The Market: Specialized Optimization Tools You Shouldn't Try to Replicate

Route optimization is a mathematically complex field that has evolved over decades. Providers such as PTV, Wanko, and Greenplan have optimization engines that calculate millions of variants in seconds and account for constraints that virtually no company could replicate on its own without launching a research project.

Our recommendation here is clear, even though we build custom software: For the optimization itself, purchase a standard product. Developing your own optimizer is practically never cost-effective for a medium-sized freight forwarding company, because you’d be competing against products that represent hundreds of person-years of development. Anyone who tries to sell you custom development for the core of route optimization is selling you the wrong project. This honesty regarding the limits of custom software is one of the questions you should ask every partner; you can find out what those questions are in Selecting a Software Development Partner.

Why Good Route Optimization Still Fails in Everyday Life

If the optimizer is so good, why do many trucking companies end up planning their routes based on experience and gut feeling after all? Because the tool is operating in isolation.

The typical workflow: Orders arrive via email and in the TMS. In the morning, the planner transfers them to the optimizer—either manually or via a partial export—updates any special cases, runs the calculations, checks the results, and exports the finalized routes back to the TMS and to the drivers. Each of these steps is done manually. Under time pressure—and the dispatching department is always under time pressure—the detour via the optimizer is eventually skipped, and planning is done manually again. The best optimization result is useless if the process to get there is too cumbersome to follow on a daily basis. The article on SaaS Proliferation.

Here’s an example: A food logistics company with 40 of its own trucks operating in local transport licenses a high-performance optimization tool. In theory, it uses it to plan optimally. In practice, the dispatching team exports the orders from the TMS to a file in the morning, loads them into the optimization tool, manually corrects any cases that were lost during the export, and transfers the finalized routes back to the system. On busy days, there isn’t enough time for this, so planning is based on the previous week’s pattern. The optimization tool is excellent but sits idle for half the week.

The Cost of the Gap: A Calculation Framework

The cost of the manual work involved with the optimizer can be quantified. Use your own values; the following are a model, not a benchmark.

Suppose a planner spends 2 hours per day entering orders into the optimizer, checking the results, and sending the routes back to the TMS and to the drivers. With 220 workdays and full costs of 55 EUR, that amounts to 2 × 220 × 55 = 24,200 EUR per year—just for data transfer related to a tool that’s actually supposed to save time.

Important: This is a theoretical gross potential, not an actual savings amount, and it is not yet the largest item. The more costly effect is invisible: the routes that aren’t optimized in the first place due to time pressure, because using the tool is too cumbersome. These extra kilometers driven aren’t factored into any calculations and often significantly exceed the pure coordination time. Only a process analysis can reveal how much of this is actually recovered.

Standard optimizer, configuration, or custom integration?

Before you consider developing your own solution, you should honestly evaluate the following points in this order.

First: Launch your standard product the right way. Most optimization tools cover more constraints than are initially used. Often, the problem lies not with the tool itself, but with an incomplete implementation.

Second: Check the standard interfaces. Many optimization tools and TMS systems come with built-in connectors. If one of them covers your order flow, don't recreate it.

Third: Are your incoming orders too specialized for the standard process? Only when orders still have to be manually transferred back and forth to the optimizer—because your incoming data and scheduling logic don’t fit into off-the-shelf interfaces—is it worth developing your own integration. Note the difference: In that case, you’re not building the optimizer itself, but rather the connection to it.

When a Standard Optimizer Isn't the Right Choice

There are cases where even a well-configured off-the-shelf product reaches its limits: highly specialized loading rules, a mix of in-house fleet and rotating subcontractors in a single schedule, or unusual time-slot or hazardous-materials logic. Even then, the first solution is rarely a custom optimizer. A more logical approach is to look for a specialized product that handles precisely this restriction; the market for route planning software is broader than the three major players might suggest. Only when no product truly addresses your core constraint—and that constraint constitutes your competitive advantage—does custom development of the optimization core become an option, and even then, it’s usually used as a supplement, not a replacement.

The Cost of Integration into the Dispo Workflow

Here, you’re paying for integration and data flow, not for an optimization engine. A single integration that automatically feeds order data into the optimizer and sends the results back to the TMS and to the drivers starts at a reasonable price of around 50,000 EUR for custom development. An end-to-end dispatching layer that integrates order entry, the optimizer connection, and status updates costs between 75,000 and 150,000 EUR for freight forwarders of this size, plus an estimated 15 to 20 percent per year for operations and further development.

Where your project ends up depends on the number of interfaces to the TMS and the optimizer, the quality of the master data, and the number of special cases. The benefit is twofold: the daily manual work associated with the optimizer is eliminated, and it is actually used every day because accessing it is no longer a hurdle. Only then does the product—which you’re paying for anyway—deliver its full value. The complete cost breakdown can be found in The Cost of Digital Transformation for Small and Medium-Sized Businesses, the article on the ROI of Custom Software. This demonstrates how to establish a data flow between email, TMS, and the optimizer without duplicate data entry, From double data entry to zero manual entries.

This integration is part of the operational layer, which sits above the TMS and the optimizer and consolidates the scheduling data. To see what this layer looks like as a whole, visit our page on Operations Platforms for Freight Forwarders; what really matters in the composition itself is the article in Scheduling Software for Freight Forwarding Companies.

Self-Assessment: Does Your Route Planning Need Its Own Integration?

Developing your own integration rather than continuing to do things manually is likely your next step if several of these points apply:

• Orders are manually transferred from email or the TMS to the optimizer each morning.
• The optimized routes are then manually uploaded back to the TMS and sent to the drivers.
• When time is tight, the optimizer is bypassed and planning is done based on experience again.
• Your optimization tool and your TMS do not have a suitable standard interface for your order intake.
• You know your tours could be better, but the tools to make that happen are sitting there unused.

She's probably not Your next step if you haven't yet fully implemented your optimizer, or if a ready-made connector between the TMS and the optimizer already covers your order flow.

What to do now

Keep the two issues clearly separate. For route optimization itself, choose a standard product and implement it correctly; building your own solution is the wrong approach here. Then measure how much time manual work related to this tool takes up, and observe how often it gets skipped when under pressure. Taken together, these two observations will tell you whether developing your own integration is worth the effort.

If the tool remains an “island” because data is manually imported and exported, the integration into the scheduling workflow is the sound investment—not a new optimization tool. If you’d like to see what this integration looks like for a freight forwarding company of your size, visit our page at Operations Platforms for Freight Forwarders, or you Schedule a call.