Packing List and Bill of Lading to CargoWise: Why the Review Step Matters

Most document automation fails the same way: quietly. It extracts a number wrong, nobody checks, and the error lands in your customs declaration three weeks later. The demo above shows the alternative we built CargoMode around: upload a packing list and a bill of lading, watch the extraction land in a booking with the gaps marked, fix what needs fixing, then send it to CargoWise. The review step isn't a bolt-on. It's the product.

What happens when you upload the documents

You drop in a packing list and a bill of lading. PDF or Excel, doesn't matter. CargoMode also takes air waybills and VGM documents the same way.

AI extraction reads both documents and pulls out three layers of data:

• Shipment level: shipper, consignee, port of loading, port of discharge, vessel and voyage.
• Container level: container numbers, seal numbers, container type.
• Line level: each commodity line with description, quantity, weight, and volume.

That data doesn't go straight into CargoWise. It lands in a booking inside CargoMode first, as a working draft you can see and edit before anything touches your TMS.

Why gaps get flagged instead of guessed

This is the part that matters. A bill of lading and a packing list don't always agree. Weights get rounded differently. A container number is smudged on the scan. The consignee address on one document is abbreviated and the other spells it out. An extraction engine that just picks a value and moves on is making a decision on your behalf, and you won't know it made that decision until something downstream rejects it or, worse, doesn't.

CargoMode flags the gap instead. If a field is missing, ambiguous, or the two source documents disagree, it shows up in the review screen rather than being silently resolved. You're not retyping the whole booking. You're looking at the handful of fields the system isn't confident about and making the call yourself.

That's the trade CargoMode makes deliberately: slightly less "fully automated" marketing, considerably more "I trust what's in CargoWise." If your ops team is going to put their name against a shipment, they should get to see what the AI wasn't sure about.

The container number check that runs every time

One check that happens automatically, with no review needed unless it fails: every container number gets validated against the ISO 6346 check digit before it goes anywhere near CargoWise.

The check digit is the 11th character of a container number, and it's calculated from the first 10. If someone mistypes a container number, or an OCR pass misreads a character on a scanned bill of lading, the check digit catches it immediately rather than letting a malformed number travel through your system.

Here's the arithmetic, worked on a real example, MSCU1234565:

Letters to numeric values (ISO 6346 letter table):
M = 21, S = 30, C = 12, U = 32

String becomes:   21 30 12 32 1 2 3 4 5 6
Position weights: 2^0, 2^1, 2^2, 2^3, 2^4, 2^5, 2^6, 2^7, 2^8, 2^9

Multiply each digit by its weight, sum them, divide by 11,
take the remainder. If the remainder is 10, the check digit is 0.
Otherwise the check digit equals the remainder.

Compare that result to the 11th character in the container number.
Match: number is structurally valid.
Mismatch: flag it, don't pass it through.

You can run this check yourself, on any container number, with the container number validator. It's the same logic CargoMode applies to every container it extracts from a document.

What never gets overwritten

One rule sits underneath the whole extraction process: a document never overwrites carrier API data. If CargoMode already has the vessel schedule, container status, or milestone data for a shipment from a carrier's API, a packing list or bill of lading arriving later doesn't get to quietly replace it. Documents fill gaps and get reviewed. Carrier data, which is the carrier's own system of record, stays authoritative.

This matters because paper (even digital paper, a PDF) is a snapshot taken at one moment, often typed by someone outside your organisation. An API feed is live. Letting a static document clobber a live feed is exactly the kind of silent failure the review step exists to prevent.

Sending it to CargoWise

Once the booking looks right, the gaps are resolved and the container numbers have passed validation, one click syncs it to CargoWise through eAdapter. No re-keying the shipment into a CargoWise screen, no copying line items across one at a time. The booking you reviewed is the booking that lands in CargoWise.

If you want the broader picture of what document-to-CargoWise automation looks like without CargoMode specifically, the general how-to is in Automate Data Entry from Packing Lists and Bills of Lading. This article is the walkthrough of how CargoMode does it, review step included.

Try it on a real container number

Before you trust any extraction tool with your bills of lading, you can check the exact mechanism it should be running under the hood. Run a container number through the container number validator and see the check digit calculation for yourself.