More Features, Simple User Interface: Insights from Uber’s Design Director Didier Hilhorst
Didier Hilhorst is a rare breed in design: he started as an economist and fell in love with design by “happy accident.” That accident turned into a very successful career working as a senior interaction designer for IDEO, the design lead for Flipboard, and now, as Uber‘s design director. He’s led a substantial redesign of the ridesharing app, grappling with the difficult design challenge of adding features while still retaining an easy-to-use user interface. We asked Hilhorst to share some of his best advice for scaling up a product without turning off your users.
What’s your best advice for designers who want to add features, but still keep their product easy-to-use?
Adding features is easy, removing features is hard. A rigorous process should be in place to evaluate whether a feature is truly necessary or not. Designers (and really anyone working on digital products) should be healthily skeptical about any new feature. As product and design leaders, we have great tools at our disposal to help guide a decision: from insights and research, to data, to testing.
At times adding a new feature actually means rethinking a flow (or functionality) entirely. Yes, this may be more work, but just adding a new feature has a price too, which typically compounds over time. No one wants to go down the path of feature overload.
In many ways design is akin to gardening: get rid of the weeds. If data is showing a feature has poor usage or unintentional consequences, kill it.
How have you been doing this at Uber?
We’re not afraid to take a good look at existing flows and decide to rethink aspects of the product, like we did with our latest redesign of the ‘Rider’ experience. This isn’t just to avoid adding new features on top of a product, but also to adapt to a rapidly changing environment. Change can be scary, but the cost of adding new features can slow things down to a crawl (or even worse) as rapidly as you added them in the first place.
A huge help in evaluating new features is clarity. Clarity can take many forms: from a strong perspective on principles and frameworks to understanding your customer’s behaviors, but ultimately it’s a combination of many things. Typically, random features tend to run amok when product and design requirements are vague, or when there is little to no understanding of the problem space. At Uber, research and data are invaluable tools to help guide conversations and create clarity through design.
What are some of the big challenges you can run into when adding features?
Any new feature has a price tag associated with it. At first, that price tag seems to only be expressed in terms of the cost to the organization: product management, design, engineering, etc. However, over time the cost for the product itself and customers may be much higher. A new feature requires much more than just building it. It also increases the cognitive load and may adversely affect other features (typically an unintended consequence).
Adding new features invariably decreases your focus on core functionality, both internally and externally. At some point it can get so bad that it affects comprehension of your interface, and ultimately hurts your bottom-line.
Another challenge I see is taking the necessary time to introduce a new feature. Quality should never be sacrificed over speed of execution. If two to four weeks isn’t enough time to get to a good result, don’t stick to your timelines. Be honest and make sure the product is not suffering (and hopefully plan better the next time around).
What will the UI of the future look like?
I try to avoid making predictions, but rather observe how people’s behaviors and relationship with technology change over time. I’ve found this to be a more effective way to think about new interaction paradigms and evolve a user interface. That said, once in awhile there’s nothing wrong with going full Minority Report for some fun and inspiration.
The iPhone launched 10 years ago, which is both laughably recent, and yet a decade in technology is a long time. During that time, interfaces have evolved greatly, especially for Uber and how interactions on the device affect the physical world. Increasingly we will see boundaries fade between digital and physical. This in turn will require new interfaces and an evolution of the ways we interact.
You can read more on how Didier Hilhorst and Uber are redesigning their app on his Medium here.