5 simple ways to improve your data visualisations
You don’t need to be a Tableau wiz to create compelling visualisations. Follow these 5 principles and give your insights the visuals they deserve!
In this article, I will use my 3-year experience as a Data Analyst to show you very simple ways to make your visualisations look more attractive, easier to read, and professional.
Let’s pretend we are creating an analysis for a shoe brand and the client wants a slide with the following information:
- They want to know what is the overall purchase intention for their brand and if it increased vs. last year.
- They also need to see the current purchase intention by store.
- They were wondering if product unavailability deterred clients from wanting to purchase even if other methods were offered.
- Finally, they were interested in figuring out what alternative methods were being offered and how this changed vs. last year.
Following the 5 principles I’ll introduce below, this is the slide I created:
So, let’s review the principles and see what that slide would look like if it weren’t following them.
#1 Don’t underestimate whitespace
With datasets becoming richer and richer, Data Analysts are often able to extract so much insight for a report that it becomes a challenge not to overload the visualisation.
But trust me, whitespace is as important as the visualisations you include in your slide. The last thing you want is important information buried in a myriad of small tidbits, fun facts, and charts.
Here’s the same slide above but with almost no whitespace:
Unnecessarily large charts, granular details that the client didn’t ask for, and content very close to the borders of the slide. All this buried the key insights I wanted to transmit and made the slide very hard to read.
For example, in the unavailability by store, notice how the increase in Store 7 which was carefully highlighted in the previous slide using very little space and a fun icon is now almost impossible to spot.
So, carefully sift through all the information you have available and distinguish what are the most important insights you need to convey. You can always add extra information in the appendix or on an additional slide.
#2 Make your colours consistent
In the interest of reducing unnecessary information, enabling the audience to understand the main insight quicker, and creating more whitespace, colours can be your best friend.
In this article, I went more in detail about the use of colours, but here is what our slide would look like if I hadn’t coordinated the colours:
In the original slide, it was possible to make links using colours. Grey always referred to 2022 data; blue was data related to unavailability, and green was the colour of positive purchase intention. You’d glance at the slide and the message would be transmitted much quicker.
Here, you have to actually read all of the text to understand what’s going on because the colours don’t aid you in that. Also, notice how the colours for the “Methods offered” chart is inconsistent with the rest of the slide and pulls attention to it when actually that’s not the main insight.
So, think about your colours carefully. Fortunately or unfortunately, if you tell stories with data, it’s useful to have some basic notions of colour theory.
#3 Pay attention to alignment
You put so much work into creating this insight, then you put it into wonky sections like this and all the audience can see is how everything is so misaligned.
In case you don’t know, on Powerpoint you can use the Align function to make sure all elements are where they should be (Drawing > Arrange > Align).
Also, notice how I removed the line that was next to the “Methods offered” chart and now it’s just floating there. Use section dividers to direct the eyes of the audience in the flow you want them to follow.
#4 Make use of alternative visualisations
I’d say there are worst data visualisation offenses and you’ll see that the slide below doesn’t look as bad as the previous ones. But repeating the same type of visualisation throughout the slide definitely takes away some of its spark:
If you’re ever feeling uninspired, check out this article I wrote with lots of creatives ideas to display data differently. I also recommend reading reports and infographics other people make to increase your repertoire of visualisations.
I feel like when you’re presenting to an audience that isn’t so data-driven or maybe even hates data, avoiding the same old charts can make more of an impact on them.
#5 Use visual cues to highlight information
Going back to the point of the large amounts of information Data Analysts often have to display, at times you can get all of the above right but the key insight remains lost like a needle in a haystack.
Here’s the original slide minus a rectangle, a warning icon, and a line:
Even if the alignment is good, everything seems to be floating. I’d still look at the Purchase Intention information first, just because our eyes tend to read from left to right, but it’s not super clear how the sections relate to one another.
Also, the very alarming information about the increase in unavailability for Store 7 is just not getting the attention it deserves.
Rectangles with a bit transparency, icons, lines, and arrows can all help you to direct the flow of your information, highlight what needs to be looked into with more attention, and just make your slide a bit more fun. For the icons, I like to download them from Flat Icons but just ensure your icons have similar line widths and design so it’s all consistent.
I’d be curious to know how you guys would improve my original slide and what you thought of the tips in this article! Check out my other stores for helpful first steps to land a Data Analyst job, how to start building your portfolio, and what it’s like being a woman in the tech world.
