Data Visualisation: A Handbook for Data Driven Design

by Andy Kirk

You might extend the wording of the data visualisation definition to describe what the facilitated understanding will be about, as illustrated in Figure 3.2.

There is no singular function of visualisation; what’s important shifts with the constraints of your audience, goals, tools, expertise, and data and time available.’

When commencing a project, probably not all of the circumstances that may potentially influence your work will be definable. Things change. That is why we need to be prepared to accommodate elegantly the impact of new factors at any point in the process.

Other stakeholders might exist as subject-matter experts, available to offer advice about domain-specific queries.

Create categories of stakeholders: - owner - user - subject matter expert

Identifying this cast and crew of people who have a stake in your work will help you anticipate the interactions and relationships you might need to manage: what personalities exist, what help you might exploit, and what obstacles might need navigating. Diplomacy will be required.

What is your audience’s relationship with the subject matter?

What is their motivation for acquiring the understanding you intend to provide for them?

What are their visualisation literacy capabilities?

If you have an especially wide-ranging and diverse profile of audience characteristics, you are unlikely to be able to satisfy the needs of each variation; you might need to commit to prioritising some audiences over others.

Across these capability groups, which attributes do you possess, or can you demonstrate? Where are your weaknesses, in terms of both gaps and potentially overly dominant traits?

Summarise for yourself.

The best functioning visualisation teams will offer a balanced blend of skills across all these hats.

Even if you do not need necessarily to adhere to a deadline, let’s say for personal projects, it can still be useful to define a target date to help sharpen your progress.

I would always recommend noting down the duration of each major task across your design process, so you can more surgically evaluate how you have spent your time and be better placed to estimate accurately commitments on future projects.

You may receive guidance from your stakeholders that certain messages need to be downplayed or amplified.

These can be difficult matters to handle: you want to respect any requirements received, but also you do not want to undermine the integrity of what you are representing.

Creative influences can emerge internally, through the unique dynamics of an organisation, and externally, through broader competition across the entire marketplace and industries.

The technology you have access to will affect how digitally ambitious your work can be and/or how efficiently you will be able to make it.

This concerns the characteristics of the setting in which your work is going to be encountered and consumed by your audience.

Firstly, is it going to be consumed remotely – away from you – or presented in person?

Secondly, is the nature of the engagement one that needs to facilitate especially rapid understanding, or does it lend itself to a more extended, prolonged engagement?

I usually think of four broadly typical settings:

Categorise your work according to these.

You will need a clear understanding about the specific format of the deliverables required.

As well as the medium, it is also important to get a sense of the project’s deliverables in terms of expected quantities.

The issue of frequency concerns how often a particular project will need to be reproduced and what its lifespan will be.

Can you justify investing time, for example, in programmatically automating certain parts of the construction process if they can be reused to save time in the future?

The definition of vision is ‘the ability to think about or plan the future with imagination or wisdom’.

As before with the origin curiosity, purpose might evolve as you progress through the process, but the sooner you can establish at least some degree of focus, the better, especially in being able to eliminate potential creative avenues that will have no relevance to your aims.

I explained how, as visualisers, we have limited control over the final phase, comprehending, which is largely determined by a viewer’s attitude and connection to the subject matter.

We do, though, have control over how our viewers perceive and interpret our visualisations. They are particularly influenced by the choices surrounding two significant design characteristics: tone and experience.

The tone conveyed by a visualisation has an influence on the perceiving phase of understanding.

A visualisation that conveys a reading tone places emphasis on optimising the precision and efficiency of perceiving the represented data.

Indeed, you might reasonably ask why would you ever not seek to optimise the accuracy and efficiency of value judgements?

Well, this is why the definition of your purpose is so significant.

In contrast to reading values, sometimes we might justifiably place more emphasis on feeling data.

The viewer might gain a general sense of major patterns that reveal things going up and going down, the major clusters of connections and the major components of a whole.

The red houses represent the small number of families who have contributed nearly half of the initial campaign funding, the green pieces are representative of the total households in the USA.

Data physicaliation?

Data is more than just a bunch of numbers and text values. Thinking about tone is to recognise semantically what your subject is about: what activity, instance or phenomenon does it represent?

Learning about the underlying phenomena of your data helps you feel its spirit more clearly than just looking at values in isolation.

Encapsulating emotional sensations like fear, disgust, fun and humanity through your design choices might accelerate the meaning of the subject and potentially affect the most elusive phase of understanding, comprehending and how viewers feel.

Secondly, Kirk Goldsberry stresses that data visualization should ultimately be true to a phenomenon, rather than a technique or the format of data.

The traits of a good explanatory visualisation are that it effectively does the job of communicating the main features you would remark on if you were there.

Some might describe this as ‘narrative’ visualisation. This is where the most tangible demonstration of storytelling is relevant.

Common to any explanatory visualisations is a need for the visualiser to possess sufficient knowledge – or have the skill and capacity to acquire sufficient knowledge – about the topic being displayed.

Explanatory projects will mainly be for audiences who do not have the knowledge, capacity or time to form for themselves the meaning of a visualisation.

The responsibility for then translating ‘what it means’, the essence of interpretation, is passed to them. This kind of experience will only be suitable if the audience have the requisite knowledge and motivation to form such interpretations themselves. Indeed, the assumption would be that the users will be better equipped to do this than the creators.

A deeper exploratory experience goes beyond just offering means to interact and more towards what might be described as offering a participatory or contributory experience.

With exhibitory visualisations the viewers have to do the work to interpret meaning, relying on their own capacity to perceive and translate the features of a visualisation.

it represents the idea of a visualiser leaving the viewer to finish the task of gaining understanding.

In contrast to exploratory visualisation, for exhibitory pieces this is conducted just by looking and thinking. But like exploratory experiences, exhibitory projects rely entirely on the audience having the motivation and capacity to interpret.

Another common context for using an exhibitory visualisation might exist in the situation of producing a visual for stakeholders who have directly requested you to create something for them.