8 visualisation tips for displaying data that people will actually understand
One of the most powerful ways to persuade people to your opinion is by providing them with indisputable facts. The problem is, if you give your audience three pages of tabulated datasets, they are highly unlikely to read them. This is one of the reasons we choose to present data in the form of charts and visualisations rather than tables of figures.
It’s simple, right? We just open PowerPoint, click the Chart Wizard button and voila, we are statistical masters.
Not so fast!
Creating visualisations is not quite so simple. In fact, it can be rather difficult at times to visualise data effectively. I’ve experimented considerably with different ways to display information to business stakeholders and this week I share with you 8 of my best tips for creating effective visualisations that educate, persuade, inspire and may even entertain.
1. Include a caption or description
No matter how hard you might try to make a visualisation as readable as possible, some people will always just prefer prose. Space permitting, it’s generally wise to provide, at the very least, a caption to accompany the visualisation and to explain its message. If possible, I would recommend including one or two paragraphs describing the data and talking about why you’re communicating this information to the reader (e.g. to inspire an action of some sort).
2. Combine multiple data points into one chart
Where you have multiple charts with one series of data in each, you may be able to combine some of them into a single graph. Combining the data into one chart will allow you to compare them and it may also save space and simplify your overall presentation. Take heed: only consider combining charts where they are reasonably simple and are without a large amount of complex data points.
3. Split the data into multiple charts
In contrast, don’t feel like you need to show all the data in one chart. Multiple datasets may only need to be in the same chart if they need to be contrasted against each other. Considering breaking out a chart into two or more smaller charts to simplify the visualisation and help with readability.
4. Highlight and show 1 key statistic
You don’t always need to show data in graphs. The idea you are communicating may be better displayed as just a statement instead. For example, rather than showing a pie chart with all the responses, perhaps just state that 74% of respondents said ‘no’.
5. Fractions are often better than percentages
We aren’t robots. So it should come as no surprise that when we communicating data in the form of digits, it can be difficult for people to interpret. Proportions listed as percentages don’t always pack the punch that they should. For example, instead of saying that 90% of kids eat Weetbix for breakfast, say that 9 out of 10 kids enjoy those little blocks of cardboard. We find it much easier to visualise 10 children sitting at a breakfast table than we would a 90th percentile statistic.
6. Be strategic with numbers that are too big (or too small)
When we see large numbers, there becomes a point where we stop appreciating how big they actually are. Rather than saying that there were 330,000 people at the protest last week, a news report might say that there were enough people there to fill the MCG more than 3 times. While we may find it difficult to picture what 330,000 people looks like, we can easily imagine a stadium full of people and then simply multiple that by 3. The same thing can be said for extremely small numbers or percentages. Try and frame the data in a way that people could accurately grasp their magnitude.
7. A picture is worth a thousand numbers
Use pictograms to add visual appeal to your data. For example, rather than creating a pie chart to show a proportion of sports people play, instead create a graphic with 13 basketballs and 7 footballs. Your readers will not need to check any legend because the data themselves are the labels.
8. Sense-check the data
There’s a saying that says the average person has one ovary and one testicle. While this is statistically true, it’s obviously extremely unlikely for one person to have this combination. Similarly, statistics may not be giving an accurate representation of the true situation. Always sanity-check the numbers to make sure they pass the sniff test before blindly showcasing them as truth. If you don’t double-check them, someone else might do it first and make you look foolish.
I hope these tips have inspired you to get out there and use data as a key tool in influencing your stakeholders. The key objective with a visualisation is for it to effectively deliver the message you’re trying to communicate. With this principle at heart and by using the tips I’ve presented in this article, you will be well on the way to developing powerful messages that connect with your audience.
