Skip to content
Nordic Labs X

// case study

From 14 hours of order typing a week to none

Orders arriving as emails and PDFs are now read, checked and entered into the ERP automatically, with a human approving only the exceptions.

The client
Kvarnby Logistik
Industry
Logistics
Year

The client

A regional logistics company handling about 900 B2B orders a month for 40 customers.

Their X

“Two people spend most of Monday and Thursday typing orders from PDFs into our system. And we still get typos in addresses.”

How we understood it

Orders came in eleven different PDF layouts and free-text emails. Off-the-shelf OCR tools failed on half of them.

But every order had to end up in the same six fields in the ERP, and the customers' delivery addresses were already known.

That meant AI could read any layout, and the known customer data could validate what it read.

What we built

An automation that picks up incoming orders, uses AI to extract the fields from any layout, validates them against customer master data, and creates the order in the ERP.

Anything uncertain lands in a small review screen where a person approves or corrects it in seconds.

Key decisions

  • Every extracted field carries a confidence score; low confidence always goes to a human.
  • Corrections are logged and used to improve the extraction prompts.
  • The ERP integration uses its official API. No screen scraping.
Architecture
    • Order inbox
    • AI extraction
    • Validation rules
    • Review screen
    • ERP
  • Order inbox → AI extraction
  • AI extraction → Validation rules
  • Validation rules → ERP
  • Validation rules → Review screen
  • Review screen → ERP

Results

Manual order entry per week
0 hrs
Before: 14 hrsAfter: 0 hrs
Orders processed automatically
87%
Before: 0%After: 87%
Address errors per month
2
Before: 23After: 2

In their words

“We expected a software project. We got two days a week back. The review screen means we still feel in control.”
Johan BergOperations Manager, Kvarnby Logistik

Built with

  • Claude
  • n8n
  • PostgreSQL
  • Visma

// next step

Have a similar X? Let's talk.

Tell us the problem in your own words. We'll reply within one business day with a straight answer on whether we can solve it, and how.

Free · 30 minutes · no commitment