Language
Search
-
Solutions
-
Services
- References
- Events
Agentic AI
Services
KI Agenten
LLMs
Automation Platform
Response Management
E-Invoicing
Mobile
Cloud
GenerAItor
Marketplace
Versicherungen
Banken
Gesundheitswesen
Industrie und Handel
Öffentlicher Sektor
Debeka Group
With its diverse range of insurance and financial services, the Debeka Group ranks among the top five in the German insurance and home savings industry. Founded in 1905, the Group has evolved from a health insurer exclusively for civil servants into an insurance group that offers coverage to all private households as well as small and medium-sized businesses.
Every year, Debeka receives approximately 16 million document pages through a wide variety of incoming channels—traditionally by mail or fax, via email, through the customer portal, or through sales. To process these volumes across all business lines, Debeka relies on the OmnIA (formerly smart FIX) solution from Insiders Technologies, supplemented by the newly developed Cognitive Classifier for combined text and image recognition.
Industry
Insurance / Financial Services
Product
OmnIA with Cognitive Classifier
Volume
16 million document pages per year
"Working with Insiders Technologies allows you to break new ground in AI while being confident that the quality will be right in the end. The future of digitalization will lie in complementing rule-based systems with AI-based systems—just as OmnIA (formerly smart FIX) already does today."
Patrick Schneider
Head of the Frontend Department, Debeka Krankenversicherungsverein a. G.
Project Overview
What was the
initial situation?
- Approximately 16 million document pages annually via multiple intake channels—with a high degree of complexity and heterogeneity across business units
- Incoming mail for the Vehicle Services and Building Society divisions consists primarily of image-based documents such as accident photos, real estate plans, and ID copies
- Text-based classification alone provides little guidance for image-heavy documents—the majority had to be indexed manually
- High volumes require the highest degree of automation and reliability
Why the decision
Insiders?
- Long-standing and successful use of OmnIA for line-of-business separation, routing, and document classification
- Insiders Technologies as a partner for the joint development of the Cognitive Classifier based on a real-world use case
- An agile approach enabled training and implementation within a few months
- Simple configuration options and a high degree of operational autonomy
How was the solution implemented?
- Expansion of the existing OmnIA solution to include the Cognitive Classifier—existing workflows and the user interface remained unchanged
- Training of the Cognitive Classifier using real input documents from Debeka via supervised machine learning by Insiders Technologies
- Intelligent combination of text and image classification: The AI decides on a case-by-case basis which method or combination to use
- Approximately 90 percent of the customization is carried out independently by Debeka’s developers
How can success
be measured?
- Manual post-processing rate in the building savings bank division reduced from nearly 100% to less than 10%
- Significant increase in the level of automation across all divisions
- Reliable recognition even of complex processes involving several hundred pages of mixed document types (floor plans, site plans, registry extracts, construction progress photos)
What are the next steps?
- Integrating the remaining lines of business into Input Management—the life insurance line of business alone will increase the volume of documents by about 30 percent
- Expansion of automation functions already implemented—such as automatic claim creation—to all lines of business
- Tapping into further automation potential through the enhanced quality of combined text and image recognition