

FOUR STAGES OF MATURITY

Stage one is data visibility: collection, standardization, dashboards. Stage two is governance: end-to-end process ownership, SLAs with carriers, regular quality-and-cost reviews (a common mistake: no single owner of transport cost, when different people buy transport for different channels separately, losing the effect of scale). Stage three is optimization: consolidation, routing, matching carriers to the right work, TMS tools. Stage four is predictive analytics: ETA forecasting, anomaly detection, AI-supported purchasing recommendations. The future of transport isn't a cheaper carrier, it's faster and better operational decisions based on data.