A thought has been bothering me for the last year: How much value do dashboards intended for public consumption actually create?
In my experience, many creators turn to dashboards in lieu of developing an explicit argument or telling a story with their data. Instead, they assume their user will do that through exploration. The argument, it seems, is that if users have all the data for all of the possible outcomes at their fingertips, they can figure out how to understand it.
This kind of exploration can be useful for some audiences in some contexts. Users do benefit from dashboards on, for example, government agency websites, websites showing public health data, websites that make data more transparent, and other targeted projects geared toward specific kinds of users. And clearly, internal audiences—especially those working with data in real-time—will find dashboards a useful, if not indispensable, in their work.
But for the more casual user, I’m leaning towards the perspective that people simply don’t explore data dashboards—they don’t select, filter, and search. I would even go so far as to argue that casual users can include the people who the dashboard is meant for: those who are very interested in the content and the data.
Our media habits are changing
Our media consumption habits have changed so dramatically in recent years that most users are unlikely to explore data. Instead, users—even the target demographic—want quick stories, analysis, and statistics they can comprehend and share quickly and easily.
People are now more likely to skim through content. Their attention moves in and out. This makes dashboards—formally defined in The Big Book of Dashboards as “a visual display of data used to monitor/conditions and/or facilitate understanding”—a tough sell.
In a recent appearance on the Joe Rogan podcast promoting his new Netflix film, The Rip, Matt Damon pointed out that viewers give a “very different level of attention” to a movie at home versus in a theater. As a result, their Netflix producers pushed the action scenes closer to the beginning of the movie and, importantly, encouraged their stars to reiterate “the plot three or four times in the dialogue” to remind people whose attention may have slipped. It’s worth watching just a couple of minutes:
Damon wasn’t arguing this was necessarily a good thing—he pointed out that this kind of viewing behavior is “going to really to infringe on how we’re telling a story.” Others have made similar arguments, arguing that Netflix is dumbing down content so viewers can keep multitasking and scrolling.
But this is the reality of the attention economy. We scroll. Movies are made to support simultaneous watching and scrolling. So we scroll more. It’s now the way we consume media, for better or worse.
This direction of change is not new, including in the data visualization field. In 2016, Archie Tse from The New York Times and Gregor Aisch at the Info+ conference both noted that news organizations were reducing the number of interactive graphics they produced because users were more likely to scroll than click. Interactivity was reserved for stories where it added clear value.
Today, a decade later:
- A growing share of people get their news from podcasts and email newsletters, neither of which are conducive to interactive data visualizations.
Dashboards assume a level of attention, screen size, and patience that increasingly does not match how people consume information.
The fundamental mismatch
Dashboards are built on a major assumption: If we give people access to the data and enough tools to explore it, they will find what they need.
But I don’t think most people want to explore. I think they want to understand—and they want to understand quickly and easily. They want to make comparisons that matter and quickly see the takeaway message, its implications, and the solution, if applicable. All while on their phone.
In the dashboarding world, the work of interpretation falls on the user. And there is value in that—people can bring their own interpretation and context to the data. But I believe people want the expert to facilitate that interpretation and provide them with an explanation.
How AI may make traditional dashboards feel obsolete
Emerging AI tools will make this tension even sharper. Instead of navigating filters and dropdown menus, users will want to ask specific questions, like “What’s the job market like in my county compared with similar counties?” (Again, there may be a big difference between dashboards intended for internal versus external audiences here—corporate governance, procurement rules, and human resources being just a few things that differentiate the internal from the external.)
A system connected to the data will generate the comparison, the visualization, and the explanation. The future of dashboards may not be interfaces people explore through dropdowns and filters, but data backends that AI systems query to produce tailored explanations. In that world, the dashboard itself is no longer the product: the data and the interpretation are.
What should we build instead?
If the creator’s goal is to enhance users’ understanding of data and act on it, we should consider investing more in:
- Explanatory visual—and, importantly, video—essays
- Guided comparisons that answer specific questions
- Story-driven graphics and reports
- Products that provide context and interpretation
- Clear discussion of implications and potential solutions
These products do not ask the user to do the analytical work; they do that work for the user. Then, maybe, a dashboard could be provided for those who want to go further.
Do you really need a dashboard?
Dashboards feel powerful because they give users control. But for many public communication goals, dashboards may not be the best use of our limited time, money, and effort. They require the user to do a lot of work to get to an answer we as creators can give them more quickly.
This is not to say that static graphics, like infographics or slide decks or social media posts, require less time to create. In fact, they may take more time to create because they require the creator to distill the analysis to a core message that meets their users’ needs. But if those alternative products more effectively make information actionable for the people who need it, that tradeoff might be worthwhile.
The question is not whether dashboards can be useful. The question is whether they are the most effective way to help people understand data and make decisions. And for many public-facing projects, I think the answer is no.
If you enjoyed this article, consider subscribing to the PolicyViz Substack newsletter. Subscribers get early access to posts, can join the discussion, and receive additional insights on data visualization.
