TripAdvisor Mobile Redesign • January 2017
After we successfully launched navigation updates on Android and iOS, it was time to take a closer look at the new Search & Discovery tab we had built. To save time and development effort in the first phase, we had launched with the most bare bones solution possible: a list of links. We knew that we could do better.
We had some big questions: Should this tab focus more on Search, on Discovery, or should it be a mix of both? What content should we show in each of the core travel modes?* We did not want the screen to turn into an endless feed, but we did want to acknowledge our users' contexts and surface information that would be valuable.
To help answer our questions, we ran a design sprint.
In September 2016, designers, product managers, and engineers from the Mobile Experience team gathered in Palo Alto, California for a three-day design sprint. We encouraged anyone and everyone to participate.
In the sprint kickoff, we used existing TripAdvisor research to identify three traveler personas to focus on. You will see the personas below.
With those personas in mind, we focused on answering the question: what could we build to make a truly awesome Discovery/home screen experience? We got out Sharpies and drew storyboards to brainstorm as many ideas as possible. Next, we grouped our ideas into themes and voted on which themes would be most interesting to prototype.
The next day, we went out into touristy areas of San Francisco—the Ferry Building and the Embarcadero—to validate our prototypes with real travelers. Then we reconvened in the office to review our findings as a group and start to plan our Discovery proposal.
While we gained so many valuable insights from the sprint, the greatest takeaway was that travelers want it all. They want a balance of “must-see, Top 10" activities along with “spontaneous, off-the-beaten-path” activities in order to not feel FOMO (Fear Of Missing Out) while traveling.
For a full a retrospective of the design sprint, check out this link: https://goo.gl/DE0yM7
We continued to prototype and test a bunch of concepts even after the sprint was over. We knew that if Discovery was going to be a hit with travelers, it must be dynamic, intelligent, and contextual. It must be personalized for each individual, reflecting her/his travel preferences through a number of signals: browsing history, completed bookings, trips s/he organized in the My Trips product, etc.
Discovery would be the very first screen that many of our users would see—we wanted it to get it right, not only for their sakes, but also for our engineers. The new, dynamic content framework required to support a Discovery screen would take several weeks of development time and we needed to be absolutely sure that it was worth the investment.
As we spec'd out Discovery, we listed which of TripAdvisor's content might be helpful in each of the travel modes (Local, Planning, In-Destination). Because the screen was going to be so dependent on individual travelers' preferences, we knew that we could not try to design screens for every possible state, the scope of the project would blow up. Instead, we began to think about Discovery in terms of a larger system of flexible UI components.
We had a choice to make at that point. We could either continue using the app's existing, several-years-old design language and save a little development time, or we could take advantage of the opportunity and truly redesign the application. To my delight, we chose the latter! For more about that process, check out phase three of the redesign. Read more→