AI for Freight Forwarders: How to Win Back Time on Dispatch, Freight Billing, and Customs Documents

A freight forwarder’s working day rarely fails because a truck breaks down. It fails because a rate request sits in a shared inbox for three hours, a carrier invoice does not match the quote and nobody catches it, and a commercial invoice arrives as a blurry scan the day before the goods hit customs. The trucks move. It is the paperwork and the coordination around them that eats the margin.

Meanwhile the people who do that work are getting harder to find. Germany is short tens of thousands of drivers and back-office logistics staff, and demographics are pulling more experienced people out of the industry every year than come in²³. Forwarders cannot hire their way out of the workload. They have to take the routine work off the desk.

This guide is for the operations lead, Disponent, or Geschaeftsfuehrer at a freight forwarder or logistics provider who wants a practical view of where AI genuinely helps: dispatch and Disposition, freight billing and Frachtabrechnung, and customs documents and Zolldokumente. It names the real tools, the honest limits, and how a Company Brain plus AI employees fits on top of the TMS and ERP you already run.

The Freight Forwarder’s Time Problem

Forwarding is a coordination business. A single shipment can touch a shipper, a carrier, a customs broker, a warehouse, and a consignee, and every handoff is an email, a document, or a data-entry step. The systems are good at storing a shipment once it exists. They are bad at the messy work of getting it into the system and reconciling it afterwards.

• The driver and staff shortage is structural - Germany is short well over 70,000 professional drivers, a number industry bodies warn would be far higher without the economic slowdown³⁴. Forecasts point to a gap of up to 120,000².
• The workforce is ageing out - 39 percent of German professional drivers are 55 or older, and roughly 30,000 to 35,000 retire each year while only 15,000 to 20,000 join². The same demographic squeeze hits dispatchers, billing clerks, and customs declarants.
• Manual billing quietly leaks margin - freight invoice error rates run 5 to 8 percent in manual programmes, and studies find a large share of carrier invoices contain some discrepancy⁵.
• Freight audit is a recognised cost line - billing errors and audit losses represent an estimated 2 to 5 percent of total freight spend⁸, and a mid-market shipper can lose 3 to 7 percent of freight spend to undetected mistakes⁷.
• Customs work is document-heavy and deadline-driven - commercial invoices, packing lists, and declarations arrive in inconsistent formats and often late, and a missing field can hold a shipment at the border⁹¹⁰.
• Email is the real operating system - much of a forwarder’s day runs through a shared inbox that no TMS controls, where rate requests, booking confirmations, and carrier queries pile up²¹.

None of this is a technology gap in the trucks. It is a capacity gap at the desk. That is exactly the kind of work AI is now good enough to take over.

What AI Actually Does in a Forwarding Office

The word AI covers everything from a spam filter to a self-driving truck. In a forwarding office the useful version is narrow and specific: it reads unstructured input, matches it against your systems, and drafts the next step. It is less a robot and more a very fast, tireless clerk who never loses the thread.

A practical forwarding AI does four things: it reads the email, PDF, or scan, it extracts the structured data, it checks that data against your TMS, ERP, and rules, and it acts by drafting a reply, pre-filling a document, or flagging an exception. A human stays on the steps that bind you legally or financially.

Where AI is strong, and where it is not

The scale is already proven at the top of the market. McKinsey documents one logistics operator that deployed 50 AI agents which automated 60 percent of check calls, 73 percent of order acceptances, and 80 percent of paper invoice payments, and handled two million quotes, saving tens of thousands of hours of labour¹.

The three jobs AI takes off the desk first

1. Dispatch and Disposition support - turning inbound rate requests and bookings into structured, ready-to-action shipments so the Disponent plans instead of retypes.
2. Freight billing and freight audit - matching carrier invoices against quotes and shipment files, and generating customer invoices from the file rather than by hand.
3. Customs documents - reading commercial invoices and packing lists, extracting the fields, and pre-filling the declaration before a declarant checks and files it.

Use Case 1: Dispatch and Disposition

Disposition is the heart of a forwarding office and the most interrupt-driven job in it. A Disponent juggles inbound requests, carrier availability, time windows, and customer promises, mostly through email and phone. Much of the load is not the planning itself but the manual work of turning scattered inputs into a plannable shipment.

What AI takes over

• Rate-request triage - the AI reads inbound enquiries in the shared inbox, extracts origin, destination, weight, dimensions, and requested dates, and creates a structured draft shipment in the TMS.
• Booking confirmation parsing - it reads carrier and shipper confirmations, matches them to the open shipment, and updates status without a dispatcher retyping reference numbers.
• Carrier and rate lookup - it pulls the relevant lane rates and carrier options from the TMS or rate sheet so the Disponent chooses from a shortlist instead of searching.
• Exception surfacing - it flags the shipments that do not fit a standard pattern, such as an unusual lane, a missing time window, or a customer with special handling rules.
• Proactive status replies - it drafts the where-is-my-shipment answers customers ask all day, pulling live status from the visibility platform.
• Handover notes - it summarises the open board at shift change so the next Disponent starts with context, not a cold inbox.

Where the human stays in charge

Dispatch AI is decision support, not a driverless dispatch board. The Disponent still owns the trade-offs the machine cannot see: a long-standing customer who gets priority, a driver who should not run a certain lane, a promise made on the phone that is not in any system.

Use Case 2: Freight Billing and Freight Audit

Freight billing is repetitive, deadline-bound, and unforgiving of small errors, which makes it one of the highest-return places to put AI. There are two sides: generating clean customer invoices from the shipment file, and auditing inbound carrier invoices against what was agreed.

Why billing leaks money

• Errors are common - manual freight invoice error rates run 5 to 8 percent, and a large share of carrier invoices carry some discrepancy⁵.
• The exposure is material - billing errors and audit losses represent 2 to 5 percent of total freight spend⁸.
• Undetected mistakes compound - a mid-market operator can lose 3 to 7 percent of freight spend to errors nobody catches⁷.
• Manual audit does not scale - checking every invoice by hand is exactly the work forwarders cut when they are short-staffed, so most invoices go unaudited⁶.
• Recovery is real when you look - freight audit typically recovers 8 to 12 percent of the spend it actually checks⁸.

What AI takes over

• Carrier-invoice matching - the AI reads each inbound invoice, extracts the charges, and compares them line by line against the quoted rate, agreed surcharges, and the shipment file, flagging anything that does not reconcile.
• Customer-invoice generation - it assembles the outbound invoice from the shipment record, pulling the right charges, references, and terms, ready for a clerk to approve.
• Accessorial and surcharge checks - it catches duplicate charges, wrong fuel surcharges, and services billed but never rendered.
• Dispute drafting - when an invoice is wrong, it drafts the query to the carrier with the evidence attached.
• Posting to the ERP - approved invoices flow into the accounting system without manual keying.
• Cash and margin visibility - it flags shipments where the margin is thin or negative before the invoice goes out, not after.

This is the use case that most often justifies the whole programme on its own, because the savings are measurable in euros recovered and hours returned, not in soft productivity.

Use Case 3: Customs Documents and Compliance

Customs is where forwarding paperwork is most punishing. The data arrives late and in poor formats, the deadlines are hard, and a missing or wrong field can hold goods at the border. It is also where AI has one of its clearest wins, because most of the work is reading documents and moving data, not making legal calls.

What AI takes over

• Document extraction - the AI reads commercial invoices, packing lists, and transport documents, including scans and non-standard formats, and pulls the fields into structured data⁹.
• Declaration pre-fill - it pre-populates the customs declaration in your customs software so the declarant checks and files rather than keys from scratch.
• Completeness checks - it flags missing data, mismatched values across documents, and inconsistencies before filing, when they are cheap to fix¹⁰.
• Tariff-code suggestion - it proposes a likely commodity code based on the product description, which a trained declarant confirms or corrects.
• Email triage - it reads the customs correspondence that arrives all day and routes or drafts responses²¹.
• Batch processing - it handles large document batches at once, clearing the volume a clerk could not process in a day.

The compliance line AI does not cross

In Germany the legal filing still runs through ATLAS and your customs software, and a qualified declarant keeps responsibility for the classification and the submission. AI removes the keying, the copy-paste, and the completeness checking. It does not assume liability for a tariff decision, and no serious deployment lets it.

The Real Tool Landscape

There is no single AI product that runs a forwarding office. There is a landscape of tools, each strong in one layer. Understanding who does what keeps you from buying the wrong thing or expecting one tool to do everything. Here is the honest map, with real vendors named.

Transport management systems (the system of record)

• CargoWise (WiseTech Global) - the dominant global forwarding and customs platform, deep functionality across air, ocean, and customs, widely used by larger forwarders¹⁴.
• Riege Scope - strong in the DACH and EU market, particularly on air, ocean, and customs compliance, popular with mid-sized forwarders¹⁵.
• Soloplan CarLo - a leading German road-transport TMS with disposition, telematics, and billing built in¹⁷.
• Descartes, Magaya, GoFreight, Shipwell - forwarding and TMS platforms adding native AI features for tracking, quoting, and documentation¹³¹⁴.

Customs and trade compliance

• AEB - the DACH specialist for customs filing, export controls, and trade compliance, often run alongside a TMS¹⁶.
• Descartes and Riege Scope customs - integrated customs modules for forwarders operating across borders¹⁵.
• ATLAS - not a vendor but the German customs IT system every declaration ultimately flows through.

Freight audit, visibility, and quoting

• Transporeon (Trimble) - carrier connectivity, freight procurement, and visibility across the shipper-carrier network¹⁸.
• Visibility platforms - project44, FourKites, and Shippeo track shipments in real time and feed status data to the rest of the stack¹³.
• AI quoting and communication tools - Wisor, Freightos, Cargobase, Logixboard, and Sedna automate rate quoting and inbox work for forwarders¹³²¹.

The Layer That Ties It Together

The tools above each solve a slice. What no single slice does is hold the knowledge that makes your forwarding business specific: your lanes, your negotiated rates, your customer handling rules, and the way your team actually works. That is what Superkind builds, and it is why the approach fits forwarding.

Superkind is a Company Brain plus AI employees. The Company Brain holds your routes, rates, customers, and processes. The AI employees, such as a dispatcher, an invoice clerk, and a back-office clerk, connect to the systems you already run, email, TMS, and ERP, and take over the routine work across all of them. It is one layer over everything you already use, not another system to replace what you have.

• Process-first, not product-first - we map how your Disposition, billing, and customs work actually run before building anything, so the AI fits your office, not a template.
• Sits on top of your TMS - CargoWise, Riege Scope, Soloplan, or whatever you run stays the system of record. The AI reads from and writes back into it.
• Connected to your real inbox - the AI works in the shared inbox where forwarding actually happens, triaging rate requests, confirmations, and carrier queries.
• AI employees with clear roles - a dispatcher agent for Disposition support, an invoice clerk for billing and audit, a back-office clerk for customs document work.
• The Company Brain remembers - your rates, lane quirks, and customer rules live in one place the AI draws on, instead of in one person’s head.
• Human-in-the-loop by design - quotes, declarations, and payments are drafted by AI and approved by your team, with every action logged.
• Outcomes, not licences - pricing is per use case with clear ROI defined before the build, not a per-seat contract.
• Live in weeks - a first use case goes into production in 8 to 12 weeks, running in parallel with your team before it takes load.

A 90-Day Rollout Plan

The forwarders who get value do not launch a grand transformation. They pick one painful, high-volume process and take it from manual to automated in a quarter. Here is a realistic path.

Phase 1: Pick and map (Weeks 1-4)

1. Week 1: Choose one use case - rate-request triage, carrier-invoice checking, or customs-document extraction. Pick the one that is most repetitive and easiest to measure in your office.
2. Week 2: Map the real process - watch how the work actually happens, including the email steps and workarounds nobody documented. This is where generic tools miss.
3. Week 3: Connect the systems - identify the inbox, TMS, ERP, and document store the AI must read from and write to, and confirm access and formats.
4. Week 4: Set the baseline and KPIs - measure current time per task, error rate, and volume so you can prove the change. Define the human-in-the-loop checkpoints.

Phase 2: Build and test (Weeks 5-8)

5. Weeks 5-6: Build the AI employee - connect it to your systems and load your routes, rates, and rules into the Company Brain. No new platform for your team to learn.
6. Week 7: Test on real history - run it against your past emails, invoices, and documents and compare its output to what your team actually did.
7. Week 8: Tune the exceptions - fix the edge cases the test surfaces, adjust the confidence thresholds, and lock the approval steps.

Phase 3: Run in parallel, then hand over (Weeks 9-12)

9. Week 9: Shadow mode - the AI drafts, your team keeps doing the work, and you compare. Nothing binding goes out without a human.
10. Weeks 10-11: Hand over the load - the AI takes the volume, your team reviews and approves, and the exception rate falls as it learns.
11. Week 12: Measure and decide the next use case - compare against the baseline, show the result, and pick the second process to automate.

Decision Framework: What Should You Actually Buy?

The right move depends on your volume, your systems, and where your pain is. Here is a straight framework.

Frequently Asked Questions

Henri Jung

Co-founder of Superkind, where he helps SMEs and enterprises deploy custom AI agents that actually fit how their teams work. Henri is passionate about closing the gap between what AI can do and the value it creates in real companies. Before Superkind, he spent years working with mid-sized businesses on digital transformation and saw first-hand how many AI projects fail because they start with technology instead of process. He believes the Mittelstand has everything it needs to lead in AI - it just needs the right approach.