8 Tips for Doing Data Visualization Right


Data visualizations come in all shapes and sizes, and range from the simple to the complex. Some are destined for the public web, others for internal eyes only. Some require sophisticated visual analytics software to produce, and others… Excel. In any case, some will work and others won't.

Some, in fact, will be so successful they'll go viral. Sadly, some will be so bad they'll go viral as well. You don't want this latter situation to happen to you -- unless you think it's OK to turn your company into a public laughingstock.

The fact is, you may not ever have had to think about presenting your data visually before, but you do now. As businesses turn to data-driven decision making, you carry the onus for delivering insight and intelligence in a way the business can absorb easily -- meaning, through data visualizations.

A visual message, done well, delivers far more punch than text -- or numbers -- alone. Study after study shows this, and experts scream it from on high. Visualizations give people something around which they can wrap their brains.

The good ones compare multiple values and put information into context, said Randy Krum, president of InfoNewt and Cool Infographics blogger, during his "Effective Infographics" session at yesterday's INFORMS Analytics Conference in San Antonio. The bad ones misrepresent, misinform, or otherwise totally miss the mark.

Krum, who makes a living at "helping companies see their data," shared eight design tips for making sure your data visualizations -- self-created or outsourced -- don't implode on you. Click on the image below to start a slideshow with his tips, the first being accuracy.

Tip 1: Accuracy
'If you mess up your data visualization, you lose your credibility' -- and yet data visualizations are littered with mistakes, Krum said. For example, using circle size to show that one category is three times the size of another is correctly done by multiplying the area times three. Yet designers often enlarge the circle by tripling the diameter, and thus misrepresenting the data.

"If you mess up your data visualization, you lose your credibility" -- and yet data visualizations are littered with mistakes, Krum said. For example, using circle size to show that one category is three times the size of another is correctly done by multiplying the area times three. Yet designers often enlarge the circle by tripling the diameter, and thus misrepresenting the data.

Do you have advice on how to make data visualizations shine? Share below.

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.


Re: Legends are evil, so are bad bar charts
  • 4/24/2014 3:10:10 AM
NO RATINGS

@MTraylen--The correct answer to your questions may be more a matter of cognition, psychology, etc.  If you have a cognitive bias to assign too much value to small percentage variations, then a non-zero baseline will play to your bias.  In the car example, if I'm a marketer and I'm aware that people will ignore the numbers and instead compare the height of bars in an absolute sense, then if I can offer $5 lower on a $25,000 purchase, and I set my baseline at $24,990 and max at $25,010 then it makes the apparent difference seem much, much larger.  This sort of thing is stock and trade for marketeers.  Full disclosure: I have had the word Marketing and Markets in my title on more than one occasion.

Re: Tips for Doing Data Visualization Right
  • 4/23/2014 11:09:25 PM
NO RATINGS

..

I especially like the pie chart example where the pieces of the pie add up to 193%.

Re: Legends are evil, so are bad bar charts
  • 4/23/2014 3:22:07 PM
NO RATINGS

Replying to eafpres about baseline at 0, I normally agree with you, but in the case of credit scores isn't the baseline somewhere way above 0?  Aren't credit scores in fact, over-inflated and therefore this is an exception to this rule?  Another I would argue would be car prices; why would you need ot show the baseline at $0 ever for car prices?  The baeline needs to be relevant, consistent and comparable I would argue.

Re: Data for Educators
  • 3/18/2014 7:35:30 AM
NO RATINGS

shit.........

Re: Tips for Doing Data Visualization Right
  • 4/21/2013 9:48:17 PM
NO RATINGS

@shehan. I haven't tried it myself, but have heard good things about it from a user I interviewed at a food company quite some time ago: IPad Dashboard Spices Up Food Maker's BI Strategy

Re: Tips for Doing Data Visualization Right
  • 4/20/2013 9:39:26 AM
NO RATINGS

I really like the way smartphone data is show on the third slide, this is one of the best ways to share information.

Re: Tips for Doing Data Visualization Right
  • 4/20/2013 9:36:09 AM
NO RATINGS

@Beth – It's always nice to see visual data on all colors in presentations and reports than the traditional data legends. Have you had a chance to try Qlikview, I think it's one of the best when it comes to representing information graphically.

Data for Educators
  • 4/11/2013 12:37:09 PM
NO RATINGS

Thanks - I really enjoyed this! Though there are differences (e.g., veering from the norms of other displays), there is much crossover between data visualization tips and the best practices for data being presented to educators (avoid legends, maintain credibility, etc. - www.overthecounterdata.com/otcd ). In both areas, tools like your tips that get people talking about the importance of design are so important. The more people realize the importance of design, the better for everyone seeking to understand the data being presented. Thanks again! :-)

Re: Legends are evil, so are bad bar charts
  • 4/11/2013 7:39:40 AM
NO RATINGS

Eafpres Any other suggestions for good visualizations or charts?

Re: Legends are evil, so are bad bar charts
  • 4/10/2013 11:58:57 AM
NO RATINGS

I see your point eafpres -- the variation would seem less significant if a zero baseline was used

Page 1 / 2   >   >>
INFORMATION RESOURCES
ANALYTICS IN ACTION
CARTERTOONS
VIEW ALL +
QUICK POLL
VIEW ALL +