During my time at TripAdvisor, one of the more frequent pain points for both customers and the business, was difficulty with searching and discovering content. Customers often found it hard to find exactly what they needed in the moment they opened the application, and TripAdvisor's hotel and restaurant bookings were of the utmost importance.
The mobile product team needed to provide an engaging and contextual experience that would immerse the customer in the destination of their choosing, and to use machine learning and data science to drive relevant content. This was a nearly year-long reimagining of home, search, and discovery.
If you're interested in exploring some of these features more in-depth, please download the TripAdvisor® iPhone® and Android® applications.
The mobile product team got all necessary stakeholders from each product team (hotels, restaurants, flights, vacation rentals, attractions, and search) together to discuss the challenge at hand. We took the lead on iterating on top of a previous test we had done.
I collaborated with four other product designers and product managers throughout the year, where we conducted user interviews, built prototypes, tested with customers, scrapped hundreds of different directions, and iterated. We eventually landed on a fully immersive home experience that grounded the customer in the location of their choosing, depending on user input (i.e. "I'm traveling to Los Angeles on these dates") or location detection (i.e. we know the customer is at home in Oakland, so show this relevant content).
We coordinated with the larger product and engineering organization to lay out the roadmap and build it iteratively. Through this phased approach, we were able to discover some issues early on that allowed us to simplify the experience. For one, the home page was perceived as overwhelming and too long, so we worked with user research to narrow down to the top use cases.
Here is a prototype I built in Principle to design some interaction and animation ideas, as well as do some light user testing during one of the iterations. The final design was altered slightly from this.
We saw a lift of about ~2% in hotel bookings, ~4% in restaurant reservations, and ~3.5% in attractions bookings.