A system for on-trip destination insights.

Destination insights prepare riders for where they're headed by providing hyper-relevant info during their trip.

Information may be omitted for confidentiality.

My role.

I owned the end-to-end design process that included: secondary research, workshop facilitation, framework, prototyping, and visual design. This work fell into a much larger product strategy for Uber Feed, where I teamed up alongside another product designer, researcher, and PM. I also worked with engineers through development.

Product timeline.

2 months.


Problem.

At the end of 2016, the Uber team had successfully launched a complete redesign (and rebuild) of the Uber app from the ground up. Because of the aggressive nine month deadline, most of the redesign was focused on the pickup experience and neglected other parts of the app — On-trip Uber Feed being one.

Uber Feed contained important information like trip and driver details, but also had less relevant content like new feature messages and city team announcements. To further add to the problem, partnership integrations were constantly being built on top of it.

Uber Feed had essentially become a junk drawer with little value to the rider. Engagement was super low and there was no clear product direction.

Objective.

If you think about the Uber rider journey, the biggest touchpoint (in relation to time) is the ride itself. Yet, it’s a surface that hasn’t been designed to its full potential. There had been talks across the org that Uber Feed would be hyper-contextual to the ride, serving riders utility, data, and entertainment. Don’t get me wrong, there were glimmers of that idea! But they weren’t brought together and fully realized. That said, our high-level goals as a team were to:

  • Create a scalable solution for Uber Feed.
  • Increase engagement with Uber Feed.
  • Differentiate the Uber app from competitors.

Gaining a perspective.

In order to solve the problem, we needed to frame it from the user’s perspective. I spoke with stakeholders to get a solid understanding of how they were creating cards/content for Uber Feed and which ones had the most engagement. Many said successful cards tend to be contextual to the rider’s trip — pickup, destination, time, weather, etc. I asked myself: How might we use context to help riders reach their end-goals? What were their end-goals? Where were they going?

Digging through research across various teams, I found that a large segment of riders prepare for their destination while on-trip. For example, riders headed to the airport might use their American Airlines app to grab their boarding pass, or they might be on Google Maps to help the driver navigate a faster route. However, if the rider frequently traveled to a destination they were familiar with, they might be browsing Facebook instead.

Questions and familiarity.

I started to think about the problem through two lenses.

Questions.

The types of questions riders might have about their destination lay on a spectrum between uncertainties and curiosities.

Uncertainties are top priority questions that usually have an impact on time and money. Without having the correct knowledge, there might be detrimental consequences.

Curiosities are interest based. Having insight into these can potentially enhance the rider’s experience; they’re the “good-to-knows.”

Familiarity.

The number of times and frequency the rider travels to a particular destination. Applying “familiarity” to “questions” alters the product to be more personalized. For example, the types of questions a rider might ask on their first trip to Brenda’s Soul Food is different from their seventh trip there.


Mashups ideation.

I knew I had to get an idea of the questions riders ask when traveling to specific places. Knowing the questions would help me craft the content for our insights concept. I invited cross-functional leads and designers across the org to a Mashup workshop to help generate a list of questions.

There were three categories of post-its:

Destination contained various types of places that I drew from Foursquare’s API.

Familiarity had random numbers that indicated the amount of times you’ve traveled to that particular destination.

Country was a bonus category for the bravest participants. It contained post-its of random countries across the globe where these destinations were located, forcing them to think about a travel-abroad scenario.

I grouped questions from the workshop into task categories that informed our interaction framework. For example, questions like “Which is the correct entrance for the hospital?”, “How do I get to my terminal after I get through TSA?”, and “Where is the pasta booth at the farmer’s market?” were grouped together under “Orient & Navigate” and required a map as a part of the insight. Using this approach, I created a scalable Destination Insights platform for internal teams and strategic partners to plug into.

Regarding familiarity, what I found interesting was uncertainties tended to be very situational and were independent of familiarity, while curiosities evolved over time and were heavily dependent on familiarity. Let’s use a rider’s first vs tenth time to the airport as an example. The first experience might require a way to navigate to the correct gate, but the usefulness of it decays by the time they’ve taken their tenth trip there (unless it’s a completely different part of the airport, of course). Now, take that same rider and let’s say they’re running late. It doesn’t matter if it’s their first or tenth time; they’d still have the same uncertainty: “Will I miss my flight? Is there a huge line at security?” Other factors outside of amount of times traveled to x destination and frequency contributed to this uncertainty — those factors needed to be considered when crafting insights.


Rapid prototyping and user feedback.

My design researcher and I shared a Principle Prototype of Destination Insights with users to test usability and gauge interest.

Feedback was overwhelmingly positive with minor usability and information hierarchy issues. Riders expressed how convenient the insights were, saving them time from having to multitask between apps.


Final iteration.

After reviewing the findings from the usability test with cross-functional leads and going through multiple design reviews, I landed on my final iteration of Destination Insights. Below are a few samples of the design.


The results.

Trip data and our existing partnership with Foursquare dictated that we focus on restaurants for our MVP. The product feature was shipped as an experiment. I left Uber shortly after engineering started development, so the product may have changed since then.

Destination Insights is the first of its kind, differentiating Uber from its competitors through elevated on-trip experience. What makes it more special is that it was built by looking beyond the transaction and addressing user end-goals. I’m happy to have contributed to an experience that brought more humanity to Uber.