Annotations

Some things can be quickly understood...


Chance of Winning Presidency

[Click image to enlarge]

Presidential hopes:

This simple chart captures a moment in history. As more results came in on the night of the US Presidential election, the prospects of the two candidates reversed dramatically. The chart requires very little explanation.

...others take slightly longer…

[Click image to enlarge]

Brexit - who voted 'leave':

This chart is a little more complex, so an active title and brief explanation are provided. Annotations are used to bring out some of the key messages.

...and some need time to digest...

[Click image to enlarge]

Patterns of global trade:

Circular flow charts offer an efficient way of presenting very detailed information. However, such complex charts require much more guidance, explanation and annotation.

...so help your audience to interpret and understand

Ann Kemery gives a good example of how to do this. This fairly simple chart is intended to show how the results of 6 organisations have changed over a 7 year period.

However, the chart is too cluttered for the viewer to really pick out any key messages. The bold gridlines distract from the actual data and it is difficult for users to quickly interpret the chart without help.

8. The annotation layer is the most important

Help viewers understand what the project is about and how to use it. Don't just leave them to figure it out.

Understanding our audience: ONS Personas


Research by the Office for National statistics identified the needs of 5 'Personas' who use statistics:

  • ‘Inquiring Citizen’: simple, high level information.
  • ‘Information Forager’: enough detail to make decisions.
  • ‘Policy Maker’: high reliance on trusted sources.
  • ‘Technical User’: expert in data analysis but not the base data.
  • ‘Expert Analyst’: wants underlying data and knows how to use it.

The main story:
for 'Inquiring Citizens'

[Click image to enlarge]

DWP recently redesigned many of its statistical publications to address the user needs highlighted by ONS. The Fraud and Error publication front page shown above provides scope and context for the statistics, then draws out the key messages with simple charts and plain English descriptions (no jargon).


Detail: for 'Foragers' and 'Policy Makers'

[Click image to enlarge]

The document is now in landscape format with one clear subject per page. This forces the producers to keep messages focused and centred around one or two key visuals. It guides readers with visually appealing graphs, visual cues, contrasting colours and clear narration.

The report has reduced from 140 pages to fewer than 20...which means more people will actually read it!

9. Layer Information to Suit the Context

End of this part of Chapter Four

So we know what chart we want and what to add on to help the user but how can we provide powerful visual cues to reinforce messages in data

That's what we'll cover in Chapter 4.5: Colour