Data Visualization: 5 Most Important Things to Know
Hello everyone. In this medium post, you’ll learn the five most important things you need to know to improve your data visualization skills. By the end of this post, I promise you will have the knowledge to level up your game in visualization. Let’s get right into it!
Simplicity is the key.
The first and most crucial element of quality visualization is simplicity. It is so important that I must emphasize its significance before getting to the rest of the post.
Think of simplicity not as a reinforcing factor but as a prerequisite for data visualization. The whole point of visualizing data is to tell a story to the business owners and help them make decisions based on this story. There is more than one way to do this. However, to get the point across right away, we must avoid any element that might confuse our audience.
Business owners don’t have hours and hours to spend on your graphics and visuals. They have a business to manage, meetings to attend, decisions to make — all of which take time. As a data visualization specialist, it is your job to make it easy for them to understand the story behind the data. To achieve this, your final visual must be as simple and comprehensible as possible, so they get the idea right away.
Now let’s start on the five most important elements of data visualization.
Choose a graph that complements the data you will storify.
To create an effective story and support our point, we need to choose a graph that is appropriate for the data. The image above demonstrates this key element perfectly. It shows what type of graphs we should use to compare different attributes or relations or to visualize a distribution.
So, what does it mean to choose the right graph for the data?
Let’s say you have data that you can use to show how a metric changes over time. What graph would you choose to visualize this data? Can you capture the time-dependent changes or the changes between two points in time with a pie chart? It is equally unlikely to see such changes with a histogram or a radar chart.
The most suitable graphs that you can use to visualize time-dependent changes are line charts and area charts. Why? Because such graphs have two different axes. We can show time on one axis, and on the other, we can observe the data realized at a specific point in time. We can also observe how the realized values change between two different points in time. In this case, we can say that a line chart or area chart is the most suitable graph type to visualize such data.
Don’t use hard-to-read fonts.
Font selection in data visualization is one of the key elements that deserve our attention, yet many visualizations fail to meet this standard.
The human eye involuntarily follows the strongest emphasis and appeal. So if you use complicated, fancy fonts, you run the risk of distracting your audience. We already know that an effective story is also simple and understandable. Since your goal is to help people make the right decisions quickly based on your story, distraction is not a risk you can take. You should not divert attention with unusual fonts if you want the focus only on the points you want to emphasize.
The figure above shows an example of the good and bad use of fonts.
Preserve visual integrity with correct arrangement.
When visualizing data, we need to make good use of available visual space. It is a common misconception that more information is better. Instead, we need to present as much information as possible with comprehensible and simple graphics. Both criteria are equally important. Good visualizations cannot compensate for insufficient information, and vice versa.
When practicing data visualization, think of each object as a container. For impeccable visuals, all containers must be perfectly aligned. If we place one graphic on top and the other across, it will look messy.
When placing visuals, keep the following in mind:
- The start and end points of all containers must be aligned.
- Font sizes must be consistent overall. Of course, you will use different sizes depending on the intended use. For example, the main heading will be larger than the graph titles, but make sure they are consistent, i.e., that all graphic titles are the same size.
- The indent and the hanging indent must be identical for all graphs. Assuming that each container has a graph, the graph must be uniformly indented on all edges.
- The white space outside the placement area should not be too large.
Quick tip on alignment:
Imagine working on a grid. Unfortunately, not all data visualization programs support the grid view, but we can focus and imagine having one while working on our graphs. And trust me, it works.
Deciding on the arrangement of the graphics
There is no written rule for placing the graphs in visual space. The only way to get better results is through experience because, in time, it gets ingrained in our heads. Are you a true fan of data visualization? It may not be the answer you wanted to hear, but if you are, you need to practice as much as possible to acquire the perfect arrangement of graphics. Why? Because we cannot possibly say that it’s always better to place the metrics in the top right corner or that the visual headline must always be aligned top and left. Place the line graphs at the bottom and the bar graphs at the top, and you are done! Does it make sense? No. It’s impossible to make such rules. It all comes down to what works for your graphs, your audience, and your data.
Choose the right color palette for data.
Before we start visualization, we must have knowledge about the particular industry on whose data we’ll be working. In this section, I will talk about why this is important and what can happen if we proceed without this knowledge.
The image above shows a data visualization of food and beverages. Do you think the use of orange-based colors in this image is coincidental?
If you have ever studied color theory, you probably know the answer. These colors were chosen intentionally. Colors have their own meanings and trigger certain emotions on a subconscious level. Orange in this image above represents sincerity, friendship, and intimacy. Considering that this visualization was created for the food industry, we can say that the color choice was strategic. It was meant to evoke feelings of candidness and sincerity and make people feel more comfortable by appealing to their subconscious.
Most well-known fast food brands use the same trick. Their logos are usually red or orange.
So what do you need to do? If you are working on a professional data visualization project, choosing the right colors can create a familiar visual language and make your audience feel comfortable. If you are working on a visualization for a specific company, ask for their brand guide to create the appropriate color palette. If the visualization is for a particular industry, do thorough research and determine what colors you should use.
Long story short, if you are a data visualization specialist or a front-end developer, you must definitely take an interest in color theory.
Know your audience.
Analyzing your target audience before starting the visualization process will increase the efficiency of your work. If you know your target audience, you can give them exactly what they need. You’ll also save time by not creating something you think is essential but the target audience does not want.
Let’s look at it this way. You have an audience who doesn’t know how to read graphs or has never seen complex graphs in their lives. What do you gain from presenting them with such visuals if they cannot understand the story you are telling? Not much. You must take a different approach. You can give brief instructions on how to read the graphs before you present your story, or you can simplify your work so that the audience can understand the data you are presenting. For example, if your audience has no science background, it would be pointless to create a visualization that is full of scientific information.
Try answering the following four questions to get to know your audience. Read this blog post if you have trouble answering them:
- Who can use the information?
- What do you want them to do?
- Who’s most likely to do that?
- How can you reach them?
Then go in deeper and ask questions like the following that can be tailored to your target audience: What will end users look for in your visualizations? Do they have enough time for you to present a detailed visualization? How much do they know about the context of the visualization? What additional information are they looking for? Are they familiar with the graphs used in the visualization? Answering questions like these will change the outcome of your work to match the needs of your audience.
Finally
Thank you for reading this far. If you are familiar with data visualization, you have probably heard these key points before, but I hope the examples I have given have made them clearer for you. Until next time!
Resources
[1] Verma, R. (2020, June 11). Data Visualization: Top 5 Most Important Things to Know. LoginWorks. https://www.loginworks.com/blogs/data-visualization-top-5-most-important-things/
[2] Geere, D. (2020, March 19). How to Identify Your Audience. Bar&Line. https://blog.duncangeere.com/how-to-identify-your-audience/
[3] Gomes, M. M. (n.d.). Data Visualization — Best Practices and Foundations. Toptal. Retrieved March 11, 2022, from https://www.toptal.com/designers/data-visualization/data-visualization-best-practices
[4] Aniruddha. (2020, July 27). 8 Data Visualization Tips to Improve Data Stories. Analytics Vidhya. https://www.analyticsvidhya.com/blog/2020/07/8-data-visualization-tips-to-improve-data-stories/
[5] Wong, L. (2020, September 23). Data-Driven Storytelling: 9 Techniques for Effective Visualization. Kantaloupe. https://www.gokantaloupe.com/blog/best-techniques-for-data-driven-storytelling
[6] Sunarto, N. (2021, October 18). 55 Facts and Statistics Showing Why Data Visualization Is Important. Piktochart. https://piktochart.com/blog/data-visualization-statistics/
[7] Csg Solutions. (n.d.) 6 tips for creating effective data visualizations (with examples). Retrieved March 11, 2022, from https://blog.csgsolutions.com/6-tips-for-creating-effective-data-visualizations
