iTromsø · IBM, VC2
10 minutes per day — time spent reviewing potential stories
Independently reportedMedia & entertainment · Document processing · NLP · In production
Document processing
Port of Rotterdam Authority · Rossum
Vendor-reportedLogistics & transportIn productionEnterpriseLLM
Vendor-reported. The customer is named and the numbers are quoted from the source page, but the account comes from the vendor. No independent confirmation.
Reported by rossum.ai (vendor self-report). Checked against the source page on 2026-09-06. 3 of 3 figures below appear on that page word for word.
The problem
The Accounts Payable (AP) team received 20,000 invoices per year with varied layouts. Legacy OCR required manual template setup, and processes were largely manual, reducing efficiency.
What was deployed
- Vendor
- Rossum
- Products
- Rossum, Rossum IDP solution, Rossum Aurora
- Models
- Rossum’s proprietary LLM
- Technique
- LLM
- Build or buy
- Bought off the shelf
- Deployment
- SaaS
- Data used
- invoices, Tonnage Certificates
- Scale
- 20,000 invoices per year
What changed
Each row is quoted from the source. Figures we could not find on the page in those words are marked — they are kept, not deleted, so you can judge them.
810AP days savedper year
“810 AP days saved per year”
Quoted word for word from the source
AP manual effort
“70.7% Reduction in AP manual effort”
Quoted word for word from the source
>90%Accuracy for tonnage certificatesafter 10 tonnage certificates
“>90% Accuracy after 10 tonnage certificates”
Quoted word for word from the source
“We were looking for a SaaS technology that enabled highly automated invoice processing using intelligent document processing. We chose Rossum for its advanced AI, data quality accuracy customizability and out-of-the-box integrations for end-to-end automation”
Difficulties and limits
- Legacy OCR software was reliant on technical teams to set up templates, which impacted supplier onboarding.
- Efficiency of the AP team was further reduced as its processes were largely manual and a high percentage of non-purchase order invoices were causing issues requiring high line item extraction accuracy.
- PoR works with many different suppliers, each submitting invoices with a wide variability of layouts.
Sources
- rossum.ai — vendor self-report, 2024-03-27
Similar deployments
Amplitude · Wolfia
4X — throughput of security questionnaires
Vendor-reportedTechnology & software · Document processing · AI agents · Scaled
Adyen · Rossum
70 — automation rate
Vendor-reportedBanking & finance · Document processing · NLP · Scaled · 2023
Skaylink · Moterra
dropped by around 80% — NIS2 preparation time
Vendor-reportedTechnology & software · Document processing · LLM · Scaled
OM Pharma · DrugCard
3 minutes — review time per article
Vendor-reportedPharma & biotech · Document processing · NLP · In production
AMPREP (Amalgamations Repco) · Cimplify.AI
100% — invoice automation
Vendor-reportedManufacturing · Document processing · NLP · In production