Sign up for our FREE Power BI Bootcap
Technology has allowed us to access, gather, and stockpile data at levels never seen before. Some call it the "new oil", a term I only partially agree with. Regardless of its popular classification, the rise of data brings forth an unforeseen frustration, our lack of ability to make sense of it all.
You've likely heard people talking about telling a story with data, yet what that means needs to be clarified. We all understand a story to follow a narrative and a narrative to be the recount of a series of events in an order that connects them (thanks, Oxford Dictionary). We commonly see narrative playing out in the written or visual media, so what about data?
If you're trying to portray a narrative within data visualisation (as opposed to building an exploratory data visualisation), we will begin by borrowing strategies from visual media. And like visual media, compelling data visualisations share two significant attributes, solid narrative structure and intriguing visual narrative elements.
A visualisation's narrative structure is the framing of design elements used to convey an idea, namely the ordering of visualisation elements (linear or user-directed), much like a movie has a beginning, a middle, and an end. It also includes the degree of user interaction and control over the story and the use of messaging (annotations, introductory and summary statements) - choose your own adventure, anyone?
Visual narrative elements refer to the strategic graphic design elements used to build a narrative experience. These design elements direct a user's attention to important events within the story, create transitions between those events, and position the user within the story.
Let's explore these elements further.
The first and most widely used visual narrative element involves summarising and drawing viewers’ attention to key story elements. Annotations can appear as text overlaid on or next to plots, accompanying documentation, or audio narration.
The Economist relies heavily on annotations to explain complex data visualisations.
Visual highlighting is any method that draws attention to critical observations, statistics, outliers or trends in the data. As displayed here in Watkins' Arctic Ice Fluctuations visualisation, the distinct green draws your attention from the grey, conveying a contrast of importance between the different lines.
Image courtesy of The New York Times.
Matching refers to the consistent use of visual elements to show the relationships between different sections of data visualisation. For example, in the case below, David Mccandless uses the traditional red and blue to group unique traits of the political spectrum.
Image courtesy of Information is Beautiful.
Progress bars are a visual indicator of the progress of a narrative that can be used in various contexts. For example, storybooks use numerical pages to measure progress and find specific information, movies use linear play bars to report time and progress, and narrative visualisations incorporate progress bars for navigation. Progress bars are thus a valuable tool for reporting and tracking the progression of the users through different events (think PowerPoint presentations, slideshows, etc.).
Consistent visual platforms: Contents within the platform are changed, while the general layout remains consistent. This creates efficiency and predictability. A superb usage example of consistent visuals appears in The New York Times, authored by Kevin Litman-Navarro.
Image courtesy of The New York Times.
Narrative visualisation utilises multi-dimensional messaging such as text, annotations, and graphics to enrich the story. This allows further detail in a visual format to accompany the narrative, creating compelling and memorable visuals with impactful messaging. It also helps to distinguish narrative data visualisations from exploratory data visualisations.
Extensive annotations and text can overwhelm a viewer. Interactive data visualisations can overcome this by providing details on demand. This includes tooltips and information appearing while hovering over data points, drill-downs, and drill-throughs.
Image courtesy of Automated Analytics (us!)
Sliders are a common interactive element used in data visualisation that encourages viewers to transition back and forth between data at different points in time. Sliding and transitioning the visualisation allows the viewer to control and experience a trend unfolding before their eyes.
Image courtesy of USGS.
Many narrative visualisations include interactive features requiring the user to interact with controls and data visualisations that they might need to be more familiar with or be immediately aware of. Think welcome tours - skip this?
Image courtesy of Automated Analytics (us!).
Semantic consistency refers to the consistent use of visual aesthetics to encode information. A clear example is the constant use of colour and size to represent a variable across related visualisations. This aims to create predictability and efficiency for the viewer as the representation of variables remains consistent.
Image courtesy of The New York Times.
Identifiers of Interactivity: Interactivity is an essential component of narrative visualisation, as it allows the viewer to engage with and experience the story more deeply. To draw attention to this feature, designers should use techniques such as highlighting, animation and providing a simple instruction tutorial. For example, requiring viewers to enter their city and age before progressing with the story ensures they can take advantage of the interactive elements.
Animated Transitions: Animation supports narrative visualisation by providing a visual cue to the viewer when a change or update occurs. When creating a narrative visualisation, one should aim for fewer animation techniques than possible; each piece should be carefully chosen and justified concerning the overall story being told.
There's a fine line between too much and too little when considering visual elements, and lucky for us, there are a few ways in which we can assess the effectiveness of visualisation to serve our end users better. We'll discuss these in part 2 of our Data Visualisation Strategy.
What are your favourite visual narrative strategies?
