Document Processing · Enterprise Knowledge Search
Boehringer Ingelheim had millions of unstructured text documents in inconsistent formats and quality states, making it hard for teams across the business to analyze the data, surface insights, and respond quickly to complex medical inquiries.
Reported impact
The platform gave employees across Boehringer Ingelheim access to previously underused insights, cut response time for complex medical inquiries by over 40%, enabled shared data-related services across the business, and is expected to deliver substantial annual financial benefits as the rollout scales worldwide.
Boehringer Ingelheim worked with Capgemini Invent to turn large volumes of underused unstructured text into a reusable analytics service. The engagement started with design-thinking workshops in 2020 to define use cases, business value, and funding, then moved into an agile build with consultants, data scientists, and software developers. The resulting solution was a fully customized, microservice-oriented NLP analytics platform rather than an off-the-shelf tool, so the team could match the company’s data, workflows, and IP requirements. Text documents from multiple internal sources are brought into the platform for NLP analysis under a scalable Machine Learning Operations approach, while continuous integration and deployment let the team release services iteratively. After an MVP in 2021 and a 2022 pilot, the platform was rolled out worldwide so trained users across Boehringer Ingelheim could access shared insights and support medical inquiry work faster.
Unstructured text documents from multiple internal sources are brought into the custom NLP-driven analytics platform, where NLP algorithms analyze the content under a scalable Machine Learning Operations approach; the resulting insights are then exposed to trained users across Boehringer Ingelheim, who use them to answer complex medical inquiries and provide data-related services faster.
"Boehringer Ingelheim continuously collects millions of text documents in different quality states and formats from various sources"
"The platform reduces the time needed for this analysis and improves the overall response time for complex medical inquiries by over 40%."
"This team developed a modularized and microservice-oriented NLP platform architecture that enabled the continuous integration and deployment of stakeholder-relevant services"
"development of an NLP platform that would transform existing data into powerful knowledge using a scalable Machine Learning Operations approach"
"the pharmaceutical company developed a custom, NLP-driven analytics platform"
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
OpenAI Sample May 29, 2026 Boston Children's Hospital Boston Children's needed to reduce the administrative burden of supply-chain, billing, and scheduling work while also helping clinicians reason through rare-disease cases where fragmented genetic data, incomplete… About 60,000 hours in time savings co-pilot geneticist · enterprise AI layer · secure internal ChatGPT environment OpenAI Sample May 21, 2026 AdventHealth AdventHealth needed to reduce administrative and documentation burden across clinical and operational teams so staff could handle more demand without sacrificing quality or clinician time with patients. 80% reduction in time spent on administrative tasks ChatGPT Enterprise · ChatGPT for Healthcare 
Capgemini Sample Dec 17, 2025 Amsterdam City Archives Amsterdam City Archives needed to make a massive paper archive usable for the public despite handwritten documents, outdated Dutch, and limited search tools that made discovery difficult. 50 kilometers of paper archives AI Search · Chat with History · Handwritten Character Recognition (HCR) 