Data visualisation

Data visualisation is turning a result into something a decision-maker can act on. It is the part of analysis that decides whether any of the work before it changes anything at all, which is why analysts who can do it get asked what they think and analysts who cannot get asked for a spreadsheet.

What the work consists of

Choosing a form that matches the comparison you need the reader to make, removing everything that does not serve it, and writing the finding above the chart so nobody has to infer it. Most of the craft is subtraction.

What employers screen for

Whether the chart resolves a question. A portfolio of handsome charts that settle nothing reads worse than one plain chart with the conclusion stated above it, because the second proves you understood what the chart was for.

• Matching form to comparison: change over time, composition, distribution and ranking each have a right answer
• Honest axes — starting a bar chart anywhere but zero is a decision you must be able to defend
• Labelling with the source and the period, which is also this publication’s house rule
• Knowing when a table is better, which is more often than designers care to admit

The honest limit

One to two months to competence, and it cannot carry a career by itself. It multiplies the value of analysis rather than substituting for it: a beautifully presented wrong answer travels further than an ugly right one, which is a reason for care rather than for confidence.