How much time do your support representatives lose searching for similar service requests?. They know someone must have solved the same problem before, but their searches come back empty because they haven’t used the right keywords…

One customer said, “The application freezes after login.” Another reports, “The system hangs before the dashboard loads.” Different words. Same issue. Unfortunately, traditional keyword search treats them as completely unrelated and this doesn’t help our support representatives.

This is exactly the problem we’re solving in our Siebel CRM 26.6 Release Update with Retrieval-Augmented Generation (RAG) and semantic similarity search. Instead of searching for exact words, RAG – an advanced generative AI technique, helps search interpret exactly what the Service Request is actually about.

Search Experience

Now let’s go back to our support representative. Instead of manually trying different keyword combinations over and over again, they simply click the “View Solution” button.

Behind the scenes, Siebel does several things automatically.

  • First, it summarizes the current Service Request.
  • Next, it performs a semantic similarity search against historical Service Requests stored in OpenSearch.
  • Within seconds, the user sees a ranked list of similar Service Requests.

The top result might be a Service Request created six months ago by another customer who described the issue very differently—but experienced the exact same problem.

The engineer can immediately review how that case was resolved, examine troubleshooting steps, and decide whether the same resolution applies.

Instead of spending fifteen minutes searching, they have useful answers in seconds.

More Than Just Matching Service Requests

The user experience doesn’t stop with finding historical Service Requests. Our solution also retrieves relevant articles from the Fusion Knowledge Base, allowing engineers to access documented troubleshooting procedures and best practices from within the same screen.

If multiple Service Requests are found from the similarity search, the user can explore a ranked list of matches, drill into individual cases, review previous resolutions, and even associate an existing Service Request as the parent record.

Everything needed to investigate the issue is available without leaving the context of current Service Request.

The image below shows the search results displayed in the Siebel UI.

Embedded into Siebel CRM

One of the biggest advantages of the solution is that organizations don’t have to build a RAG framework from scratch.

Siebel CRM provides an integrated framework for configuring Retrieval-Augmented Generation. The framework supports both cloud and on-premises deployments, giving organizations the flexibility to adopt AI RAG-based search using their existing architecture.

The illustration below shows how the search is performed:

Bringing AI to Everyday Customer Support

The real value of AI isn’t simply generating answers—it’s helping people make better decisions faster.

By combining Large Language Models with semantic vector search, Siebel CRM transforms the way support engineers work. To learn more about Retrieval Augmented Generation (RAG) for Siebel CRM, read our documentation or contact me directly.