Data Visualisation Tips for PowerPoint charts and diagrams


Dealing with complex data in PowerPoint and presenting it in a way that is easy to understand is a challenge that we face every day here at Presented. Get your data visualisation right (or indeed your data visualization if you are an American).

Read a few of our tips to help you with data visualisation in PowerPoint.

1.    Know your message:

Before you even start thinking about how to visualise your data, you need to know what story you want to tell with the data. What do you want your audience to take away after seeing your chart?

For example: If you’re showing the annual revenue for your company for the past ten years, there could be a number of stories you are trying to tell. Your story could be that annual revenue is growing year on year, or it could be that despite annual revenue growing year on year, the % growth each year is actually decreasing.

2.    Use message headings instead of titles:

We often see charts with titles such as:

“Annual overhead costs 2010 – 2015” or “Price variation of Edam in European cities”.

These titles tell the audience next to nothing about your story. Instead of titles, choose to use descriptive headings instead. A heading should grab your audience’s attention and immediately explain your story. If you want the audience to understand and remember your message, then don’t hide it in a topic title, spell out the message for them. “Overhead costs have halved in 2015” or “London pays more than every other European city for Edam” are message headings that tell the audience exactly what they need to know, without even looking at the data.

Titles Message headings
Annual overhead costs 2010 – 2015 Overhead costs have halved in 2015
Price variation of Edam in European cities London pays more than every other European city for Edam

 

3.    Select the right chart or visual for the data:

There are pie charts, bar charts, scatter diagrams, line graphs and probably hundreds of other methods for data visualisation. How do you know which one to choose? This is where The Graphic Continuum comes in handy. This poster (widely available on the internet) breaks down which graphs, charts or visuals are the best to choose depending on what type of data visualistaiton you need. Whether your data tracks changes over time, compares categories or shows how a variable is broken down into its constituent parts, The Graphic Continuum will show you which method will display your data the most clearly. Although it doesn’t provide an exhaustive list, there are plenty of options there to get your imagination flowing!

Data visualisation - the graphic continuum

4.    Select the right data for your message:

Don’t just use the charts and data that are ready made, drill down and make sure your data is working to show the story you want. Keep on track and don’t show irrelevant data – even if it’s pretty or interesting, stick to the message you want your data visualisation to make.

e.g. Imagine you have data for your company’s annual revenue for the last five years, but the story you want to tell is that the % growth in annual revenue is decreasing each year. There is no need to show your audience the annual revenue data – simply do the calculations and show the audience the data for % growth instead.

5.    Choose your scale:

Even if the data doesn’t change, the scale that you choose for your charts can have a surprisingly big impact on the message. There are few hard and fast rules when choosing a scale for your charts, but it is important to be aware of scale and perhaps try out a few different options. The question you should ask yourself is whether this scale shows an honest representation of your data.

One rule to definitely heed is how the scale of a column chart that starts at zero can affect your data visualistaion. We subconsciously calculate the area of the chart based on a zero start value – but if the scale doesn’t start at zero then the data might be automatically misrepresented. You can make your data highlight big differences with a higher start value. Or small differences with a zero start. Be both careful and wise with this!

6.    Simplify!

If you can simplify or remove bits of the data without detracting from the integrity of your message, then do it!
Ask yourself:

  • Will including certain data add anything to help your audience understand?
  • Does it provide useful context?
  • Do I need to keep it to offer a level of detail necessary to convince the sceptics in my audience that my conclusion is accurate and reliable?

If the answer to these questions is no, then delete it! The simpler the data is to understand, the easier your audience will find it to follow what you’re saying.

7.    Use colour sparingly and purposefully:

Colour can be brilliant at providing focus and reducing confusion, but too much colour can have the opposite effect and will confuse unnecessarily. Any more than around 6 colours is too many. Remember, if you are presenting information live, you can add animation to colour sections of a graph. That way you can explain the overall data first and then use colour to highlight the key area that you want your audience to focus on.

8.    De-clutter the Chart Junk:

Check out Edward Tufte for all things to do with data visualisation. Edward Tufte coined the phrase chartjunk. He knows his onions.

What do you want your audience to focus on? Whilst axes, legends and gridlines are no doubt useful, they should not be the focal point of your data. Use colours like light grey for gridlines and axes so that they do not draw attention away from the important data. A good rule of thumb for any data labels is the closer you can have them to the data the better. Why put the legend all the way to the side of a pie chart when you could label each segment directly? The closer you have labels to the data the easier it is for people to see which label relates to which segment.

Likewise, you often won’t need both numbers on the axes and on top of your chart series. Consider losing such elements of “chart junk” to keep your data visualisation working well for you.

 

We hope these data visualisation tips will help you!