Designing Agentic Journeys from Prompt to Booking
An UI/UX solution for a new gen hotel booking agentic app, an experience that moved from design to production in about a week. The exercise was not to create another travel search screen, but to imagine how an AI copilot could guide a user from an open-ended request to a confident booking. The experience was designed around a clear journey: understand the user’s intent, recommend suitable hotels, compare options, explain the best fit, guide room selection, confirm details and complete the booking inside one continuous flow. This direction fits agentic AI’s broader shift from simple chat responses to systems that can plan, use tools, make decisions and complete multi-step actions across a journey.
Turning choice into guided decision-making
ICD’s thinking focused on reducing the burden of comparison. Instead of asking users to scan endless options, the copilot highlights what matters: location, price, family-fit, amenities, room type, policies and hidden advantages. The UI makes the AI’s recommendation visible and explainable, so the user is not just given an answer, but understands why a choice is being suggested.
Designing fast, believable POCs
The work shows ICD’s ability to think beyond conventional product patterns and quickly prototype new behaviours for emerging AI use cases. In a short one-week cycle, the team shaped the journey, interaction model, UI states, comparison panels, secret-tip cards, step tracker and booking confirmation flow into a product experience that felt testable, credible and production-ready for a POC.
share
