Let me begin by asking the following: What do you think defines a data visualization expert?
Think hard about this, because your answer may simply reflect that you need a person who can help improve the type of visualization products or output you currently create; or it may reveal that you are searching for a person who can help you create a new type of product altogether. Either way, a dedicated data visualization expert may not be someone you need to hire, because odds are you already have people who can make good data visualizations but don’t yet know it.
Here’s how I define “data visualization expert”: A person who can process and interpret information in order to visually communicate it to an audience effectively.
Using this definition, you probably have people who can process and interpret data already, so the way to turn them into data visualization experts is to teach them how they can best communicate their data, their models, and their findings to an audience. (By the way, I’m not sure the term “expert” is even appropriate when it comes to data visualization. The field encapsulates so many different skills -- data processing, programming, design, and communication -- that I wonder whether anyone really can be an expert.)
The problem with turning over data visualization tasks to a dedicated person is that the people who get their hands dirty with data and their models best understand the results and thus the story. Turning over responsibility to someone who may not understand the intricacies of the data and models asks for misinterpretation and biased presentation of results.
I recently worked with analysts from a large agency who expressed dismay at the fact that once they finished creating their output and sketching their visualizations, the “graphics team” would modify their work in ways it saw fit. That is not a good way to effectively communicate to an audience -- the people who work with the data should direct the construction of the visualization.
So how do you turn your analysts into data visualization experts? Have them read books and blogs or maybe take a course (perhaps online). They may need to learn a new software package such as Tableau or StatPlanet; or they may need to learn a new programming language such as d3, Processing, or R.
You'll also find visualization capabilities from providers of standard programs used by analysts in my field -- SAS and Stata. Remember, if the tool doesn’t help you create the visualization you want, you need to get a new tool. Most importantly, encourage them to think more strategically and more creatively about presenting data.
Where else can your analysts start? Books by data visualization experts Edward Tufte, Stephen Few, Naomi Robbins, Alberto Cairo, and Andy Kirk, among others, are great places to start.
Creating good visualizations will require some software skills, but online tutorials are abundant these days: You can find great Excel tutorials by Purna “Chandoo” Duggirala and John Peltier; great R tutorials by Anthony Damico and Nathan Yau; and great d3 source code and tutorials by Mike Bostock, Jerome Cukier, and Scott Murray.
Data visualization courses, workshops, and consultations are available from myself, Alberto Cairo, and Andy Kirk, not to mention a growing number of online -- often free -- courses. And an abundance of presentation skills resources are available from such authors as Nancy Duarte, Carmine Gallo, and Garr Reynolds.
Presenting data effectively is not rocket science: It requires a strong sense of good visualization techniques and communicating through data, both of which can be learned. Not everyone can be a great graphic designer or an interactive programmer, but any analyst who is familiar with data can create great, effective visualizations.
Do you agree or disagree? Read Hyoun Park's Point piece and share your thoughts below.
4 Objectives for Your Data Visualizations
SAS Visual Analytics: See Demos Galore & More
10 Data Visualization Thought Leaders