Data Analytics

Getting better at communicating with data

21 July 2020 by Heidi Marshall

We’ve all been there… sitting through a presentation with endless charts, full of extraneous information, feeling no wiser by the end of it (and possibly a little less well informed than at the start).

Some of us are lucky enough to have experienced the contrary. The presenter who understands the data so well and the context of that data, and has translated it in a way that makes you understand the issue better. But most importantly of all, has delivered the message in a way that makes you care and elicits a response from you. 

As a participant your take-away is at opposite ends of the spectrum – from feeling confused and disillusioned, to feeling like you could make a difference with your newfound knowledge and insights.  

The difference lies in the presenter and their ability to communicate using data. The underlying principles of communicating with data are exactly the same as with any other form of communication – it is the presentation of quantitative information in an easily digestible manner. 

In 1949, two employees at Bell Laboratories – Claude Elwood Shannon and Warren Weaver published a seminal article called ‘The Mathematical Theory of Communication’. In it they identified three types of communication problems:  

  1. The technical problem – how accurately can the symbols of communication be transmitted.
  2. The semantic problem – how precisely do the transmitted symbols convey the desired meaning 
  3. The effectiveness problem – how effectively does the received meaning affect conduct in the desired way. 

The effectiveness problem is probably the most important of these three – the ‘so what’? If data is perfectly created, captured, transmitted, and understood but the recipient doesn’t care at the end of it, then ultimately the communication failed. So, how can we communicate data in a manner that is not only understood, but that fulfils the original goal? 

There are six principles of Communicating Data to adhere to:

1 Know your goal

Who is your target audience? What do you want them to know? What action do you want them to take?

2 Use the right data

Sometimes a whole presentation can be focused around a single number, other times you need to use enough data to ensure your findings are meaningful and statistically significant. However, always be conscious of overwhelming people – just because you’ve done a huge amount of analysis doesn’t mean you have to share the finite detail. This serves to confuse.  

3 Select suitable visualisations

There seems like an infinite number of options for visualising data, but there are frameworks such as Jock Mackinlay’s ‘Quantitative, Ordinal and Nominal’, that recommend the best visual approach depending on your data set and type. The main objective to keep front of mind is ‘does this visualisation get across the point I am trying to make’. 

4 Design for aesthetics

Willard Cope Brinton referred to the ‘judicious embellishment of charts.’. The trap we commonly fall into is we beautify data visualisations to the extent that we distract, confuse or worse distort the data. However, it is possible to elegantly design and around interest / enhance memory. Stay away from multiple fonts, clashing colour scheme, sloppy alignment, vertical or angled labels, thick grid lines or borders and useless images and clip art.  

5 Choose an effective medium and channel

Take care when selecting the ‘how, when and where’, so that you can reach the right audience and achieve your goal. Ask yourself whether you need graphics or narration- or both? Will you present live or pre-recorded or stand-alone? Generally, the more complex the data set, the more likely you will need to combine media types and present in person or remotely. There is also the trade-off between cost and impact to consider. Your goal will inform these decisions. 

6 Check the results

Make sure you achieved your goal. Did your audience receive the message? Did they understand it? Did it have the right impact on them?  

When Hans Rosling took to the stage in 2006 to present his GapMinder scatterplot he redefined how to present and communicate data. The data was complex, the communication was complex, but it was brilliantly executed and left a deep, long-term impact. 

Communication of data is fundamental to Data Analytics. In the words of Willard Cope Brinton: “As the cathedral is to its foundation so is an effective presentation of facts to the data.” 

Share this with your network

Recent posts

You might also be interested in...

Stay connected with Corndel
Keep up-to-date with the latest information from the team, including all of our
must-read news and insights.