There’s a joke that goes around in our team.
What’s scarier than rogue or incorrect data?
Bad data visualization that complicates understanding and doesn’t add any value.
And just like any other joke, there is a dark undertone of truth in this statement. One might have loads of interesting data gathered from years of research. But if people can’t gather valuable insights from it, it’s as good as not having it. This is where effective data visualization comes into picture. If your data doesn’t tell a good story, there’s little chance that it will solve any problems.
It’s funny that I realize it now. A few months back, I was scared of data. I had this untameable fear of designing dashboards, or anything that included data. It made me anxious because I didn’t know where to start.
But then, my mentor recommended that I read this book – Storytelling with Data by Cole Nussbaumer Knaflic.
After reading this book, I was blown away by the simplicity with which Cole has explained such a complex topic. When I started reading it, I expected numbers and complex data-driven concepts but to my surprise it took me to a ride of enjoyable language play and practical examples. After every concept, I was in awe of her ability to evoke emotions through words and the way her concepts can breathe life into any data.
Quoting a line from the book ‘There is a story in your data. But your tools don’t know what that story is. That’s where it takes you – the analyst or communicator of the information – to bring that story visually and contextually to life. That process is the focus of this book.’
Inspired by this book, I thought of listing down a few of the concepts that I feel are important for all designers. Please know, this attempt will only bring you an idea of my understanding about this book and I highly recommend you to pick up this book.
Know your audience
To design something that helps users solve a problem, we must first know who these users are, what their needs are and how they will interact with this data.
Designing without understanding your audience is not only wrong process-wise but also unempathetic. While interacting with designs made from assumptions, users are most likely to feel confused, frustrated and unmotivated to accomplish their tasks.
Therefore, be empathetic to the needs of your audience. Try answering these questions before you present your data-
- Who are you communicating to?
- What information does your audience need to know?
- How can you help users arrive at a decision using the data?
And if you do not have the luxury to directly talk to your end users (quite common situation) try answering these questions-
- What biases audiences might carry with them that can make them supportive/ non supportive of our data?
- If you had to tell your audience what they need to know in a single sentence, what would it be?
Therefore, knowing your audience will help you understand what information to look for and what data to present. Often we leave it to the audience to understand the data by presenting a difficult graph full of numbers. But it is actually on us to help them see what they need to see. Our research will help figure out if users need to see quantitative data such as revenues, numbers, change in percentages or if they need to see qualitative data such as processes or quotes or one line conclusions.
Choose the right visualisation
Just as clothes and shoes come in different sizes, we need the right visualisation for different kinds of data. It depends on various factors like user demographics, user knowledge or expertise with the subject, amount or type of data that needs to be represented.
Since this is an important factor, Cole addresses this topic in the book with great detail. She shares that being selective of the visualization is a skill that you master with time and experience.
Below are two examples:
- Pie Chart with too many slices is a disaster
- 3D effect on the chart violates Data-ink-ratio rules
There is no denying that pie charts are simple and easy to read where each segment corresponds to the proportion it represents. However, it should be used only when the total of all dimensions is 100%, and users want an approximate idea about a portion of the whole.
However, a pie chart with more than 5 divisions becomes a nightmare for the user to relate the portion with the legend and it becomes really difficult to judge the volume of any specific portion.
Let me explain this through an example, let’s say here we present a survey data on the popularity of different job roles in a country in the year 2019. Can you confidently compare the segment of psychologist and architect ?
It’s clear how a 2D bar chart is representing the data with more clarity and it’s easier to compare the data this way. Don’t get me wrong, not always you can choose a bar chart but you can always choose when not to go for a pie chart!
Bringing design concepts into the data world
This point is definitely one of the strongest concepts in this book, because how much we designers think about “Users” before bringing any experience to the table and therefore, in the world of data representation – Audience is what you need to care for. Therefore, understanding what we want our audience to be able to do with the data and then creating a visualisation that brings this on the plate with ease, that is what it means by ‘think like a designer’.
The concept of Affordance, Accessibility and Aesthetics are the key concepts in the design world that focus on making design usable, accessible and visually better.
In design, Affordance reveals how an interaction should be used or interpreted. One simple example could be, in data visualization the use of red or green color guides users to take notice or interpret the right meaning of data.
The concept of Accessibility is quite important and loudly spoken about in the design world. It is about creating designs which are usable by people with diverse abilities. Methods of practicing it can be a separate topic of discussion but just to put a few here – support text with visuals and labels, choose a better legibility font and maintain contrasts according to WCAG guidelines.
And the next one – concept of Aesthetics talks about how aesthetic designs are not only perceived as easier to use but also readily accepted by users of different demographics. Two important points to remember while implementing it – pay attention to colors, alignment and use visual hierarchy to enhance aesthetics.
Therefore, combining the three concepts of design can bring a better experience of data representation for the audience.
I know, my thoughts are all over in this blog and maybe I haven’t covered each and every point, mainly because this book is a journey of its own. It gave me a different perspective to look at the data representation now. It’s not about boring numbers and ugly looking graphs but it’s about the ‘audience’ and now I know that we need to tell a story through our data to the audience.
Cole has successfully convinced me that once you create data for your audience, you are no longer serving just the data on the table, you are telling a story and that’s what makes sense. I recommend that even if you’re not a designer but need to look at graphs and work on those quite often, go and enjoy this journey! You’ll find a fresh perspective like I did!
Hope you enjoy it as much as I did.