One of the important tasks any data analyst will have to master is the representation of data i.e. ‘data visualization’. It is a process through which data can be depicted using charts, plots, graphs, and animations. These representations must help the users to visualize a story that was otherwise hidden within the data. Quite often, it is easy for data analysts to get carried away with the creation of dashboards filled with vanity metrics and illegible graphs that are difficult for the users to interpret.

What change in approach should you consider if you find yourselves ending up in this situation over and over? This is where design thinking comes in. Understanding your users, questioning presumptions, and coming up with creative ideas you can prototype and test are all steps in the iterative process known as “design thinking.”. Nielsen Norman Group, defines the design thinking flow in 3 major steps which are a) understand, b) explore and c) materialize. The flow can be subdivided into six phases – empathize, define, ideate, prototype, test, and implement. The “Empathize” and “Define” phases help us understand the problem and product. Ideate and prototype to explore a solution and finally test and implement phases to materialize the solution in the real world.

Let’s now explore how to apply these six phases of design thinking in data visualization.

❤️ Empathize

This step primarily focuses on requirements gathering and understanding the audience and their needs.

i. Who are the key stakeholders? What do they want to know from the data? What do they know already?

ii. What are the pain points the end-users are facing? What problem does your product solve? Are they unhappy with the current implementation if any?

iii. Get clarity about the data. Ask questions. Ask questions to external stakeholders. Ask questions about the data model to engineers and designers within your team.

? Define

With all the information you have gathered from all the stakeholders, define the key outcomes for the visualization. What are the key metrics and KPIs for the product? Identify and define the problem statement in this phase.

? Ideate

This is one of the most exciting phases of design thinking. This step involves brainstorming and storyboarding with the entire team. Understand how the end-users use the product, study the data collected and choose the metrics you want to display on the dashboard. Decide the graphs that are best suited for visualizing a piece of information. Here is an excellent article on choosing the right chart for your data. Once you have charted a journey for the dashboard users and decided ‘what’ information and ‘how’ it needs to be visualized, move to the next step.

? Prototype

In this step, you bring together all the different visualization components and create the final frame. By now you must have decided on the color scheme, information hierarchy and how to position various graphs and charts within the dashboard. Remember, the visualization must be minimal and must tell a story. If production data is not available yet, dummy dashboards can be created using test and synthetic data. Loop in designers in your team to get feedback about the visualization. The standard templates can be created which will have details about which fonts and font sizes to be used for titles, subtitles, how much space should be kept between the charts, and any other details that can be required for building the dashboards.

? Test

Deploy your dashboard to a small subset of team members first. Get feedback from cross-functional team members. Send invites to external stakeholders as the next step. Keep a track of the bugs and issues that users are facing. Remember, even negative feedback is valuable. Ask questions to the users. Did they get what they were looking for? Did they face any issues with the visualization tool?

? Implement

Finally, after multiple iterations and finalizing the visualizations, deploy the same to production. One of the challenges you might face is the magnitude of data between test environments and production. Keep a track of the loading time of graphs and make optimizations if needed. 

Design thinking allows you to think out of the box – the box represented by your usual approach to data and visualization. You can make data visualizations that are user-centric and incredibly effective for any audience if you adopt design thinking. But design thinking is not a silver bullet that will magically solve problems with the creation of data visualization. It is an iterative and cyclical process and not a linear one. You may adapt and improvise these steps based on your needs.

The key takeaway is to put your users first and understand their needs and the product you are building.