Benjamin Wiederkehr is an Interaction Designer with a focus on information visualisation and interface design. With his work, he simplifies complex data and tells stories with a meaningful impact. Benjamin: “The designer’s job is done when the reader is able to intuitively understand the underlaying phenomena expressed with the visualisation”.
Graphic Hunters: What is the power of a good visualisation? In what way can a visualisation help to understand or to communicate information?
Benjamin Wiederkehr: First, it’s important to note that «good» will mean different things under different circumstances. The quality of a data visualisation should be evaluated by considering not only the graphical representation but also the target audience for it, the context where it is deployed, and the goals the author hopes to achieve with it. When we evaluate visualisations we typically do this with four dimensions grouped into Data and Design. Data Relevance examines the importance of the data for the specific context of the user. The data must hold substantial meaning for the user and must motivate her to actively engage with the visualisation. Data Integrity evaluates the accuracy, consistency, and honesty of the data source and its provider. Design Form analyses if the data is presented appropriately. The presentation needs to be clear, readable and holding up to high aesthetic standard. Data Function assesses the overall user experience as a combination of usability, simplicity and immersion.
I have read that someone has to understand a visualisation within seconds, others say visualisations should leave something to explore. What is your opinion on this?
Again, we believe that this should be answered by considering the specific use case. Let’s assume that a visualisation will live on the web and is targeted at the general public like it would be the case for an online newspaper. Then I would stick to the 20/20 rule: The visualisation should be able to get a reader’s attention and intrigue her curiosity within the first 20 seconds. At the same time, the visualisation should have enough depth and provide enough context for the reader to spend the next 20 minutes exploring it. But this is just one use case among many. Others can be seen somewhere on a spectrum between immediate understanding on one side and extensive exploration on the other.
What is the first most important question one should ask before starting visualising the data?
Why. Why should this data be visualised? Understanding the purpose of our work should be the foundation for every decision we take further down the road. I believe starting with why is important not because the answer will be all you need to know in order to do good work, but because it will lead you to more detailed and equally fundamental questions like: For who should the data be visualised? When and how will these people use the visualisation? What are they expecting to learn from this visualisation? Knowing the answers to these questions will enable you to answer higher level questions like what is the right medium to visualise the data and which is the right visualisation technique to apply. Following this path of questions and answer will eventually equip you with the necessary information to make all design decisions with ease.
What tools are important to learn when you want to work with visualisations?
Personally, I always start with drawing sketches using pen and paper. Although very basic, sketching is an essential tool to ponder, draft, and discuss speculative potential of diagrams and visualisations.
What is the role of design in a visualisation?
At our studio, we refer to the design of a visualisation, or of any artifact for that matter, as the sensible combination of form and function. Aesthetic requirements need to be neatly integrated with functional requirements to render a good design solution. Becoming overly attached to one or the other can hurt the overall elegance of a design solution. It might leave the user with something that works, but is undesirable or something that’s beautiful but not useful. The goal should be to combine these diferent requirements and create a solution that’s both useable and desirable. The role of design in data visualisation is to optimize the visual display for the human’s perception and interpretation. The designer’s job is done when the reader is able to intuitively understand, that is to recognize efortlessly and to interpret correctly, the underlaying phenomena captured with the data and expressed with the visualisation.
What is the secret on how to balance design, data and the story?
As mentioned above, I believe that there are certain requirements for the data and the design that need to be satisfied in order to create good visualisations. I don’t see these requirements in conflict with each other, but I agree that it needs attention to find the right balance between them. Now, if we consider the story to be a third component in the creation of a data visualisation, we should again tie the question about the balance back to the question about the audience and their context. Very few visualisations tell one single story. Most of them tell multiple stories at once or different stories to different people. Oftentimes the reader will discover their own stories when interacting with visualisations. At Interactive Things, we’ve found the following tactic to be helpful for creating effective visualisations.
First, we provide the user with an introduction into the topic, the data, and the visualisation technique. This could be done in the form of a written story, a set of annotations, an animated buildup of the visualisation, or any other technique to make the user feel at home. The introduction might be a great place to highlight a first takeaway that can be gleaned from the visualisation. This way, we combine a helpful introduction with an insightful interpretation.
Second, we allow the user to freely explore the visualisation by herself. Where appropriate we might let her select and filter the data independently, personalize the visualisation to match her circumstances, or customize the graphs and charts to view the data from a different point of view. This exploration of the data should be guided enough that the user does not feel lost and free enough that she does not feel constricted.
Third, we enable the user to take action and apply the gained understanding in a way that matches their use case. This can be as simple as sharing the insights with their network or exporting and storing it for further use. But it could also be as elaborate as engaging in a dialog around the insights, reaching out to the data provider or visualisation creator, or taking action in a political, economical, or social manner. Here is where we connect our tactic back to our initial discussion of the importance of the question why we should visualise a data set. The type of action and its resulting impact should thought through and planned for from the very beginning of the creation process.
What lessons do you want the participants to take home from the training?
With our training we enable the participants to build visualisation systems in a more user-centered way. We will introduce the tools and methods we use to do this, explain how they work and how to apply them, and we will do this in a very practical manner with examples and scenarios. Ideally we will be able to share with the participants our enthusiasm for considering the people first, before answering questions about data, design, or technology.