audience, data journalism, data visualization, graphic design, infographic, Interactivity, journalism

What I learned at Jan Willem Tulp’s workshop at the NODA16

Lacking data visualisation design skills can lead to distorted data and prioritising aesthetics before information. For this reason, Jan Willem Tulp organised a data visualisation design workshop at NODA and Tutki! 2016, a joint conference held last weekend in Helsinki.

Tulp’s workshop focused on improving the visual literacy in order to critically evaluate our and other people’s work. Below are five takeaways to keep in mind when designing a data visualisation:

1. Being an expert in a tool is the starting point

Even though tools are important to create visualisations, using them perfectly does not guarantee the “correctness, quality and aesthetics of your design.” It’s important to know why do we want to represent the data in a particular manner that will attract the audience.

The following maps show the same information – the US wind- through different tools. However, the design of the first map makes more difficult to explore the data and seems less likely to invite the reader to explore it:

2. Put the data in context

From Tulp’s point of view, there are three elements that create a good data visualisation: the data (what do we want to tell), the concept (why are we using a visualisation) and the design (how the data will make sense).

Data can help to tell some story, but journalists have to understand it before turning it into some representation. In the following picture, numbers or text are pointless if we don’t know their meaning:

3. Graphs, sound and taste

Visual representations can help us understand the world. However, we are not only talking about graphs but other channels such as music or taste. Below are some examples:

Fractions of a Second: An Olympic Musical by The New York Times:

Screen Shot 2016-04-26 at 15.50.01

The Data Cuisine, an interesting project on expressing 🍴 as a means of 📊 expression:

output_sXhgNv (1)

4. Create alternative designs

Learning by doing. A good exercise to practice data visualisation design skills is to gather graphics that have already been published and redesign them. This will help us to see what does or what does not work, and how can the visualisation be improved.

WTF Visualisations is a website that gathers graphics that make no sense. The following pie charts were suggested as bad data visualisations, and Jan Willem challenged the audience to redesign the second graph around money and football:

5. No definition of ‘interesting’

What makes a data visualisation interesting? People differ on this concept and it has changed throughout the years.

To explain that, Jan Willem showed two interactive visualisations at the beginning of the workshop and asked the audience to pick up one. Both representations gather plenty of data and present it as a story. Which one will you choose?

U.S. Gun deaths created by Periscopic:


Drone strikes victims in Pakistan, created by Pitch Interactive


However, the audience can engage with the design part very different. More than half the class chose the second interactive. (I didn’t. I still think that the first one created by Periscopic is easier to follow and extract the story, without looking for extra information).

Even though there are plenty approaches to define what is interesting, Jan Willem said that any data representation needs novelty and comprehensibility to caption the reader’s eye:

More about Jan Willem Tulp

He defines himself as a data experience designer creating data visualisations for his company, Tulp Interactive.

Follow him on Pinterest, he gathers plenty of sources and good visualisations.

Desing vs art

Jan Willem Tulp says that design has to be used to represent meaningful data in a story. However, design and aesthetics in visualisation have been discussed (and is still going on) by many experts in this field.

Andrew Vande Moere and Helen Purchase investigate the requirement of attractiveness in visualisation on their publication On the role of design in information visualization. They conclude that visualisation design is not an ‘add-on’:

“Design reasoning is not an activity that can be added ‘later’ after most other decisions have been made; visualisation developers need to be conscious of the role of design at the outset.”


3 thoughts on “What I learned at Jan Willem Tulp’s workshop at the NODA16

  1. Pingback: Vizually | What I learned at Jan Willem Tulp’s workshop at the NODA16

  2. Pingback: What I learned at Jan Willem Tulp’s workshop at Tutki! 2016/NODA16 | Online Journalism Blog

  3. Pingback: AUDIO: “The best data journalism combines traditional reporting with data work” says Simon Rogers | dinfografia

Leave a Reply

Please log in using one of these methods to post your comment: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s