Some considerations for data visualization in professional settings
Data visualization is important in professional setting. It has been around for a long time. Even before the appearance of data viz software and tools like Tableau, people used Excel and PowerPoints to produce spectacular dashboards and reports. Regardless of the tool, there are always a few simple but important considerations when we are producing reports in professional setting. In this post I will try to discuss a few of them based on my experience.
What to display: Like most of the things that we do in life, decision making is also essential in data visualization. In data visualization we need to decide what is the best way to represent the data. Sometimes a simple table is good enough. In other cases, we use charts: bar chart, pie charts, scatter plots, histograms, etc. But not all of them are useful in every situation. Depending on the nature of the data (e.g., discrete or categorial) and the amount of information (number of dimensions/variables), we have to decide on the correct chart type. However, audience is a key consideration in the selection of chart types. E.g., if the audience is not accustomed with boxplots, it might be a better idea to use something else even if boxplot might be the most appropriate for that dataset.
Consistency: If you are using data viz for a large size of audience with a lot of charts, then it is very important to remain consistent across charts/pages/slides. E.g., I used to prepare reports for clients that included sales data (among other things) at company level, brand level and SKU level broken down into different sales regions and distribution channels. Even though the person who was in charge of region A had little interest of what was happening in region B, but the senior management were very curious about the performance across regions. Some of these presentations and reports were large (more than 100 slides). So, imagine a senior manager looking at slide 80 after going through slide 10, and if the colors of the competitors in the charts were inconsistent, it would have been very confusing for them.
Format: By format I am referring to report structure. Sometimes, your audience is so used to a certain format that they would prefer if the format and chart types were the same from one period to the other. This is especially relevant for periodic reporting. One of the reasons could be that people do not like change. But most importantly if the format remains the same, then the audience will have to put less time and effort in understanding the format and can focus on the story instead.
Tell more with less: It is always a good idea to simplify things for the audience. Rather then having the audience going through texts, it is better to represent the story in a chart so that it can be easily captured. E.g., for my research project I decided to use bar chart to display the outcome of my experiment. However, If I did not use different colors (green and red) for two types of features, the reader would have to go through the texts to understand what it meant. Instead, marking important and noise features with different colors allowed the reader to capture the information right away.
This is a very simple example, but similar ideas can be implemented to other cases to ensure that audiences grasp the information easily and quickly.
Improvisation: You might ask, if there are so many considerations (or restrictions) in data visualization, then is there a way to showcase my data visualization skills. The obvious answer is it depends. If you are working on a one-off report/presentation, and the audience and your organization allow for creativity, then you can improvise as much as you can. However, if the audience like set format then it is much more difficult to improvise. In that case, you can include a high-level summary page where you can showcase your data visualization skills.
Modern data visualizations tools and software have made it easy for analysts to make stunning visualization in quick time. Despite all the innovation and technical progress in the space, your audience remain central. It is not about what I want them to see. It is about what they want to see. It has always been like that and there is no reason to believe that it will change in the near future.
