About the client
United European Gastroenterology (UEG) is the leading non-profit organization dedicated to excellence in digestive health across Europe. As a European umbrella organization, UEG brings together 50 national and 17 specialist member societies. Each year, UEG hosts a major medical congress that welcomes over 14,000 healthcare and industry professionals from the field of gastroenterology.
UEG entrusted CONTEXITY, a company specializing at the intersection of information retrieval, semantic reasoning, and artificial intelligence, to develop the backend for a recommendation and search platform, that delivers personalized educational content based on users’ behaviour and interests.
Client
Website
Our services
Software Semanteer Search, Intelligence, Recommendations & Personalization
Partner
Challenge
UEG has been producing high-quality educational materials for over a decade. This includes on-demand recordings and scientific posters from the annual congress (UEG Week), online courses, in-depth medical articles (the “Mistakes In…” series), guidelines from specialist Member Societies, podcasts, and webinar recordings. The content was hosted across multiple sections of UEG’s main website and a digital library that included only a portion of the available resources.
UEG faced multiple critical challenges
- Unify the fragmented content sources, structures and formats into one database
- Build a search and recommendation engine, capable of ingesting the unified database
- Incorporate user behaviour patterns and expressed interests to weight search
results and recommendations, ensuring the most relevant content would rise to
the top.
The Solution
To set up the data backend for the new search and recommendation system, the following technical steps were carried out:
- Deployed the Semanteer Search server in two environments (production and staging), with data schemas tailored to the existing CRM data structures.
- Implemented an automated content push from SugarCRM to the Semanteer server, including ingestion of PDF files and speech-to-text transcript handling.
- Integrated historical user-item interaction data from the legacy library system to improve recommendations for returning users.
- Configured domain-specific synonym rules to ensure accurate search results, particularly for domain-specific abbreviations.
- Developed test cases in Quepid to evaluate the quality of search results for frequently used domain-relevant queries.
- Leveraged user profile information—such as interests—to deliver personalized recommendations.
- The tasks were jointly defined during weekly jour fixe meetings and executed in close coordination thereafter
Business impact
Gutflix represents an innovative use of AI and algorithms to reduce the friction between users and the educational content they’re looking for. It’s one of the first functioning solutions of its kind within the medical society and its educational ecosystem.
Gutflix went live at the end of 2024. By the first quarter of 2025, we had already reached the number of active users previously recorded in the old library — three months ahead of schedule.
Comparing Gutflix’s Google Analytics data from Q1 2025 with the previous library’s data from Q1 2024, we observed significant improvements across key metrics.
+46
number of active users
+46
number of sessions per user
+66
more engagement in sessions
+21
average session length per user
”Communication with CONTEXITY was consistently efficient and cordial, and the project remained on schedule throughout. Alexandros Paramythis and his team have outstanding expertise and technical know-how, which helped us tremendously in understanding all the key details necessary for the successful development and launch of this project. Beyond the tech, his project management skills were equally impressive — he kept everything running smoothly, ensured milestones were met within budget, and somehow managed to keep everyone sane and focused throughout. A real pleasure to work with!
Dung VuongTechnology & Innovation at UEG
