Document Processing
Medtronic’s teams were spending too much time manually matching invoices, proof-of-delivery documents, and purchase orders, which slowed cash-flow-related processes, introduced errors, and kept staff away from more strategic work.
Reported impact
Medtronic dramatically reduced processing time, improved matching accuracy, and freed staff for strategic work while also helping reduce open accounts receivable and improve cash flow.
Medtronic partnered with IBM Consulting and IBM Neudesic to deploy the Document Intelligence Platform Accelerator, an AI-powered automation layer built on Microsoft technology. The platform ingests invoices, proof-of-delivery files, and purchase orders, then uses machine learning and generative AI to extract and match information even when documents contain poor scans, stamps, or inconsistent layouts. IBM Neudesic customized the accelerator so it fit Medtronic’s existing workflows rather than forcing a new manual review process, and the first departmental rollout was completed in only four weeks. Because the system automates the matching step rather than simply flagging issues, teams can process documents at a much higher rate while reserving human effort for exceptions and higher-value analysis. The project therefore improved both operational accuracy and the financial workflow around open receivables, showing how document intelligence can produce immediate business value in a global healthcare operation.
Invoices, proof-of-delivery files, and purchase orders are ingested into the DIP Accelerator, AI extracts and matches the relevant fields across the documents, and the resulting match decisions are pushed back into Medtronic’s credit-services and supply-chain workflows for faster resolution.
“By leveraging advanced machine learning (ML) and generative AI (gen AI), the DIP Accelerator handles complex formats, stamps, poor scans and non-standard layouts with ease.”
“Down from up to 20 minutes per invoice to 8 seconds”
“In just four weeks, Medtronic automated document matching within one department”
“Together, they deployed the Document Intelligence Platform (DIP) Accelerator”
“automates the ingestion and matching of invoices, proof of delivery and purchase orders.”
Review how this reference was scored and trace each normalized claim back to its documented source.
Scoring methodology
Customer Reference Value measures usefulness as a customer reference. Evidence Confidence measures the quality and traceability of available evidence, not independent verification.
Traceability
Field coverage shows which facts each source supports. Dates are shown only when documented by the source.
Related
IBM Sample Oct 16, 2024 Medtronic Medtronic, a leading medical device manufacturer, identified an opportunity to optimize their process for identifying critical raw materials within their extensive range of products. Medtronic achieves 90% accuracy IBM Sample Apr 18, 2024 University Hospitals Coventry and Warwickshire NHS Trust UHCW NHS Trust needed to reduce outpatient delays and missed appointments while managing an elective backlog because existing scheduling and validation processes were too slow to keep up with demand. 10%-15% reduction in the overall backlog Celonis · IBM Consulting · IBM Garage 
Google Cloud Sample Archived · Oct 9, 2026 Angelus Founded by a dentist whose first product was based on a patent, Angelus competes on differentiation, developing bioceramic products and fiberglass post lines that require clinical proof, intellectual property… 50,000 pages of proprietary data now powering AI-generated documents Gemini 