United StatesTravel and aviation2025 – ongoingGlobal Travel Collection
Atlas travel information assistant
A Global Travel Collection advisor estimated Atlas saved 10 to 15 minutes per query
Problem
Travel advisors searched fragmented systems to prepare customized itineraries. Manual navigation consumed time that could otherwise go to clients.
Approach
Global Travel Collection built Atlas using Azure AI Foundry and Copilot Studio. It retrieved travel information and SharePoint content with Azure AI Search, then generated grounded answers through Azure OpenAI. Advisors used a company-built interface.
Results
Takeaway
The query-level saving was an individual advisor's estimate. The projected annual saving across the network was not a measured aggregate.
Categories
- Size
- Large enterprise
- Industry
- Travel and aviation
- Business function
- Internal work and knowledge managementCustomer service
- Innovation type
- Process innovation
- AI technology
- Generative AI
- Adoption
- Off-the-shelf tool or platform
- Evidence
- Vendor case study
Sources
Source summaries
Microsoft described Atlas in use and explained its retrieval architecture. It quoted an advisor's estimate of 10 to 15 minutes saved per query and separately presented network-wide annual savings as a projection.
Full details are in the original sources.
