Choosing the Right Chart: Data Visualisation for Business Presentations

Analytics dashboard with line and bar charts for business data visualisation

When a chart fails in a presentation, people tend to blame the styling: the colours are off, the fonts are too small, it looks dated. Sometimes that is true. Far more often, the chart is simply the wrong type for what it is trying to say.

A chart is an argument in visual form. Pick the format before you know the argument, and no amount of polish will rescue it.

Start with the sentence, not the data

Before you open the chart tool, write down the one sentence you want the audience to take away. “Revenue has grown every quarter.” “Enterprise customers are now our biggest segment.” “Our costs are lower than the three main competitors.”

That sentence tells you what kind of comparison you are making, and the comparison tells you the chart.

Match the message to the format

Change over time → line or column chart

Use a line when the trend matters more than individual values, especially across many periods. Use columns when you have fewer periods and each value matters on its own, such as quarterly revenue.

Comparing items → horizontal bar chart

Bars are the most readable chart there is. Horizontal bars leave room for long labels and sort naturally from largest to smallest. If you are comparing competitors, regions or products, this is almost always the answer.

Parts of a whole → stacked bar, or occasionally a pie

Pie charts are fine with two or three segments where one clearly dominates. Beyond that, people cannot compare slice sizes accurately. A single stacked bar or a simple ranked bar chart does the job better.

Relationship between two variables → scatter plot

Useful for showing that two things move together, such as marketing spend and pipeline. Rare in pitch decks, valuable in analytical and board presentations.

How a total builds up or breaks down → waterfall

Ideal for explaining how you got from last year’s profit to this year’s, or where a budget goes. Finance audiences read waterfalls instantly.

One number that matters → just the number

Not every figure needs a chart. “3x year-on-year growth” set large on a clean slide often lands harder than any graph.

Five fixes that improve almost any chart

  1. Put the conclusion in the headline. Replace “Revenue by Quarter” with “Revenue has grown for six straight quarters.” The reader knows what to look for before they look.
  2. Highlight one thing. Grey out everything except the bar, line or segment that proves your point, and give that one element your accent colour.
  3. Remove the clutter. Gridlines, borders, 3D effects, shadows and legends you could replace with direct labels. Each one taxes attention and adds nothing.
  4. Label directly. Put values and series names on or next to the data rather than in a legend the eye has to travel to.
  5. Round sensibly. $4.2M is easier to hold in your head than $4,187,352. Precision belongs in the appendix.

Common mistakes worth avoiding

  • Truncated axes that exaggerate small changes. Sophisticated audiences notice, and trust drops for the rest of the deck.
  • Dual-axis charts that force viewers to decode two scales at once. Two simple charts side by side are nearly always clearer.
  • Too many series. More than four or five lines on one chart becomes spaghetti. Split it, or show only the lines that matter.
  • Inconsistent colours across slides. If “Product A” is blue on slide 6, it should be blue on slide 14.

Presented versus read

A chart you present can be simpler, because you are there to explain it. A chart in a deck that gets emailed needs a clearer headline and perhaps a short annotation, because nobody is standing beside it. Know which one you are building. We covered the wider version of this problem in Why Your Slides Have Too Much Text.

Good data visualisation is not decoration. It is the fastest route from a number to a decision. Get the format right and the design has something worth emphasising.

Have a deck full of data that is not landing? Data visualisation is a big part of what I do.

Leave a Comment

Your email address will not be published. Required fields are marked *