Delivery route planning

Planning routes by hand works as long as there are two trucks and a planner who knows everyone. Beyond that, the time spent every evening balloons, and a single absence throws the day off. A planning tool does not replace the planner: it saves them time and makes their decisions transferable. Here is what it needs to take into account, how I build it, and what it costs.

List the constraints before talking about software

An optimization result is only as good as the constraints you feed it. Before any development, I take inventory with the operations team:

The last row matters most: many planning rules are not written down anywhere. Scoping is precisely where they get put on paper. It also makes operations less dependent on a single person: the day the usual planner is away, the rules are still in the tool.

Automatic optimization: rely on a proven engine

I do not build in-house optimization algorithms: it would be expensive and less reliable than existing engines. I use, for example, Google’s Route Optimization API, which, according to its documentation checked in October 2026, calculates route plans for one or more vehicles taking into account time windows, capacities, driver shifts and breaks, and returns the stop sequence, vehicle assignments and estimated arrival times. Its usage is billed by Google under its own pricing terms, to be checked in its documentation at the time of the project.

The custom work consists of:

• automatically pulling in the day’s orders from your ERP or ordering tool;
• translating your rules into the optimization engine’s format;
• presenting the result on a map and a schedule the planner can adjust;
• sending approved routes to the drivers’ apps;
• replanning during the day when an urgent order comes in or a vehicle is out of service.

The planner stays in control

A tool that imposes routes without explanation quickly gets worked around. I always include the ability to move a stop, lock an assignment (this customer with this driver), rerun the calculation on the rest, and compare two proposals. Basic metrics, miles or kilometers, hours, load factor and estimated delays, are displayed for each version.

On the driver side, the approved route arrives in the app with the order of stops, instructions and the proof of delivery to complete: that is the subject of the driver app. And your customers can be notified automatically with real-time delivery tracking.

The budget

Starting point: the floor for a business app, $6,800 (€6,000)excl. VAT. A sample breakdown for an operation with 10 to 50 users (planners and drivers):

Without automatic optimization, with map-assisted planning instead, the budget drops toward $16,800 (€15,000)excl. VAT. On top of this come the optimization engine, mapping service and hosting usage, billed by the providers. For your setup, the estimator gives you a range in one minute.

When this is not the right solution

Off-the-shelf route planning software exists, usually billed per vehicle or per user. If your constraints are standard (time windows, capacity, hours), you do not need deep integration with your ERP and your fleet is stable, one of them will do the job, faster and for less in year one.

Custom software makes sense when your business rules cannot be configured in those tools, when re-entering data between the ordering tool and the routing tool costs you time every day, or when planning has to fit into a broader system (inventory, invoicing, fleet management). On that last point, see also custom fleet management software.

Where to start

1. Gather a week of real routes and the planner’s rules, even informal ones.
2. A free 30-minute discovery call.
3. Scoping (1 to 2 weeks): constraint inventory, choice of engine, integrations, fixed quote.
4. Test on real data: we compare the suggested routes with the planner’s own.
5. Development in batches, go-live and 3 months of support included.

The full offer is presented on the route planning app page.