Resuming the DataViz 101 series started last year, I want to revisit some basics for data visualization, showing when we get value from having numbers visualized, and when such a visualization is inappropriate. One of the main reasons that data visualization exists at all — be it smooth infographics, or slick project reports — is the fact that it saves us time needed to digest some quantitative information, i.e. the information that has numbers in it. Visuals present numbers in an appealing way, making them easier to read. Sometimes, however, they use visualized numbers with no substantial ground. If no meaning is ingrained into the graphical cuteness, a visual would make no sense. Some other technique for information rendering has to be used then, such as a text.
Take a look at one such case where numbers pretend to be visualized with some meaning, while actually failing to provide real value to people who look at them.
One can see this pattern with numbers highlighted quite often on web-sites for conferences or gatherings. Such a visual is supposed, presumably, to convince potential attendees that this conference holds some value for them. However, I don't see how it will help decide if a conference is worth attending or not. There's no universal converter that would work for each and every individual, and translate those hours of keynotes, workshops, trainings, and the count of speakers into a meaningful answer to one question: "Will I learn something new and useful for me personally at this conference?" How are these flat numbers capable to attend to the unique knowledge landscape of any given individual? No way, they can't do it. Those people who are looking to decide for themselves if a conference is worth attending or not, might as well skip this "hippish" part with meaningless numbers, and proceed straight on to the text piece about the speakers, keynotes, workshops and training. Bad news for someone who did this visual: they've wasted both their time, and the time of the site visitors.
Here's the other example that shows how visualized numbers can help in project management:
This is a sparkline report, and while it includes numbers that seem to hold no meaning to an external observer, an insider who looks at the graph is likely to know the project context: how user stories and bugs are sized in general, how much effort does it take to have them completed, and how these numbers can be rendered into a diagnosis report on the project health. Compare the sparkline graph and this text: "This report covers the last 16 weeks. Designers had their backlog full with 13 user stories in the first week, with fewer and fewer new stories added in the next weeks. They completed 3 stories, and had 2 more added to their backlog in the current week". Of course, the sparkline renders this info in a more compact and time-saving way.
As a summary, before we hurry to create a visual report, or an infographic with numbers, we need to consider if a user or a reader will get the info they want fast from this visual. Some information can be rendered best as a piece of text, like in this first example from a web-site of a conference. Words would have taken readers to the core of the matter faster. In the second example, it's the other way round. It would take more time to convey the same information in words.