4 Quick Tips for Data Visualization Newbies


You know you need to get more visual with your data. That's been a big theme for 2013. But just because you're comfortable working with data doesn't mean you know beans about presenting it visually -- and that may have you scared.

The last thing you want happening is for business users to dismiss the insights you've worked so hard on because your heat map is a mess or your bar charts too confusing. That's just as bad as tossing a massive spreadsheet at them and saying, "Here, you find the pattern."

So yes, the thought of creating visualizations can be daunting for many data analysts. Panicking won't help; preparation will.

Taking the time to learn the right way to present data visually will keep your business users from seeing red over the work you produce. For advice on how to get started, we turned to Tricia Aanderud, president of BI consulting firm And Data Inc. and writer of the BI-Notes blog. Aanderud, who specializes in helping companies transform their data into meaningful visualizations, joined us earlier this week for an All Analytics Academy lecture on the "what, when, and why" of data visualizations. (You can view the lecture on demand).

Here are her four tips for getting started with data visualization.

Study with the experts. This discipline comes with its own set of gurus, folks who have been studying the art of presenting data visually long before "data visualization" became this year's buzz. These include, but are not limited to:

Read their books. Watch their YouTube videos. Check on the visualizations they have on their websites, Aanderud said.

Reproduce visualizations. Head to the web, find data visualizations that interest you in some way, and play around with recreating them. Here's one, for example, that Aanderud shared during her lecture. Can you spot what's wrong? How would you do it differently?

Click here for a larger version: www.visualizingeconomics.com.
Click here for a larger version: www.visualizingeconomics.com.

Experiment. Download some data (there's enough available from public sources) and play around with it. "There are free tools out there, but Excel works just fine" for this purpose, too, Aanderud told us. "Look at the data, ask questions, and figure out, 'OK, can I display it? How would I? Does a line chart work better? Does a pie chart work better in this instance?' "

Allow critiques of your work. And don't get bent out of shape. Remember, she said, "you're going to learn more if something fails than if it succeeds... and experience is what you get when you don't get what you want."

This all sounds like sound advice to me. Do you have any of your own beginner recommendations to share?

— Beth Schultz, Circle me on Google+ Follow me on TwitterVisit my LinkedIn pageFriend me on Facebook, Editor in Chief, AllAnalytics.com

Related posts:

Beth Schultz, Editor in Chief

Beth Schultz has more than two decades of experience as an IT writer and editor.  Most recently, she brought her expertise to bear writing thought-provoking editorial and marketing materials on a variety of technology topics for leading IT publications and industry players.  Previously, she oversaw multimedia content development, writing and editing for special feature packages at Network World. In particular, she focused on advanced IT technology and its impact on business users and in so doing became a thought leader on the revolutionary changes remaking the corporate datacenter and enterprise IT architecture. Beth has a keen ability to identify business and technology trends, developing expertise through in-depth analysis and early adopter case studies. Over the years, she has earned more than a dozen national and regional editorial excellence awards for special issues from American Business Media, American Society of Business Press Editors, Folio.net, and others.

Midmarket Companies: Bring on the Big Data

The "big" in big data is no reflection of the size of the organization embracing its potential.

Push Yourself to New Analytical Discoveries

Take inspiration from Christopher Columbus as you pursue your analytical journeys.


Visualization
  • 11/19/2013 4:42:46 PM
NO RATINGS

Visualization of info done properly can help show critical info better than any chart. Starting with good quality data and implamenting it onto a map, as you've done, makes it easier to see a problem and solution. While I saw some people commented on the color choices as an issue, I can still see the data and know that if given more room for this chart it would be easier to see difference in colors.

Re: Start at 0
  • 11/12/2013 5:43:00 PM
NO RATINGS

Jeff, when you say you make what you're doing clear in the presentation do you mean the oral presentation that comes along with delivering the visualization or in the visual presentation itself? I'm wondering, because while the former helps there is the problem that the visualization lives outside the presentation, no? 

Start at 0
  • 11/12/2013 3:24:13 PM
NO RATINGS

I'm guilty of focusing a graph in on the interesting part and not showing from 0.  But in my defense, I make it very clear in the presentation what I'm doing.

You see this kind of thing on Yahoo finance all the time and it's often misleading.  Great job Tricia!

 

Re: Fine Points
  • 11/11/2013 8:38:54 AM
NO RATINGS

@CandidoNick, to me this came off as rather amateurish -- somebody playing around with heat map capability without really thinking about the data presented and readability (and who knows, maybe that was the point -- to play around with heat maps but not for "real" use). There are way too many categories, and therefore colors, to make this easily digestible, which makes it quite ineffective. Plus, as Tricia pointed out during the lecture, the salary ranges aren't equal in all categories. Another no-no.

Re: Fine Points
  • 11/10/2013 11:15:12 PM
NO RATINGS

Absolutely. With these heat maps, especially with such subtle contrast, there was no need for so many levels in the legend. Only added to the confusion.

Re: Fine Points
  • 11/10/2013 3:39:26 PM
NO RATINGS

I agree candidonick - more contrast!

Re: Fine Points
  • 11/10/2013 3:29:05 PM
NO RATINGS

I wasn't so sure that having so many levels was necessary.  Is there a tremendous difference in 40k and 50K?  Especially when I look at the data distribution.

Re: Fine Points
  • 11/9/2013 1:06:43 AM
NO RATINGS

I understand that heat maps are meant to get more vibrant as the concentration increases, so I assume differentiated colorings would defeat some of the purpose. However, greater contrast would be a welcome prospect.

Re: Fine Points
  • 11/8/2013 2:59:08 PM
NO RATINGS

That's an interesting point, CandidoNick, about heat maps and the color gradations being too subtle. Do you think different colors and higher contrast would make heat maps easier to comprehend? 

Fine Points
  • 11/8/2013 1:36:01 PM
NO RATINGS

How info appears may be the most critical part of presenting compelling data. These points are all fine, quality ideas that deserve consideration when preparing data for the big screen.
As for Aanderud's data map, the issue lies in the color choices. They are far too similar to one another, in my mind. And Alaska is very stretched out, hahaha.

INFORMATION RESOURCES
ANALYTICS IN ACTION
CARTERTOONS
VIEW ALL +
QUICK POLL
VIEW ALL +