Transportation is a key forest logistics component and is a large proportion of the overall cost. Often, the transportation cost is determined by contractual agreements and based on the loaded distance from a supply to a demand point. Many alternative routes provide different distances (e.g., shortest route, fastest route, minimum fuel consumption), but these distances are approximate in the contractual agreement; hence, there is a mismatch between approximated costs and actual pay. It is necessary to match supply with demand when planning to use optimization models, as these models must cover many supply and demand points. From this point, many distances need to be established. These can be generated dynamically before optimization or generated a priori as static distance tables. The former can take a long time, whereas the latter needs to use aggregated zones that remain static over time because supply points, such as harvest areas, change continually. To enable fast optimization, distances between zones and demand points can be precomputed; however, they represent only estimated distances between the actual supply point and the demand points. We propose a hybrid method to improve quality and estimate accurate distances in a quick process, using a large case study from a company in Sweden with a standardized system that directly uses computed reference distances as contractually agreed. Results show that many distance estimation approaches give poor cost estimates (1–20%) and increase transportation costs (0.2–0.6%).