Reading Bills of Lading automatically: how a Belgian logistics player sped up its import file process with AI
Every day, B/L documents read manually, container by container, field by field. Repetitive work that slows down import file turnaround. AI can solve it, but only if it actually works on the variety of B/L formats every forwarder sees daily.
Summary
A Belgian logistics player focused on forwarding had a challenge every operator in the sector recognizes. Bills of Lading come in every day, in dozens of different formats from just as many shipping lines and agents. The data on them has to be transferred exactly and completely into the forwarding system, because errors cost time and money downstream.
We ran it the way we always do: start pragmatic and build a focused PoC in a short sprint that genuinely validates the core technology. With positive results, we moved straight into production and integrated the AI directly into the client’s existing forwarding software.
The challenge
Bills of Lading are at the heart of every seafreight import file. Container numbers, Port of Loading, Port of Discharge, Liner, cargo details, all of it on the document. And all of it has to land correctly in the system.
The problem is the variation. Every shipping line has its own layout. Some B/Ls are cleanly structured, others are scanned PDFs with handwritten additions. Some have container info in a table, others as running text. A file handler has to pull the relevant fields out and retype them into the forwarding software, for every file, every time.
Three bottlenecks were stacking up. Time-heavy manual data entry for data that already exists digitally. Error-prone extraction of container numbers and codes where a single typo has downstream impact. And a process that scales linearly with volume, while logistics margins don’t.
[OPEN: concrete pain point numbers if available, e.g. minutes per file, files per day, error rate]
The approach: start pragmatic, validate fast, deliver value immediately
Our methodology runs in three phases that flow into each other without seams.
Pragmatic start. No half-day workshops on vision and strategy. Just a short kick-off to nail down scope, surface edge cases, and lock delivery criteria. Which B/L formats need to be supported? Which fields are critical and which optional? When do we call extraction reliable enough?
High-output PoC. A short, intense build sprint. We built an AI extraction engine that automatically reads an incoming B/L and pulls out the relevant fields: container information, Port of Loading, Port of Discharge, Liner, cargo details. The PoC was tested on a representative dataset with B/Ls from different shipping lines and formats, against criteria locked upfront. End of the sprint: a working prototype, validated against those criteria, and a grounded go/no-go moment.
Straight to production. When the light went green, we didn’t pause for re-budgeting. We kept going. The AI extraction was integrated directly into the client’s existing forwarding software. An incoming B/L is now automatically converted into a Seafreight import file or Transport file, with container information and cargo details already populated on the dossier.
That cadence is the whole point. Most consultancies stall between PoC and production because the PoC team gets disbanded, the production team has to ramp up from scratch, and another scoping phase wedges itself in. We keep the same team from PoC through production, with the expertise to see it through.
The result
A working system that reads Bills of Lading directly and converts them into a structured import file. Container by container, field by field, automatically populated. Not as a black box, but integrated into the existing forwarding process where the file handler sees the output and can validate where needed.
The difference for the file handlers: instead of starting every file from a blank screen, they get a pre-filled baseline. Their attention goes to the exceptions and the commercial context, not to retyping data that’s already on the document.
The impact
Faster turnaround on import files. The reading work that used to take minutes per file now runs in the background. File handlers validate instead of retype.
Less risk on critical data. Container numbers, port codes, and liner details are extracted consistently, regardless of B/L layout. One typo downstream is always more expensive than one prevented.
Scalability without matching headcount. Repetitive entry no longer scales with volume. Capacity goes to files where human judgment matters.
AI embedded in existing tooling. Users don’t switch between systems. The forwarding software stays the working screen, the AI runs underneath.
Why this works for Belgian logistics players
Logistics is a high-volume document processing sector with low tolerance for errors and margins that don’t grow with workload. That makes it an ideal place for AI, provided the solution fits inside existing workflows instead of living next to them.
Our approach bridges that gap. Start pragmatic. Build a working PoC that genuinely tests, not just shows. Green light, straight to production, integrated into the tooling users already know. Red light, a validated answer that lets management decide, with confidence, not to proceed.
Working on a similar question? Book a 30-minute intro call.