Defining the Gap in Post-Purchase Personalisation

This project started with a simple observation: most travel platforms personalise the search and booking experience, but once a trip is booked, the personalisation stops.

Through initial exploration, I noticed a pattern across platforms like Expedia, Booking.com, and even emerging AI tools. They focus on getting you to complete a purchase, offering filters, smart pricing, and recommended packages during the search phase. But after booking—when you're preparing for your trip—interactions become generic and transactional. Confirmation emails and checkout pages rarely adapt to who you are, where you’re going, or what would actually elevate your experience.

This disconnect became the foundation of the problem space I explored.

For this sprint, we weren’t aiming to reimagine the entire booking journey. Instead, I focused specifically on the post-purchase phase—moments like the checkout screen, confirmation page, and pre-trip reminders. I asked:

Given the sprint’s scope and pace, I framed the problem around contextual personalisation opportunities that could be unlocked using existing data travel companies already collect (demographics, trip details, preferences), rather than introducing entirely new user flows.

This approach allowed me to zero in on a specific, actionable gap in the user journey:
→ The missed opportunity to turn post-purchase interactions into moments of added excitement and relevance through AI-driven personalisation.

Given the fast-paced nature of this project—a design sprint focused on AI exploration—we didn’t have the luxury of conducting primary research, user interviews, or building detailed personas. Instead, my approach was to work within constraints and leverage desk research to understand:

• What data travel companies already collect.
• Where in the user journey personalisation tends to stop.
• How AI could activate existing data at under-utilised touchpoints.

Framing the Research:

Rather than trying to invent new user data pipelines, I focused on data viability—examining the types of user data already available to travel companies:

• Demographics: Age, nationality, lifestyle info from booking forms and loyalty programs.
• Trip Context: Travel dates, locations, purpose of visit (business, leisure, adventure).
• Travel Party Composition: Solo, couple, family, or group bookings.
• Purchase Behaviours: Add-ons like insurance (e.g., ski insurance suggests activity intent).

This approach was less about what we could collect and more about what’s already sitting in company databases but underutilised.

Insights Derived from Desk Research:

• Personalisation is front-loaded: Platforms focus personalisation efforts during search and selection but go “generic” post-purchase.
• Existing AI implementations are reactive: AI chatbots or recommendation engines require the user to take initiative.
• Users expect seamlessness after booking: Post-purchase touchpoints are treated as confirmation steps, not opportunities for value-add.
• Data exists, but is under-leveraged: Companies hold enough context to make hyper-relevant suggestions but lack mechanisms to deliver them organically at the right moments.

Research Takeaway:

With no time for persona-building or surveys, the research lens became:

Opportunity Mapping, Not Competitor Listing

When evaluating existing travel platforms and AI implementations, the goal wasn’t to compete feature-for-feature, but to identify whitespace—areas where current solutions fall short.