Data storytelling needs visuals. As humans we process and understand visuals far more quickly than we do words. That means that it’s more efficient to use visuals to convey complex ideas, and ensures that the stories being told aren’t misunderstood or missed altogether.
For instance, important patterns, trends and correlations that might go undetected or unappreciated in text-based commentary, can be exposed more quickly and effectively using well-presented visuals.
Having said that, it’s also important to match the visual with the data you want to convey. Using the wrong visual can easily lead to a greater level of confusion than if you’d simply written your data story as a series of bullet points.
Communicating Correctly
Being able to select the most effective visual for a particular data set ensures that you can communicate the right message. In turn, this means that insights and actions taken as a result of your data storytelling are more likely to be relevant and realistic.
If your data and its visualisations are going to have the desired impact, the narrative around it should guide the audience towards what needs to change by explaining, engaging and enlightening.
In my last blog post, we explored the relationship between data, visuals and narrative. If change is the sweet spot of storytelling, we must understand the role of and relationship between all three if we are to reach it.
Data Visualisation: Getting Started
There are many factors that influence the best way to visually present your data story. The most effective way of understanding these factors is to consider the bigger picture before settling on the format of your visualisations. They are:
Goal - Know what your goal is before even coming up with an idea. Are you trying to engage your social followers, increase traffic or drive transactions or goal completions?
Story - An effective data story isn’t just a smattering of stats, it involves a clear “data ask”. What are the insights you want the viewer to leave with?
Volume of Data – Be wary of the tendency to go “data-crazy” and extract too much data. Planning before analysing reduces this risk.
Audience - Who are you trying to reach with your data story? What is their level of knowledge or understanding?
One of the best ways to sabotage your data storytelling is with incorrectly or ill-designed data visuals. Data visualisation is meant to make the data as easy to understand as possible. Data visuals are the foundation; without strong data visualisations to support your story, it will crumble.
Making the Right Decision
The following decision trees can help with the process of choosing the most effective way of visually reflecting the data in the most impactful way.
There are 4 data analysis themes, shown below:
Comparison Analysis
This technique is used to measure the relationships between variables over two or more reporting periods. Do you need to compare variables over time? Do you need to look at the growth in sessions, by month vs. last year? Year on year change in number of sessions by month? This powerful analysis uses visuals to show trends, outliers and change over time.
Composition Analysis
By looking at the component parts of the whole, this technique conveys relative information, including probabilities, proportions or percentages.
Do you need to understand the relative importance or contribution of variables? With this technique, you can look at share, contribution and any split that adds up to 100%. For example, What is the device split of sessions on a website?
Relationship Analysis
This measures the relationships between 2 or more variables using statistical techniques. This is a less common analysis that looks at the relationship between two variables. Is there a positive or negative relationship? If session traffic goes up, do transactions go up significantly?
Distribution Analysis
This is a list showing all the possible values of the data and how often they occur. When a distribution of data is organised, patterns emerge and the number or percentage of individuals in each group become evident.
Conclusion
Data visualisation expert and author of Now You See It, Stephen Few, highlights the human element of data analysis saying, “Numbers have an important story to tell. They rely on you to give them a clear and convincing voice.”
In the area of digital data, comparison and composition representations will be the most useful. Define audience and goals, plan before tackling the numbers and decide on the type of analysis best suited to the message.
Give the numbers a voice, get the visuals just right!
