by Frederik Ruys
A short guide to visual storytelling
With an explosive amount of data and the insatiable need for visualizations it can be hard to find the best tool to share your story. But besides the proper tool, it’s also about the correct approach. How do you map impact with your data in our age of information overload?
Beautiful images but no story to tell: the internet is loaded with artistic, computer-generated data visualizations. Colourful but incomprehensible eye candy, based on big, multivariable datasets provided by companies and institutes who hop on the hype of data transparency.
The biggest disadvantage of these beautiful visualizations is that the actual message does not get across.
Of course it all depends on the exact purpose of the image. If it’s just for research or investigation purposes, a data dashboard is fine. But in all other cases; if your goal is to inform and/or convince your audience, your visualization becomes part of a storyline.
Luckily it’s not that hard to elevate your map from a data dashboard to visual story.
Obviously it all starts with gathering the data. Arranging and cleaning the set is an important first step in making the content more accessible. But despite the neat structure of rows and columns it’s nowhere near information; it’s just data.
Information is data in context. In this case the set appears to contain locations, defined by latitudes and longitudes. So picking a map projection to plot your dots is an obvious start. And by adding a key the visualization becomes informative.
To add knowledge to your map, you need to know what these locations represent. You might reach out to an expert in the field to understand the events and to decide what kind of visual forms (size, shape or colour) fit the story. By adding labels your visualization is already more appealing.
One data set might not be enough to give insight. So by combining your data with additional layers of information offers the ability to analyze the events from a broader perspective. This is the functionality that most dashboards are limited to. Unfortunately this approach is too complex for a wide audience as users have to analyze the data by themselves.
To better serve your public you need to make clear choices. Not by simplifying your message, but by clarifying your data. This can be achieved by writing a clear title (expressing the urgency) and some proper annotations. And, if it concerns an animation or a sequence of slides, by building up your story step by step. This way you lift your visualization from a dashboard to a visual story that has an impact on your stakeholders.
Knowing the difference between dashboarding and visual storytelling, it now comes down to finding the appropriate visualization. In the newsroom we commonly use the 6 basic journalistic questions which all represent a certain type of image.
WHO is it?
Visualizing people or organizations, by using symbols or icons.
WHERE is it?
Plotting location-based information like geographical data on a map.
WHEN did it happen?
Using a timeline to project chronological information.
WHAT is it?
Illustrating an object, like a technical cutaway, often based on qualitative information.
HOW does it work?
Reconstructing a process or an event in a sequence, like a comic
HOW MUCH is it?
Visualizing quantitative information, like charts and diagrams based on statistical data.
But most visualizations reveal the answer to more than just one question. Therefore I’ve been working on a matrix which shows the result for each combination.
For this blog we’ll focus on the different types of maps, therefore the combinations with WHERE (all visualizations connected with the green line). These geographical visualizations may look like this.
WHERE and WHO
To show where people are living, you’re plotting people on a map.
WHERE and WHEN
When your dataset consists of both geographical and chronological information, you’re able to visualize tracks – to reconstruct for instance a flightpath or an escape route.
WHERE and WHAT
When you combine the where and the what, you’re mapping points of interests – like restaurants, ATMs or museums.
WHERE and HOW
To reveal how locations are connected – like the relocation of people or connecting phone lines – you combine the where and the how.
WHERE and HOW MUCH
To show quantities within an administrative area, like election results, you’re generating a choropleth.
Therefore; finding the proper visualization is not rocket science, it’s just the combination of the basic journalistic question and the structure of your dataset. And as long as you use a relevant title and informative annotations, you can be confident your message will get across.
But did you ever wonder how visualizations may look like when combining multiple questions? This polyhedron reveals more combinations.