CargoWise data extraction

Get the data out of CargoWise.
Three ways, by effort.

Putting data into CargoWise is the easy part. Getting a clean answer back out, with every charge row intact and the gaps visible, is where the hours go. Pick the route that matches the question you are asking.

The three routes

No build

Ask Claude

Connect Claude, ChatGPT or any MCP client to CargoMode and ask about a job number. The answer comes back with the routing, the containers, the parties and what is still missing.

Fit: Best for ops, sales and management who need one job explained now, not a report.

Limits: Read only. Retrieves known jobs from your assigned company. Job costing needs separate consent. It does not scan every job or run continuously.

See Claude on CargoWise →

Build it

Extract through eAdapter

Retrieve organisations, known shipments and existing accounting batches with the documented Native and Universal XML requests, then land them in a reporting database you can reconcile back to CargoWise.

Fit: Best for developers and BI consultants building margin, invoice or operations reporting.

Limits: You need eAdapter enabled on your CargoWise account and someone to own the pipeline: completeness checks, replay and freshness are your job.

Read the extraction guide (Vol. 3) →

Free tool

Export to CSV

Point the CargoWise Data Exporter at a job range with your own eAdapter connection and download shipments and containers as CSV files.

Fit: Best for a one-off pull into a spreadsheet, a migration check or a quick audit.

Limits: Review retrieval coverage before you trust the result. An empty export is not proof of an empty job range.

Open the CargoWise Data Exporter →

The no-build route, in 43 seconds

Ask Claude about a CargoWise job.

Connect once from CargoMode Settings. Then bring a job number to Claude and ask. The routing, the containers and the parties come back, with what is missing flagged. No screens to hunt through, nothing to export.

Whichever route you take

Six ways a CargoWise report goes quietly wrong.

Repeated charge codes

Two freight charges can share one code. A parser that keeps one row per code overwrites the first amount. Keep every source row.

Mixed currencies

A margin report that adds USD costs to GBP revenue is wrong before it starts. Keep currencies separate and record the conversion basis.

Missing is not zero

A cost that has not been posted yet is unknown. Show it as unknown and leave the job margin incomplete until it arrives.

Operational costs versus posted accounts

Job costing and posted accounting transactions are different sources. A shipment profitability view is not a ledger or a cash report.

Joins that multiply

A job with several invoice lines doubles its totals if every line joins to every usage row. Pick the grain first, then compare totals before and after.

Stale extractions

Show the time of the last complete extraction next to the report, so a dashboard refresh cannot hide old source data.

The worked examples behind these, including a five-row reconciliation you can run yourself, are in CargoWise Data Extraction & Accounting via eAdapter. Chapter 1 is free.

Questions we get

Before you pick a route.

Can I get data out of CargoWise without eAdapter?

Yes, two ways. Ask Claude through the CargoMode MCP connection, which reads known jobs from your assigned company. Or use the free CargoWise Data Exporter for a CSV of shipments and containers in a job range. Both still use an eAdapter connection behind the scenes; the difference is that you do not have to build or maintain anything.

Does the MCP connection replace a reporting pipeline?

No. It answers questions about known jobs on demand. If you need every job, every night, reconciled to CargoWise, that is an eAdapter extraction and the Volume 3 guide covers how to build one that stays honest about missing data.

Which CargoWise XML do I use for extraction?

Organisation search needs Native XML. Known-job shipment requests and existing accounting batch exports have documented Universal XML patterns. Keep the interface, company context and response status explicit in every request.

Can CargoMode write data back into CargoWise?

CargoMode syncs reviewed bookings into CargoWise through eAdapter as part of the document and quote workflows. The MCP tools and the exporter are read only.

Is my data stored by CargoMode when I use the MCP connection?

Requested job data is returned to the AI client you authorised and is subject to that provider’s terms. Your CargoWise credentials stay on the CargoMode server and access covers the assigned company only.

Not sure which route?

Bring one job and one question.

Twenty minutes on a call and we will show you the answer coming out of your own CargoWise company, read only, through Claude.