Introduction

Data Science Central (DSC), a web-based Wiki-like hub for data professionals, hosted a webinar on data dashboard design. The webinar was sponsored by Tableau, hosted by Bill Vorhies, editorial director of DSC, and featured a talk by Alyshia Olsen, a UX designer at Tableau. Following the talk was a Q and A session where the host relayed questions posted digitally by the attendants of the webinar.

It's not you fault

While positioned as a dashboard design webinar, the crux of the talk was mostly focused on user-centered design strategies, though it was filtered through the process of designing data dashboards. Right from the start, Olsen brought up a concept that is central to the work of user-based designers, and somewhat codified into the space by Don Norman in his book, The Design of Everyday Things (2013): Frustration with a product is not the user’s fault, it is the designer’s. This idea and the idea that user experience design is all about the psychology of the user are the backbone of the talk reviewed here. Olsen went on to spend the bulk of the hour discussing a design-thinking process for dashboard design, in four sequential phases:

  • Discover Phase
  • Distill Phase
  • Ideate Phase
  • Validate Phase

These four phases are now discussed in depth.

User-centered Design

Discover Phase

In user-centered design, the Discover Phase is not about discovering what the system needs to do, as in system-centered design, but is instead about discovering who the system is for (Talja and Hartel, 2007). For Olsen, the Discover Phase is just that, a phase for discovery about the intended users. She uses the example of building a dashboard to support her father at his business as a way to frame what the Discovery Phase should yield to the designer.

Olsen states that the Discovery Phase for dashboard design is the opportunity for the designer to discover what the target users really need, not what they think they want built. The mantra for this phase is “question the obvious,” and it is manifested in the concept of the Five Whys, or, ask ‘why’ five times. The Five Whys method is a form of root cause analysis widely in use today to understand the deeper drivers of needs, actions, and results (Norman, 2013). It implies that even after you as a designer have found a reason for the wants and needs of the users, you should keep asking why until the root need is reached. There are many methods for employing this technique in user-centered design, but the outcome is the same: the designer understands the underlying issues and needs that contribute to why a user needs something designed.

An example of asking Why 5 times to understand the underlying need

Distill Phase

The Distill Phase is the process of taking all of the notes, learnings, ideas, and anything else gleaned during the discovery phase and distilling it all down to answer three questions:

  • Who are your users?
  • What is the primary goal of the dashboard?
  • What does the dashboard absolutely have to do (to satisfy the first two questions)?

During her review of the Distill Phase, Olsen gave insight on a number of strategies a designer could employ to distill their notes and answer the three main questions. These included Affinity Mapping, Pattern Identification, among others. The main goal of all of these strategies gets at the main crux of this sort of design: What do your users need? Designing for the user instead of for the system is a fundamentally different form of design, and Olsen’s entire process starts with the users, understand their needs before she even begins to design a product.

Ideate Phase

The third phase is the process of idea generation, rejection, and acceptance (with the caveat that ideas get iterated on). This is where a designer will come up with ideas to address the last question from the distill phase, while not violating any of the tenets established by the answers to the first two questions. Again Olsen presented a few different strategies for ideation, but I only want to mention one of them here, Blue Sky Thinking.

Blue Sky Thinking is an ideation strategy where a designer comes up with a basic idea for the design of something. This idea is known as a ground idea. Next, the designer comes with a very ‘out there’ idea, potentially something impossible to implement and is more of a dream. The results from this strategy come from combining elements of the ground idea with the blue sky idea to reach a middle ground. This sort of ideation strategy can push the designer into more creative solutions.

Validate Phase

The Validate Phase comes after something has been designed. This phase is centered around feedback and usability testing. It’s in this phase that a designer will be able to understand where they’ve succeeded and where they’ve failed, allowing for a return to the Ideate Phase to improve upon the product. This phase once again directly involves users, since watching and listening while a user uses the product can result in valuable feedback and improvements to the design.

Conclusion

As a Data Analytics and Visualization student, the dashboard design element of this talk was the bulk of my interest, and I was not expecting something as process oriented as the talk ended up being. However, having an expert walk through the process of designing dashboards from a user-centered perspective was enlightening. According to Norman in his book ‘The Invisible Computer’, one of the biggest problems in HCI is the fact that computers are rigid, precise, and thus require rigid and precise inputs. Humans are inherently imprecise, flexible, and ambiguous (Norman, 1999). Seeing the process of user-centered design outlined in such way has made obvious to me how attempting to design from the human up, considering all of the flexibilities and ambiguities inherent in us can lead to better, more flexible systems that work the way we need them to instead of us working the way machines want us to.

References

Norman, D. A. (1999). The invisible computer: Why good products can fail, the personal computer is so complex, and information appliances are the solution. Cambridge, MA: The MIT Press.

Norman, D. A. (2013). The design of everyday things. New York: Basic Books.

Talja, S., & Hartel, J. (2007). “Revisiting the user-centered turn in information science research: an intellectual history perspective” Information Research, 12(4) paper colis02. [Available at http://InformationR.net/ir/12-4/colis02.html]

Leave a Reply

Your email address will not be published. Required fields are marked *