Breaking down the coolest Data Visualization!
Data Visualizations are the best ways to get a feel for your data. In any data scientist’s or data analyst’s work, they predominantly own the EDA processes and crunch data in a way like no other process does. They go from the most basic bar plots to kernel density estimation plots (don’t bother what they mean) and can decode any patterns, outliers, details, etc. about your data.
With software like Tableau, PowerBI, Qlikview, and the capabilities of programming languages like Python and R in terms of Data Visualisation, the possibilities are unimaginable.
Before we dive into the article, I want you to look at some cool visualizations from the Iron Viz competitions hosted by Tableau, with people contributing aesthetic and informative visuals to the data community while showing off their skills!
Major note: You don’t really need complicated data visualizations to present your insights to your stakeholders or show data in your projects, these are visuals that are very timestaking to understand at times and carry enough information to take up the entire presentation as a whole.
So if you frequently have to make a data viz in your work, I wouldn’t really recommend any of these but nevertheless, they’re still pretty amazing to be acquainted with.
Check out all the submissions to the Iron Viz competition in 2022 here:
Interesting, aren’t they?
That being said, let us look at 5 visuals that you can make with a little bit of effort that is better than bar charts, pie charts, etc., and break them down so that you can take your data visualization skills to the next level!
The visuals discussed in this article are as follows;
Sunbursts
Sankey Charts or Alluvial Diagrams
Waterfall Charts
Radar Charts
Word Clouds
Sunbursts
Sunbursts are an upgraded version of the pie chart and are not recommended if you have too much data to show. For example like this:
But otherwise, if you want to combine some information within a circle and want to portray different segments of things with multiple layers, the sunburst chart is for you. They can break down different segments further and show the proportions held by each segment and its sub-segments!
For example like this:
Fairly easy to make on software like PowerBI, Tableau, or even Ms-Excel, try using a Sunburst chart in your work sometime soon, and let me know how it works out for you!
Find out how to make a Sunburst here:
Sankey Charts or Alluvial Diagrams
Sankey Charts are one of the most informative graphs I have come across in my entire life. The reason for this is very simple. Consider you have customers (taking the example because it's the same in the image as under) and you have regions. So you want to find out, what part of your customers are in which region based on the Sum of Sales they give.
The fairly simple to understand Sankey chart, shows the following information:
- No. of customers in each segment (percentages)
- Regions and their percentages
- What part of sales comes from Consumers to each region based on the size of the curves
- The sum of sales is easily differentiated by the size of the flow of each component on the left to multiple corresponding components on the right.
In all honesty, while a Sankey chart might take you some time to build, it's a lot of value addition to your presentations if it covers your focus areas. Find out how to make your own Sankey chart using Tableau here:
Waterfall Charts
If you work with data like sales and have used bar charts to depict your data, you know how it can show you the changes in sales, say over different months in a year. The same thing applies to a setup of a Waterfall chart, only that the difference comes when the bars are eliminated everywhere except the first and last data point, and the rest depict changes relative to every preceding data point.
For example, in the picture below, the partial blue bars depict additions to the values preceding them, and the red bars, the subtractions which are clearly shown with a minus sign.
While showing the overall sales for every month would be similar to waterfall charts with all of the bars starting at 0, a running total makes it easier to identify where the rise and fall are very evident and by how much. This makes pinpointing a data point to emphasize it easier.
One of the easily understood data vizzes of the 5 discussed in this article, use Waterfall charts if you think your audience will understand them and see how your presentations deliver the message in a better way.
Make your own waterfall chart in Tableau using the tutorial below:
Radar Charts
Think of a line chart but more defined towards specific categories and very useful for comparison. That’s where the Radar Chart comes in. A Radar chart is one of the more useful versions of a line chart with multiple categories (preferably 4 or 5) being placed in the shape of a square, pentagon, or another polygon depending on the number of categories being taken.
The quantifying value (for example Ratings as shown in the picture below), will then be measured across the different edges of the polygon depending on the different values taken. Different colors can be used to depict the same ratings or values for different people/hues and can be easily compared.
As you can see in the picture above, it's clearly evident who is more punctual and who delivers more knowledge of the three trainers being compared. Try using it for your data and see how your comparisons become easier with a Radar Chart.
Make your own Radar chart using the tutorial below:
Word Clouds
Word clouds are another easy-to-understand but easily complicated visual that you can use in your work. A word cloud can show multiple elements that are textual and resize them based on an underlying calculation. For example, if the calculation is ‘count’, then the size of the word will correspond to the number of times it occurs in the data.
The larger the times it occurs in the column or data, the larger the word is. The words are then placed in a cluster for a viewer to essentially read a word that is highlighted or simply, the biggest words.
Try making a word cloud using this tutorial below, for making one on Tableau:
With impossible not being a term in the world of Data Viz, this is your call to go out there and find out insane visuals to depict your data with and master the art of visualization.
Try making some of these yourself and let me know what you think about them in the comments below!
(Note: I use Tableau Public to make my visuals and you can try to make these on the same! It’s fairly simple to use and takes barely any time)
Also check out my previous articles here, if they interest you!
For more such articles, stay tuned with us as we chart out paths on understanding data and coding and demystify other concepts related to Data Science. Please leave a review down in the comments.
Check out my other articles at:
Do connect with me on LinkedIn at — Yash Gupta — if you want to discuss it further! Leave a clap and comment below to support the blog! Follow for more.
