How to use data visualization to tell compelling stories

We’ve all been in those meetings where the topic should be interesting. It could be really important, only for you to lose concentration as one slide of bullet points after another fly by. There might even be a few stats thrown in for good measure.  

What if there was a way to make the information more exciting? They’re talking about the future of the business, after all, shouldn’t this be engaging?  

Data visualization is not only a way to keep the attention of the stakeholders you’re presenting to, but also a necessity. When you’re trying to convey the true impact of a statistic or relay the importance of a business factor, data visualization is the only way to tell the story in a compelling way. 
 

Start simple 

This may seem like an obvious one, yet one of the biggest mistakes made by Data Analysts or Business Analysts is creating confusing visualizations. A graph might show 3 or 4 different variables that aren’t relational, which makes for chaotic viewing and doesn’t provide a clear picture of what the data is saying. 

The best approach is to begin with one KPI at a time. It may be a few minutes of extra work, yet the overall effect on the presentation will be huge. The stakeholders will clearly see what is being represented, and they’ll be able to digest and remember it. 

After this, you can add in further visualizations displaying the relationships between different points, such as multiple scatter graphs or a line graph showing two trends, if it provides substance.  
 

The right visualizations for the right job 

There are an extensive number of data visualization tools out there, and it can be easy to get carried away. You feel that you don’t want to be boring, so you might overuse pie charts, line graphs, 3D stacked bar charts and however many others.  

However, this is quite often the worst choice. A simple line graph in differing colours is quite often perfect, or a stacked bar chart will show the data better than a three-tiered donut chart. 

Going back to basics; line charts show trends, column charts show side-by-side values, and bar charts break things down simply. Scatter graphs show relationships and distribution, and pie charts show proportions. 

Always make sure that the choice you make is due to the clarity of the data. It might be obvious to you if you work with data as a job or use it on a regular basis, but that might not be what is best for the client or the stakeholders. Double-check the visualizations and have a colleague look them over. 

Use easy-on-the-eye patterns 

Humans are naturally drawn to patterns and are able to draw information from properly structured visualizations quickly. If random patterns are used, then it takes longer and becomes far more difficult to comprehend what is being shown. 

When you’re working with data showing year-on-year, then the graph will jump up and down, yet the years are ordered so it’s understandable. If the chart is showing the top sales performers, order it from highest to lowest. Whether it be alphabetic, numeric or sequential, the data presentation needs to make sense, and quickly, to the people studying it. 

Size & colour 

Whilst colour is often used to make visualizations more aesthetically pleasing and easier to understand, size is an often-underused visualization technique.  

Colour can be used to show contextually good and bad values, it can be used to differentiate variables from one another on a graph, and it can be used to show gradient values. Whether it be judging performance results, comparing year-on-year values, or displaying temperature changes, it’s simple to convey the information quickly. 

Size can operate in a relatively similar fashion. If you’re comparing values on a scatter graph or map, such as product interest across different countries, you can see with a glance where you might invest or where to focus your efforts. 

Combining these together makes the best use of both. Plotting size on a scatter map to show population, and then using colour to show another variable, such as volume of social media users, allows your insights to become immediately clear.  

Be careful with text 

Text can be essential but it can also crowd the visualization. It’s necessary for axes are and giving more clarification on data points. Often, the data is obvious with 6 words in total and doesn’t need a lengthy title or explainer box on every value. 

If text is used, make sure it’s relevant to what is being shown and that it doesn’t make the visualization difficult to read. There might be 100 products or stores on your pie chart, but if you can’t see the data through all the names you’ve tried to squeeze in, then it’s useless to stakeholders.  

Consider other ways to break the data down, such as hierarchical or donut charts. You’ll appreciate the little extra work you might need to put in further down the line. 

Remember the narrative 

Finally, one of the most important factors in telling a compelling visualization story, is to remember what the point of it all is. Don’t get caught up in creating lots of graphs that don’t relay what you’re trying to show. Furthermore, ensure that the salient information isn’t also lost amongst unnecessary data.  

Always look back at what you’re doing and think, “is this still providing the narrative that the data is telling me? Can someone else read that story if I wasn’t here?” 

Should the answer to either of those be no, then you need to reevaluate how you can amend the visualizations to do so. Sometimes that might be removing one that you’ve spent hours working but just isn’t telling the story right.  

If you’re interested in a career in Data Analysis and you would like to discuss any of these points with a Career Consultant, then register your interest below and we’ll send out a career guide for you straight away!