Large travel agency
Client project, AI travel advisor, 2026
For a large travel agency we developed an AI travel advisor that runs as an embed on the existing site. The advisor holds the conversation with the visitor, puts a trip together and leads on to the booking page.
- Role
- Design and build
- Stack
- Serverless, vector search, OpenAI
- Status
- Demo delivered
- Type
- Client project
The problem
The agency runs on a white-label platform without CMS access. Any addition therefore had to be an embed on the front end.
Visitors had to find their own way through a large offer of destinations and hotels.
Approach
We designed a chat widget with a personality of its own, added with a single script line. Answers are based on the agency's destinations and hotels and link to existing pages and referral links.
The agency adjusts tone and content from a management dashboard. Data and settings are separated per site, so one installation can serve several sites.
Result
The widget and dashboard are built and delivered as a demo. Live prices and direct booking links are planned as the next phases.
Worth noting
-
One script line
Deployable on a platform without CMS access.
-
Grounded in the agency's offer
Answers come from the agency's own data.
-
Managed from a dashboard
Tone and content can be adjusted without a developer.
How it talks to visitors
Sample conversation. The real client name stays private.
Facing something similar?
Describe briefly what is done by hand today. You get an honest answer, even if the answer is that it does not pay off.