Get your POINT across - 5 Steps to elegant data visualization

Consider this: You have created a report with graphs and charts to explain trends to your users.  Your report doesn’t make sense – in fact, it is confounding people.  You are frustrated that they don’t understand what you are trying to say.  Where did the message break down?

Presentation communication – Tell a story with your data.

We have so much to say with our data.  We make it robust and thorough.  We cache it and freeze it and store it, and then report on it.  Our reports have the ability to be accurate and dynamic. How do we communicate effectively when we present the data to our users?

Treat your data like bits of a story: To you, the data shows a beginning, middle and (potential) end.  It has depth and can be styled to tell you what you need to know.  The skill is in leveraging the bits, nuancing and crafting them into a narrative that can be told to and understood by your users.  The graphics should grab the user and talk to them immediately; the best graphic depictions tell that whole story in one glance. 

Getting to the POINT of your report is easy if you keep this in mind:

  1. Plan the outcome ahead of time; establish the message – what story are you trying to tell?
  2. Organize the data elements in a way that will accurately answer the question at hand; this may include adding calculated fields to talk about the data before it is added to the report
  3. Information - When the report is created, can you remove all data labels, and still get an idea of the trends you are reporting on?
  4. Notation - Summarize the report: if it takes more than 3 bullets to pull out the key metrics explained in the report, it may be too complex.  Add another graph to detail into a particularly interesting point
  5. Test your work – do the graphics truthfully represent the data? Do the key metrics answer the root question? Is the display interesting enough to become a conversation point?

The way we use information can be elegant and attractive.  Even complex data can combine into simple, informative, and functional graphics and charts.  Building a beautiful report is the key to getting user buy-in, and understanding the users’ business needs ahead of time is essential to telling a story with your data.

Comments

This is an excellent set of things to keep in mind with data visualizations. I'd also like to add that we need to make sure that we're using labels that our users will understand. Particularly with data, we often use words that are not familiar to our end users, or don't mean the same thing to outside audiences as internal ones. There's also an opportunity to use labels which do some of the work for us, by carrying a deeper meaning than just one word typically does. Hipmunk, a travel search engine, does this really well with their "Agony" sort order. It's just one word but it stands in for a lot of content.

That’s absolutely true! Data labels act to frame your analysis, and must be informative in creative ways.  Also, as in Hipmunk’s case, our labels can act as marketing tools; aknowledge the problem and provide the solution all in one label.  It’s also best if you can use labels in the users’ own “language”.  It communicates better with them, and translates complex data more easily.