Five Things We Learned About DIY Data Vis


If you work in communications at a think tank or a research organisation, being able to create data visualisations in-house is a useful string to your bow. At the eighth Breakfast Club, we got together to discuss different types of data visualisations, how and where to use them, and shared tips on the best free or affordable tools available to produce them.

‘DIY Data Viz’ featured two speakers: Jeff Knezovich, a data visualisation expert who is also Digital Communications Manager for health think tank the Nuffield Trust and an editor of On Think Tanks; and Rita Beden, External Affairs Manager at the think tank Centre for Cities.

As usual, our Breakfast Clubs are held under the Chatham House rule, so we can’t say who said what, but here are the top things we learned from the session:

1. There is a big difference between data and information

To most people, raw data doesn’t really mean much at all. We need to give it meaning by translating that data into information, and good visualisations help us to do that. We learned that data vis is “a way of visually conveying information – often quantitative in nature – in an accurate, compelling format” (if you’re interested in reading more about this, see this useful article by Scott Berinato). In terms of thinking about data vis we also liked ‘The Chart Doctor’ at the Financial Times.

2. Be clear about your purpose

Broadly speaking, visualisations can be characterised in two ways. The first type is declarative, having a specific message in mind. In this case, it’s your job to look out for trends, patterns and outliers in the data you are working with in order to hone the message. Ask yourself: if you could say one thing with the data, what would it be?

What you’re about to read might sound surprising, but you should spend more time on concepts and messaging than your data visualisation itself. Remember, you don’t always need quantiative data to talk visually about an important concept. Once you’ve got your message, start to build your story frame by frame.

Tip: you can also break down frames into tweets and tweet them sequentially. The Nuffield Trust has lots of excellent examples of simple infographics.

The second type of data visualisation offers users the chance to explore information for themselves. If it’s this type, think very carefully about your user interface, and what filters and buttons users will need to explore the data. We loved this interactive map tool of crime rates in the Czech Republic, and this interactive map showing the export rates of 63 different UK cities by Centre for Cities.

Either way, make sure you provide enough relevant context and background so that your audience can fully understand it. Finally, mobile responsiveness is not always possible with more complex data vis and maps, so consider your user needs and ultimate purpose before committing to this or that platform.

3. Focus relentlessly on the message…

…and edit to make sure that message comes across – at least for those of the declarative type. Think about the order you have presented the axis labels, data points, and so on, as this affects comprehension. Always highlight the key message where it appears on the visualisation. You can also label elements directly to aid this. Edit to maximise legibility and use appropriate colours, also considering the deeper meaning of icons, images and colours so that nothing can be misconstrued. Don’t be afraid to keep things simple – if the data is interesting in the first place, a humble bar chart can make waves.

4. Be a team player

If you work in communications, get to know what your research staff are working on, which usually means bugging them until you find out what juicy data they have in their upcoming reports. It’ll help you plan ahead and figure out what kind of visualisation will work best.

And do make the most of skills elsewhere in your organisation. Your researchers should already know how to show data in a clear, unbiased way, so often it’s just a case of recreating their data vis and adding your message or heading. Don’t forget to use your branding – where possible. Then once the visualisation has been created, make sure all the right people know about it, because it’s the kind of thing that will add a bit of jazz to your colleague’s presentations and blogs.

5. There are loads of great resources out there.

You can achieve a pretty high level of data visualisation using free tools although we found that most are limited in terms of how much you can brand them using your organisation’s colours and fonts. One suggestion was to train one staff member to use Illustrator if you’re very keen to avoid brandalism – the heinous crime of deviating from your brand guidelines!

If you want to create visualisations for a large amount of data, and build beautiful graphical analysis and dashboards, get stuck into Tableau. There is a version that is free to use but take heed – it automatically makes your data public. If you’re looking for infographic design, check out Canva and Piktochart. For maps and visual discovery, give Carto a shot. It’s user friendly and supplies great visuals.

For everyday data visualisations – try Highcharts, good old Microsoft Excel and Google sheets. These are all good for getting the basics right. You could also look at amCharts and Datawrapper. If you’re curious about more tools and how to use them, see the resources section of the On Think Tanks Data Visualisation Competition. Good luck!

Tagged with:
Posted in breakfast club

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: