Data Analysts and the Art of Communication

by   |   April 10, 2015 5:30 am   |   0 Comments

Bill Shander, CEO and founder, Beehive Media

Bill Shander, CEO and founder, Beehive Media

If a tree falls in the forest and no one is there to hear
it …

Does your data live in a vacuum or does it need to find an audience? It’s obvious. Your data makes noise – it must be heard!

How often do you think your audience receives your data and it resonates clearly and immediately, telling the story you hoped it would tell to them? How often do you even consider this as part of what you do in your work?

Data analysts have to start to think about not only what their data is revealing, but also how they present that data to their audiences. The first step in this process is to realize that, even if your job description may not include this term, if you are a data analyst, you are also a communicator. And to communicate data, you must visualize it.

We have all heard the arguments before, but let me reiterate the most basic one, which is that with rare exceptions, a visual display of data always will be more effective at communicating overall trends and patterns than the raw data itself. Hopefully I don’t need to spend any more time on this point in this article – that’s an entire article (or book!) in itself. But it seems to be a largely accepted axiom at this point, not requiring much debate.

Assuming you accept this, your next argument might be. “I already visualize my data! I kick out great Excel or Stata or SPSS charts as it is! You don’t need to improve on that!” Well, I would argue that even though you are already visualizing your data, odds are that you are not doing it as well as you should be. But you are not alone. Most people who visualize data aren’t really doing everything they should be doing.

So how to improve? Here are four key tips to keep in mind when communicating your data to anyone.

Related Stories

7 Key Ingredients for Knock-out Data Visualizations.
Read the story »

How to Use Visualizations to Support a Data Story.
Read the story »

What’s Wrong with this Picture? The Art of Honest Visualizations.
Read the story »

How to Tell a Story with Millions of Rows of Data.
Read the story »

Think Like Your Audience

Have you ever tried explaining anything to a four-year-old? What do you do? Let’s say you are trying to explain how to drive a car. You wouldn’t start with the rules of the road or legal issues or even with a tutorial sitting in the car. You’d probably instead start talking about how it’s like their favorite iPad driving game but that instead of tilting the iPad to turn, they rotate the wheel. You immediately begin to think about what they already know, so you can use the right similes and metaphors. And you’ll think about what they can’t possibly know (or understand) so you don’t waste any time or confuse them (for example, you wouldn’t explain the mechanics of a transmission). In other words, you channel your inner four-year-old – you think like your audience!

When you are communicating data, you need to do the same thing. You have to think about whether you are communicating to an audience that has deep expertise in the subject or is new to the topic. Do they have great familiarity with the data itself, or is this entirely new? Can you compare a lot of things at once or do you need to work through one idea at a time? The more you think like them, the more you can communicate in a way that will resonate with your audience. You have to resist the temptation to believe that you already do this – take the time to really consider your audience as you think about how to communicate with them. Don’t do it in your rote, usual way.

Be a Communicator

Once they have channeled their audience, communicators craft a message. They realize that they aren’t simply regurgitating facts and data. They think carefully about what they are trying to communicate and what their audience is trying to understand. And they make sure what they produce aligns with what the audience expects and needs.

You might argue, “But I am not telling a story,  I am just sharing facts.” And I might agree – sort of. A collection of facts is useless without some structure. You are not throwing numbers over a fence. You are providing insight and knowledge. What about? That’s your story. Maybe you are sharing quarterly sales figures, so your collection of facts is within that context. Your story is about sales performance and you need to organize and display your data to tell that story, not the story of employee engagement or lead generation or a hundred other collections of facts you might tell in another report.

KWYRWTS

To properly align what you are communicating with your audience, this is the most important tool in any communicator’s tool belt: Know What You Really Want to Say. This is absolutely critical. How can you communicate anything when you don’t know what you are trying to say?

For data analysts, this is often the hardest part. A few key challenges around this:

    • Your boss asked for the quarterly sales figures, so you prepared the quarterly sales figures. But is that the data she actually needs? As an analyst, you have to understand the goal – if the question is around profitability, perhaps the profits are a better metric. And rather than presenting only quarterly figures, you can provide quarterly numbers but with weekly snapshots because you know that there was that one outlier week that affected the overall average.

 

    • You want to include as much data as possible because you know you like to work with more, rather than less, data to reach your conclusions. This goes back to the first point above: Your audience probably doesn’t need more data, it just needs the right data. And if the audience does need more, you should provide it in chunks, not all at once. Tell a story with data, don’t dump it all on your audience in one image.

 

  • Your company and the competition have always done it this way. As an analyst, you are used to looking at the data in a certain way – and you expect your audience is too. But you should sometimes resist this assumption and try new things. There is risk in introducing new ways to communicate data, but sometimes it is outweighed by the opportunities that it presents.

 

Know Which Design Principles to Follow, and Which Not to Try

Good design is hard. Trust me, I am a self-taught designer and it has taken me 20 years to get to the point where sometimes I design something pretty nice. But it does not come naturally or easily. I follow a few basic principles and I also know which design practices I shouldn’t even bother attempting on my own. Here are a few basics:

    • The eye likes things to line up. So when you are placing graphics and text on your charts and slides, align the edges of things to each other. Whether the title is left-aligned with the chart or right-aligned with the right edge of the last bar or some other alignment, just align things to each other. You have some flexibility here, but know that the more you take the time to line things up, the better they will be perceived.

 

    • Don’t be afraid of white space. Let me rephrase that. EMBRACE WHITE SPACE! You don’t want to fill every pixel of available space with content. Create a chart, leave some space around it. Don’t label everything. Make the axes and the tick marks on the axes a nice light shade of grey, use as few ticks and gridlines as possible. Draw the eye to the important stuff, not the axis labels and other junk! And leave empty space for the eye to rest, to help it refocus on what actually counts.

 

  • Don’t try to come up with your own colors, but don’t accept the default colors Excel is giving you either. Color selection is really hard. Most good designers aren’t that great at choosing colors. Never mind that most people don’t have experience picking colors for optimizing data display. But you can Google “optimized color palettes for data” or something similar and you will find colors that work well. If you have to adhere to brand standards, then get your branding or design team to work with you to find the best colors for your data.

 

Designing a great data visualization that will win you awards may not be at the top of your list – and it doesn’t have to be. What should be is designing a visualization that reaches your audience and communicates what you intended it to communicate. This post and the few tips I’ve provided are a pretty good start.

If you want to learn more, you can check out my newest course on Lynda.com, “Data Visualization for Data Analysts,” where I talk about these and other key principles for thinking like a communicator while sharing your data. If you don’t have a subscription to Lynda.com, you can use this link to take a look at the course and get one week of free access to Lynda’s entire library of thousands of courses.

Bill Shander is CEO and founder of Beehive Media, an information design and data visualization agency. Beehive helps its clients convert abstract concepts into tangible and understandable experiences via smart information design and data visualization.


Subscribe to Data Informed
for the latest information and news on big data and analytics for the enterprise.



Improving access to data across your company/partner ecosystem



Tags: , , ,

Post a Comment

Your email is never published nor shared. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>