Product design question
As the product manager at Booking.com, how would you leverage generative AI to create personalized travel itineraries based on user preferences, including destinations, activities, and accommodations?
- Booking.com
- Product design
- Hard
Practice this question out loud. An AI interviewer asks it, follows up like a real interviewer would, and scores your answer. Type or speak.
Start a mock interview on this question · Mock interview from a job description
What this question tests
Tests applied generative AI product design in travel: can you design a genuinely useful personalization loop rather than a chatbot wrapper, while accounting for booking-accuracy risk.
How to approach it
- Clarify the user: a Booking.com traveler overwhelmed by choice who wants a coherent, personalized itinerary rather than piecing together flights, hotels, and activities separately.
- Identify the core value: generative AI can synthesize preferences (budget, pace, interests like food or hiking) into a structured day-by-day plan that also links directly to bookable inventory Booking.com already has.
- Design the flow: a conversational preference-gathering step, followed by a generated itinerary with editable day blocks, each linked to real bookable options (hotels, activities, transport) rather than generic unlinked suggestions.
- Address the accuracy risk directly: ground the AI's suggestions in Booking.com's actual live inventory and pricing (retrieval-based, not pure generation) so it never recommends a sold-out or non-existent property.
- Design for iteration: let users refine the plan conversationally (swap a day, change pace) rather than regenerating the whole itinerary from scratch each time.
- Define success as itinerary-to-booking conversion rate and time-to-book compared to the traditional multi-tab search flow, since the value proposition is speed and coherence.
What a strong answer includes
- Explicitly grounds AI output in real bookable inventory (retrieval-augmented) rather than pure free-form generation, addressing the real risk of recommending unavailable options.
- Designs the itinerary as editable day blocks tied to bookings, connecting the AI feature directly to Booking.com's core transaction, not just a standalone chat experience.
- Proposes conversational refinement rather than one-shot generation, matching how travelers actually iterate on trip plans.
- Picks conversion and time-to-book as metrics, tying the AI feature to a measurable business outcome rather than engagement alone.
Common mistakes
- Designing a generic travel chatbot disconnected from actual bookable inventory, risking hallucinated recommendations.
- Treating this as a one-shot generation feature with no way for users to iteratively refine the plan.
- No metric tying the feature back to bookings, the actual business outcome Booking.com cares about.
Likely follow-up questions
- How would you handle the AI suggesting an activity or hotel that becomes unavailable between generation and booking?
- How would you evaluate whether the generated itineraries are actually good quality?
- How would you monetize this feature versus offering it free to all users?
More product design questions
- Design a TV for a car.Google · Product design · Hard
- Design a refrigerator for the blind.Google · Product design · Hard
- Redesign Twitter for eCommerce.Shopify · Product design · Hard
- You are a PM at Flipkart. The company has decided to increase the userbase of age >60 years on the platform. What will you do?Flipkart · Product design · Hard
- Develop a seller score algorithm for Walmart Marketplace.Walmart · Product design · Hard
- Design an application launching in the US marketplace that enables patients to find doctors.Meta · Product design · Hard
More questions from Booking.com
Learn the skill behind it
Chapters of the AI PM course that teach what this question tests.
- Chapter 4: Discovery and strategy for AI products
- Chapter 7: AI UX and human oversight: design for a system that is wrong sometimes
- Chapter 14: Get the job: the AI PM interview loop