Follow Noah on Twitter @noahi
This is a continuation of a series of posts covering the Four Pillars of Visualization. Please read my previous article for an overview of the four pillars for an effective visualization.
What is your purpose? Why are you here? And what do you want to accomplish with this visualization?
These are the sorts of questions you must ask yourself before you begin drawing your visualization*. In fact, defining the purpose and selecting the content are critical, and often overlooked, steps in the visualization design process. We must understand what we're visualizing before we figure out how to visualize it. Defining your purpose helps you figure out what that whatis.
This posts addresses creating visualizations for presentation. If you're in the exploration phase, you should just try a bunch of things and see what works and not worry about process yet.
There's a reason the purpose phase comes before the other three phases. A well-defined purpose not only gives you a target at which to aim, it provides guidance to start in the right direction and keep you on track along the way. It will tell you what data is important to include and what doesn't matter. It will tell you which relationships to emphasize, and which aren't as relevant. It will tell you which data points to highlight and which aren't as interesting.
A useful purpose must take into account the following questions:
**Why am I creating this visualization?
**Who is it for?
**How are their needs, understandings, biases, vocabulary, interest level, and availability different from mine?
**What actions are they trying to take, and what do they need to understand or learn from this visualization?
**What circumstances, biases, and technologies will affect how they consume it (e.g., black and white printed report, small screen, huge screen, executive summary, in-depth seminar, etc.)?
The more you know about your customer and how they will consume your visualization, the more clear and accurate your can make your purpose, and the greater your odds of success.
So what does a good purpose look like?
Specific and detailed; think of it as a specification document for your visualization. A graph that is intended to “show the sales figures” presents a much broader (and therefore less useful) target than “show the sales figures for our flagship product, for the top three and bottom three regions, for the last 8 quarters, in a format we can use online and in print.”
Note how the second example specifies boundaries, key data, and usage considerations? That will not only help you get past the dreaded blank page, it will give you a standard to test against as you make design choices.
To be most successful, different uses (purposes!) will probably require different implementations. For example, if I'm driving from Seattle, the map to get me to New York and the map to get me to Baltimore look almost identical. However, the purpose is different enough that I really do need a different set of directions to get me there. Note also that a more general purpose of “drive to the east coast,” while accurate and concise, is highly unlikely to be sufficiently specific as to be useful.
Luckily for us there are a few very common general purposes when it comes to graphing data. Check out the headings from IBM's Many Eyes graphing tool.
Those headings, about seeing relationships, comparing values, etc., are fantastic statements of purpose. It's likely (but not quite guaranteed) that your purpose falls under one of those headings. If you're having trouble defining your purpose, consult this page and think about how one of these purposes may fit your needs and data. Not all needs will be met by one of these purposes, but they do encompass the most common graphing scenarios.
Finally, remember that your purpose will probably change as you proceed with your design process. You may discover new needs, or see something unexpected that should be revealed. That's a good thing; your purpose will evolve with your understanding. Once you have figured out your purpose you can begin selecting the content of your visualization; that's the topic of my next post.
For further discussion on this topic, download my recent whitepaper, “Choosing visual properties for effective visualizations.” In the whitepaper, I’ll discuss the huge number of design decisions you need to make upfront before creating your visualization that will impact on the ability of the visualization to communicate knowledge accurately and efficiently. This paper addresses one key aspect of the design process: how to choose an appropriate visual property (position, shape, size, color and others) to encode the different types of data that will be presented in the visualization.
Why stop the insight with this article? Visit IBM’s visualization hub, IBM Many Eyes and join over 100,000 like-mined visualization enthusiasts, academia and professionals, including additional insights from Noah Iliinsky and other IBM visualization luminaries.