6
Mar 13

## Before you hit the ‘chart’ button…

By and large most charts originate from a set of absolute figures, that is to say the actual numbers of people, or \$ of GDP, or barrels of oil.

However I’m suggesting you should pause before you hit the ‘chart’ button having highlighted these figures in your spreadsheet. Here’s a small – but useful I hope – section from a training course I ran recently, illustrated with examples dug up from my archive.

Consider if one of these three options might better illustrate your point:
% share Show figures as a share of the whole
Ranking Rank figures largest to smallest
Relative Show the figures relative to something eg a specific year, the value for a country of interest, the average

And you don’t need to be a data scientist, know a programming language or have a statistics degree. It’s all easy to achieve with a formula in the next column over in your spreadsheet. (Good tips and tutorials in Excel from Chandoo and GoogleSpreadsheets from School of Data).

Any of these options can lead to more illuminating ways of displaying a set of figures, though it goes without saying that it depends on the figures and your goal.

% share

North v South spending, Regeneration & Renewal magazine, August 2006

It can be useful to show absolute and % share alongside one another

Renewable energy need, The ENDS Report, April 2008

Rank

1981: The way we were, BBC News Online, November 2010

Relative to a specific year

Degree results, BBC News Online, July 2010

Relative to a specific country

How other countries compare to India, Nesta report on research and innovation, July 2012

Relative to a set number

Aging population, Human Resource magazine, September 2007

Relative to the average

What is growing in Britain’s gardens? BBC News Online, May 2010

1
Jan 13

## The best bits of 2012

As 2013 starts I thought I’d pull together some of the best things that I’ve said, seen, heard and learnt in 2012.

Starting with a definition of infographics I still stand by my first tweet back in March that stemmed from an exercise I created for some students: The INFO gives the GRAPHIC its form. Otherwise the ‘graphic’ is most likely acting as an illustration.

In the same vein – with the goal of creating better infographics – I’ve been encouraging anyone designing them to make sure they can answer yes to two key questions: does the graphic tell a story? and is that the same story that you set out to tell?

What has emerged strongly this year are increasingly sophisticated interactive page designs and delivery (examples under my tower graphic comment further on). However the stark simplicity but instant clarity of a recent collection of animated gifs brilliantly demonstrates the power of visuals to explain things, another fundamental that underpins great graphics regardless of layers of fancy interaction or not. Yes, some are a little clunky, but they’re really effective.

A couple of other clunky-but-effective examples that crossed my radar this year show an application of interaction that I think is underused: interaction with the goal of aiding understanding. Two examples of this ‘learning by having a go’ are here, by SolutionRealm and Tom MacWright. Click on the images to have a go.

As for formats, tower graphics seem to be thankfully on the decline. Either injecting subject matter that suits the format or playing to the storytelling strengths of interactivity a new, much more engaging breed of tower graphic is emerging. Two good examples I’ve seen are featured below, from the BBC and Angela Morelli. Click on them to go to the originals.

I still love motion graphics when they’re done well. This one about the Titanic below – another BBC production – is a good example:

And if you just want one piece of guidance as to the key to motion graphics’ success I’d offer up the BBC’s Jonathan Spencer and Mark Edwards comment at the Information Design Association: The time you have to view a graphic is the key constraint for TV data viz.

Data journalism is all around us (you can find links and resources I recommend here). By way of an example, this was a simple piece of data journalism I did this year during the Olympics – click images to see full graphics – that in contrast to endless medal tables threw up some interesting and original observations, two good reasons to keep encouraging and helping journalists and designers to engage more with data.

A theme that’s close to my heart that came out of the Information Design Association’s conference was design with a cause. Later in the year this was then well illustrated by Spinifex working with the Australian Bureau of Statistics to project interactive data onto buildings to promote the release of Census data. (A continuation of one of my all time favourite pieces of work).

Some good reading will be found on two blogs that have emerged in 2012 about the design decisions behind some of the best work we’ve seen – NYT and NatGeo – and a third, The Why Axis, promises to reveal more of the same too.

After another year is done I still think the New York Times are consistently the best when measured against my criteria: clear, relevant, engaging, useful, original, elegant. And you can enjoy a review of 2012 as illustrated by their graphics.

Happy New Year, see you in 2013!
:)

2
Oct 12

## ‘Say what you see’

I recently found myself inviting a workshop group to ‘say what you see’. And then I had a flashback… 1980s… Roy Walker… Catchphrase… anyone?

This came about from a litmus test I have for successful visual journalism: can you still piece together the story from the visual if you remove the words? Running this as an exercise – displaying examples with all the words blurred out, getting people to ‘say what you see’, and then revealing the words – has been fun.

Try it:

Visual with words blurred out – say what you see.

And with the words revealed. It passes. They say the same thing.

And this one, with words blurred:

Words revealed below. It nearly works, you will have recognised the process depicted, but it turns out the headline is about costs. But these haven’t been visualised at all:

It’s beautiful, and it’s not trying to be heavy journalism, but the point is still there.

And this (blurred left, revealed right):

When the pieces of info to be presented together are disparate the visual can only introduce the theme, but add nothing to the telling of any story. The visuals in this case are illustration, not visual journalism.

For my money, a successful piece of visual journalism is when the visual functions as a piece of journalism itself, telling the same story as the headline it sits under.

22
Aug 12

## What are you trying to say?

Infographics improve tenfold – or just remain as charts – once people get the hang of the idea that it’s better to commission a graphic when you have a story in mind, something specific you want to say, as opposed to the commission ‘I want an infographic’. The Cabinet Office’s recent chart (featured at the end of this post), and now animation, I fear is a victim of the latter.

Let me try to explain.

If you’re saying “Here is a detailed breakdown of the £5.5bn savings” it would be best presented as a bar chart.

If you’re saying “These are the five areas where we made savings”, group together similar items and present the simplified version as shares of a whole (I’m trying to get close to the original here).

If you’re saying “Over half the savings came from looking at staffing” present it as a pie chart.

If you want to say all three, and more, by all means do. Link them together, use an animation with a voice-over to guide your audience through.

But if all someone says is “I want an infographic” you’ll be presented with the likes of this.

How easily can you find out (questions I imagine it would be useful for this graphic to inform):
What were the biggest and smallest savings?
How much of the savings come from any one area eg staffing?

Commission a story, not a format. You’ll get better results, I promise.

17
Mar 12

## Tower graphics

I get at least two enquiries a month to create these* so it’s time to put down my thoughts about them.

In a nutshell remember tower graphics are a piece of journalism as much as they are a piece of design, and with that comes a set of standards.

From TechCrunch

Infographics v editorial design & illustration
If we’re talking semantics, tower graphics are not infographics, rather editorial design, heavy on illustration.

If we’re talking plain english, what infografistas do better than editorial designers and illustrators is to grapple with information. It is this which leads them to a graphic, or several, that best show(s) visually what the information is saying.

What editorial designers and illustrators do better than infografistas is to create designs that can hold together all the elements of a whole page or screen, some, all or none of which can be an infographic.

The two fields are not mutually exclusive, tower graphics would benefit from both, but on the whole they fall between the two, erring on the side of style with little regard to substance.

An example from the other end of the spectrum that showcases what can result from bringing together people from both areas of expertise would be Eureka and Bloomberg Business Week. These are my current favourite examples of editorial design that seamlessly integrate journalism, infographics and illustration. I can stop worrying about semantics and just enjoy a good read.

Pros and cons

Tower graphics are good for going viral. They’re friendly and engaging. Their style is ‘of the moment’. Everyone wants one.

Tower graphics are bad for their treatment of information. See the one from earlier this week on Pinterest as a case in point. For starters the three bar charts centrally are all incorrect which leads you on to question the integrity of the whole thing. And with that many sources cited – listed at the bottom – are the figures they compare really comparable?

A form of journalism
I don’t mean to single out the above Pinterest tower graphic, and I don’t need to. Those points apply to the majority of tower graphics I’ve ever seen. But (I get the impression) tower graphics are by and large generated as a piece of visual collateral, by designers with no analysts or journos involved. Hence the ‘con’ above: the quality of journalism is poor.

Scrolling infographics
A tower graphic is distinct from a deep infographic that plays to the scrolling nature of a browser to tell it’s story better. A nice recent example being the BBC’s Ocean trench.

Tower graphics: Let’s make them better
*I get at least two enquiries a month to create these, and being of a purer infographic persuasion they are not really my cup of tea. If there is a talented illustrator or editorial designer out there who’d like this kind of work please get in touch. If you’re good with numbers so much the better. If not, I’m happy to collaborate on that bit.

23
Feb 12

## The use of colour

And Mr Brinton has good practical advice too, that’s as relevant today as it was back then.

23
Feb 12

## Magic in graphs

There’s a lot of blog-debate over what’s good and what’s not in the fields of data visualisation, data journalism, storytelling, infographics at the moment. Given it’s a growing field this isn’t hugely surprising, the old hands having their experienced feathers ruffled by faster, cheaper, easier, shallower, stylised mass production for mass consumption. I could have re-posted several treatises on this recently, but if you’re interested you’ll have found and read them anyway.

The net effect of this on me since Christmas is that it’s getting me down: it makes for too much heavy reading at the expense of showcasing good work. I haven’t come across anything I’ve really wanted to post about. Until, that is, …

The commentary under each example doesn't hold back be it positive - as in this example - or not.

…Willard C. Brinton’s two primers on the ground rules.

They’ve been around for a long time, but were recently posted on Chart Porn. These gloriously thorough books Graphic Methods for Presenting Facts (1914) and Graphic Presentation (1939) remind me of the underlying reasons I do what I do which is summed up by the 1939 preface’s title Magic in graphs.

The first principles haven’t changed. I’d have liked to meet Willard Cope Brinton.

And this all puts me in mind of someone who’s work I was browsing recently, an example of a modern day practitioner who manages to cut the crap and just get on with doing lovely work: Nicolas Rapp.