AI automation for transport companies

You have ten trucks, a planner, and a phone that rings all day. Customers calling for status updates. Drivers calling back when something goes wrong. At the end of the week, a pile of CMR documents still waiting to be processed manually. Meanwhile, only 7% of transport companies have a concrete AI policy, according to research by Van Oers (Outlook 2026). The other 93% lose hours every week to work that could be automated.

Where does your time go?

Ask owners of regional transport companies where they lose time, and you hear the same answers. Not in driving or planning, but in the admin around it.

• Manually scanning, transcribing and archiving CMR documents and waybills
• Calling drivers for status updates that customers also call in about
• Checking and entering incoming invoices from subcontractors one by one
• Daily route planning in Excel or from memory every morning
• Calling back customers who want to know where their shipment is

CMR documents processed in under a minute

A completed CMR waybill contains 25 to 30 fields. Manually transcribing them into your TMS or accounting software takes 5 to 8 minutes per document. With 20 deliveries a day, that is up to two hours of pure admin, every single day.

AI can take that over. The driver photographs the signed CMR with their phone. The system automatically reads the relevant fields: sender, receiver, reference number, weight, date, signature. That data goes directly into your system, whether that is a TMS, Exact Online, or a simple database. No manual entry, no errors from end-of-day fatigue.

Route planning for small fleets

Enterprise tools like OptimoRoute or Routific are built for fleets of 30 trucks or more. They cost €150 to €400 per month and take weeks to implement. If you have 8 trucks covering a regional area, that is too much and too expensive.

For small fleets, the first step is simpler: automatic processing of customer orders into an optimised daily plan, accounting for time windows and load order. That is possible from €50 to €100 per month. And if you are currently planning in Excel, almost any step forward is an improvement.

Fewer calls about order status

A large share of phone calls in transport are status enquiries. The customer calls you, you call the driver, the driver calls back. For a shipment of two pallets. That costs you 15 minutes, your driver 5 minutes, and the customer is still waiting.

With a simple automation, the system automatically sends a message to the customer whenever a delivery status changes: in transit, delivered, delayed. The driver scans on delivery, the customer gets a confirmation via WhatsApp or email. No phone call needed. With 8 drivers each making 6 deliveries a day, that eliminates 10 to 20 messages or calls per day.

Invoice processing for subcontractors and toll costs

If you work with subcontractors or freelancers, you know the problem: at the end of the month, ten invoices arrive that all list things slightly differently. One puts the trip date, another a reference number. You have to match them manually against your planning.

AI-based invoice recognition reads incoming PDFs automatically, matches the data to the corresponding trip, and queues it for approval. Toll costs are automatically booked per vehicle. What used to take half a day now takes 10 minutes of checking.

What does it cost, and when does it pay back?

Realistic figures for a transport company with 8 trucks and 2 to 3 staff:

• CMR automation: €80–€150 per month for a SaaS solution, or €2,000–€4,000 one-off for custom integration
• Customer notifications via WhatsApp or email: €30–€60 per month
• Invoice recognition and accounting integration: €80–€200 per month
• Total for all three combined: approximately €200–€400 per month

At an admin hourly rate of €22, that is a saving of €176 to €264 per week, or €9,000 to €13,000 per year. The payback period on an investment of €3,000 to €5,000 is 3 to 6 months. With ongoing SaaS costs of €300 per month, you pay yourself back in the first month.

Where do you start?

Not with everything at once. The most effective approach: pick the one process that costs the most time right now, automate that first, and see if it works. Then move to the next.

• Step 1: Add up how many hours per week go into CMR processing, phone calls and invoice handling
• Step 2: Choose the one process that costs the most time, that is your starting point
• Step 3: Test for 30 days with real data before expanding
• Step 4: Then expand to the next process