Tired of using the same charts? Here are some creative ways to visualise data
These examples will inspire you to display data in more innovative and insightful ways. Plus, I have some tips for you to jazz up the good old bar and pie charts.
For starters, this article is not going to go into storytelling and the whole “know your audience” kind of tips. Here, I’m going to focus on how to make your data visualisations more exciting while keeping them simple and insightful. If you want tips on how to put your data story together overall, I would recommend checking out my list of helpful articles on the topic.
Get away from the pie and bar, if it feels right
Let me start by saying I have nothing against pie and bar charts. In fact, later on, I’ll share some tips on how to give them a more modern look. Sometimes, they will do the job just fine and there is no need to use a super extravagant chart.
But I also think it’s useful to remember that other options do exist. If I’m putting a presentation together for an audience that is very used to reading data and is quite numerical, I do tend to try and explore more unusual types of charts, if they fit the purpose.
Colours speak a million words: get them right
A lot is said about minimalism in data visualisations. We’re constantly making sure there aren’t too many labels on the charts, or unnecessary gridlines and or numbers formated incorrectly. But we often forget to apply minimalism to a very key part of the composition of a chart: the colours.
This article talks in depth about how to apply colour harmony to data visualisations and explains further the use of the colour wheel. But I’d just like to pick out the most important things to remember when choosing the right colour for your visualisations.
Pick a palette and stick to it
First thing: choose the main colour that will set the tone for your presentation/dashboard and will be present in every slide/chart. When I’m presenting to companies, I tend to choose their main colour as the dominant colour. You can also choose a colour that expresses the sentiment you’re trying to transmit within your story — more details on that on this article.
After that, you have to decide the additional colours that will form your palette. I have been using this website to create palettes. But you can also choose your additional colours based on the relationship between the elements in your chart. This article offers a very comprehensive guide on how you should choose your colours depending on the type of analysis your’re doing.
Colour is also about culture and bias
When thinking about colours, it’s very important to take the culture and context of the audience your visualisations are intended to. However, I do like to challenge certain stereotypes with my charts. You’ll never see me using blue for Male and pink for Female when referring to gender, for example.
We should also be careful not to let our biaises show - specially in demographic charts and maps - by using “positive” and “negative” colours according to our subconscious (or even conscious) beliefs.
Compacting a lot of data efficiently
Every data analyst has gone through the conundrum of having to show a lot of data for the story to make sense but without making the slide or dashboard too busy.
Unfortunately, I don’t have a magic solution for this problem and I’m always learning (aren’t we all?). But I wanted to bring you a few examples of this being done well.
Using colour and positioning in your favour
This is a report with tons of data spread across several pages. However, they created the visualisations in such a way that it’s very easy to follow the story. I’ll be using more examples from this report in this article — and I so recommend you to check the full report for inspiration — but here I want to call out the use of colours and positioning to separate two sections of the page. Key takeouts of the page on the left with a green background and more in-depth data on the right with a white background. So simple, yet so effective.
The same report used the visualisation above to put three different proportions into context very effectively. Note how the percentages shown don’t add up to 100%. However, as they were gathered from the same pool of respondents, they used semi-circles around the same centre to show these percentages. Also, notice how each semi-circle has a different colour to denote that they are not part of the same category. Again, colours speaking more than a million words.
Watch out for this when putting charts side-by-side
Depending on the situation, though, using two separate charts next to each other would make more sense. When doing that, make sure that they are not misleading the audience.
For example, for the presentation below, I had to put two bar charts next to each other. As they were all talking percentages, I had to be careful with the scaling so the 30% bar on the left wasn’t taller than the 36% bar on the right, for example.
Do you really need all that data in there?
In the situation above, I did need to include all the data in the charts. But how many times have you placed an entire chart in a presentation just to call out one particular number? This is something to think about when trying to reduce clutter in your presentation.
I’m sure you don’t want a number just loosely floating on a slide, though. Neither do I, so here are a couple of ways to call out a single number in an exciting way.
Over-indexing put simply
Analysing over-indexing categories against a measure can be super insightful but also super complex to explain to someone who is not used to reading data.
I often struggle to find a way to effectively represent over-indexing in a chart without making it look too busy. The report I mentioned previously did it in a very clear way, though, using only two bars, two shades of green and two font sizes.
Tired of eating pie, why not try these instead?
Pie charts are not the only way to demonstrate percentages of a total. Again, I’ve got nothing against them, but sometimes by replacing them you get an analysis deck with less charts, more creativity and an equal amount of insight.
For example, you can use illustrations related to the data you’re showing to demonstrate proportion. This can be very easily done by downloading an icon from Flat Icon using their Edit Colour feature. Download a white microphone icon and a pink microphone icon, place them on top of each other as needed, and boom! A great visualisation with little to no design skills.
Giving bar charts a more cool and modern look
Nothing upsets me more than seeing a bar chart with the default Excel colours, font and gridlines. Ok, there’re actually more upsetting things in life. But let’s make one thing clear: you don’t need a Tableau certification to make your bar chart look fresh and exciting.
Be minimalistic
You may not be a minimalist in life but you’ll want to be minimalistic when it comes to bar charts, trust me on this one. Removing noise from your bar chart not only makes it look fresher but it also helps to highlight the key information in the data.
So, don’t be afraid to get rid of unnecessary gridlines, labels and borders. I’ll go even as far as saying that sometimes not even the axis are necessary. Obviously, don’t take away relevant elements of the chart, but you’d be surprised how cleaner charts tell a much clearer story.
Use Excel’s chart setup options to jazz up your bar charts
I’m not going to go in detail about the ins and outs of customising Excel charts — maybe this could be the topic for another article. However, I highly recommend you take a look at the possibilities next time you create a chart.
By altering the labels, the series name and fonts (I like to use the client’s font type in the charts, for example), you can make the old boring bar chart look fresh, modern and, most importantly, more insightful.
All in all, there are no rules to data visualisation, but…
You can’t let your own visualisations sabotage all the hard work you put into an analysis. Fortunately or unfortunately, a bit of design knowledge and creativity can go a long way to help you tell more clear and insightful stories.
PS: I constantly stumble across new, innovative ways to display data. So, if you have any other examples to share, please feel free to share them in the comments or send me them via Linkedin.
