Document Processing
Claims Connection Group needed to help carriers, agents, policyholders, and repair partners interpret dense insurance policy language quickly and consistently because manual review delayed claims decisions, drove no-pay disputes, and increased inbound calls after storms.
Reported impact
LEXXARI accelerated policy review, reduced no-pay claims and post-storm call volume, and gave field professionals faster, more confident guidance during claims handling.
CCG and IBM built LEXXARI as a policy-interpretation platform that uses IBM watsonx and watsonx Orchestrate to read, summarize, and guide users through complex insurance documents. In just 14 weeks, the teams modeled the workflow around real interactions among carriers, agents, and policyholders so the assistant could do more than answer isolated questions. Watsonx was trained on real-world insurance policies to generate fast interpretations, while watsonx Orchestrate coordinates the AI agents that guide users through step-by-step actions during claims handling. The resulting platform turns hours of dense policy review into immediate, decision-ready insight that insurance professionals can use in the field and customers can understand more clearly. Because the experience is grounded in policy content and operational workflows, it reduces confusion, improves claims consistency, and creates a reusable document-intelligence pattern CCG plans to extend into other document-heavy industries.
Insurance policy documents are ingested into watsonx for summarization and interpretation, watsonx Orchestrate routes the resulting insight through workflow steps that mirror claims handling, and users receive immediate guidance or answers instead of manually re-reading policy text.
“AI agents from IBM watsonx Orchestrate were integrated into the platform to provide users with actionable, step-by-step guidance.”
“Clearer coverage understanding has led to a 40% reduction in no-pay claims and 60% fewer post-storm calls.”
“IBM watsonx was trained on real-world insurance policies”
“LEXXARI speeds up policy reviews by 70%”
“LEXXARI was co-developed in just 14 weeks.”
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 Apr 22, 2024 Edger Finance Edger Finance wanted to help investors collect, summarize and interrogate quarterly-report data much faster so users could make better investment decisions without waiting on manual analyst work. 90% improvement in the turnaround time for quarterly report data extracts IBM watsonx Assistant · IBM watsonx.ai 
OpenAI Sample Oct 16, 2026 Benefit Street Partners Benefit Street Partners needed to answer investment-committee follow-up questions faster, compare lender marks across shared loans, and review 300-page credit agreements for clauses that could shift value away from… 10–12 hours total. BDC mark tracker · Codex · Credit agreement analyzer OpenAI Sample Oct 16, 2026 Itaú Itaú needed to make financial guidance simpler and more personalized for millions of customers while also scaling document-heavy internal workflows, all under the governance, observability, and economic controls… 100% of the customer base is the goal by the end of 2026 Customer DNA · IARA · OpenAI models 