User data has become a global currency. An incredible amount of information about our interactions and habits gets collected every day by companies that want to create personal digital experiences for their users. Your shopping history, favorite restaurants, heart rate, and workout plan are being used to influence how we shop and what we buy. Some of these — Amazon, Spotify, Netflix, FitBit — already dominate our daily lives.
But these developments have brought up an issue: Most data-centric user interfaces aren’t very good. Usually, interfaces look beautiful but they don’t communicate data effectively, or they have great underlying data sets but they don’t appeal to users. This is largely because UX design and data science work independently. While both UX designers and data analysts share the same roof in a digital agency, they approach their objectives from very different angles. User experience is often grounded in art, design, and psychology, while data science is concerned with technology, engineering, and quantitative data analytics. Because of this, UX designers and data scientists often don’t speak the same language, let alone share a common understanding of the desired user experience.
"With data-rich apps becoming the norm, it's important to consider how we define good data-centric design."
Data will continue to change the way people engage in physical and digital environments. In the meantime, the opportunity to engage with, browse, and explore data sets is not just the domain of specialists and analysts anymore. This territory now belongs to consumers as much as it does to marketers, and these changes are driving a revolution in the way information is collected and consumed across different platforms and devices. With data-rich apps becoming the norm, it’s important to consider how we define good data-centric design.
Part of the explosion in data is because we’ve shifted from web pages — made up of mostly static content and with limited interactivity — to data-centric web applications that display dynamic content and allow users to take action.
UX designers play an essential role in making data-centric apps accessible by creating interfaces that inform and empower users to take action. The problem is: designers don’t usually understand the data structures and methodologies that drive the experience – that’s the domain of data scientists who use analytics and measurement strategies to help companies understand their customers by analyzing and predicting patterns of human behavior.
UX designers need to see data as something that can actively shape and enhance the user experience, while data scientists should pay more attention to the value of visual thinking and presentation dynamics for analyzing and formatting data so that the general public understands it.
Rather than data-centric user interfaces that present information boxed in static containers, design elements could be shaped in real-time and allow for functionality based on information about the user. To provide the best experience, more sophisticated apps could evolve to determine how data is presented. In the meantime, the following principles can help UX designers improve how they design data-centric user interfaces.
- Use multiple sources of data. While big data is the main driver of a user’s experience on data-centric apps, it only paints half the picture. Big data is made up of large machine-generated quantitative data sets, such as site analytics, shopping history, social media activity, and website click maps. This sort of data helps inform design by explaining how users move around the Internet, but it won't help explain why they make certain decisions. This is where thick data comes into play. Thick data is qualitative data, such as insights from user interviews and focus groups, and quantitative data like online surveys. Using both big and thick data is the key to understanding both the ‘what’ and ‘why’ of user behavior.
- Think of data as a storytelling device. UX designers need to craft a compelling narrative to help users find relevant patterns and trends. Storytelling is also important to steer users through large amounts of information. Rather than having a ton of information fighting for the user’s attention, content should be prioritized and support the underlying narrative.
- Know your user’s abilities. Establish the right level of complexity for your target audience. Even within a specific user segment, users will have different levels of analytical skills and experience and will process and engage with information in a different way. The interface should be flexible enough to accommodate everyone, from first-timers to power users.
- Trust your user’s abilities. Don’t shy away from presenting dense information because you worry that users won’t understand it. As long as the information presented is valuable, useful, and conducive to aid the user’s understanding of the information presented, they will be willing to engage with complexity. The right amount of control and support can encourage users to explore more advanced ways of interacting with the data.
- Establish clear paths to discovery. Define how users should engage with the information presented. Ben Shneiderman’s visual information seeking mantra provides a great framework for designing data-centric user interfaces.
- Overview. Providing an overview of the entire information set can very helpful. It allows users to see the bigger picture and find patterns and outliers in the data.
- Zoom. Once users identify interesting clusters of information, they need the ability to focus in on these areas and examine them more closely.
- Filter. Filtering out uninteresting items is an important part of the discovery process. Give users the ability to reduce a large data set by ruling out things they do not want.
- Details on demand. When users have honed in on a particular item of interest they will want to examine it in full detail.
- Relate. The data presented is more meaningful when users are able to view relationships and compare information.
- Minimize UI distractions. Every element on the screen should help users complete their tasks. Allowing users to click, drag, and swipe content objects can provide a more tangible and immediate way to interact.
- Make data social and relatable. Data can help users feel connected to a larger community by providing information on how other users engage with the platform. Interfaces should deliver a more fulfilling experience by using data as a social connector.
- Make the future your priority. There are many data sources that allow us to predict what a user’s next move may be. Data should be treated as a living, breathing element within any user interface with the goal of optimizing the user interface and streamlining user experience.