Unlike computers, humans do not have great capacity to process huge amounts of data, especially if it is represented in the form of extensive reports with countless rows and columns. However, when information is represented visually and aggregated in some way, the human brain is able to quickly detect patterns, understand trends and detect anomalies. For this reason, data visualization is an essential characteristic of any analytically minded executive.

‍

The representation of data is very important in communicating results , as it allows a large amount of information to be succinctly presented, thus focusing on the substance and not the technology or methodology. However, this representation and visualization of data is equally important in the initial phase of data analysis, as a descriptive tool, giving the analyst the opportunity to explore the data and understand them, in different dimensions and levels of detail, even encouraging the detection of patterns and anomalies. A good example of how important data visualization is is what Francis Anscombe presented in a scientific paper in 1973.

‍

‍

In this, the statistics expert demonstrates that with four different data sets, apparently with similar statistics, their diversity is easily identifiable if the data is represented visually .

‍

‍

The Anscombe Quartet , as this data set is now commonly known, exposes how humans can effortlessly see patterns and counterpoints in a graphical representation , but have difficulty doing so when the same data is presented in data tables, in report form.

For example, in a study of hotel reservation cancellation patterns, it can easily be seen that cancellations vary between different market segments and distribution channels. In the image below, you can see that almost half of reservations made by travel agencies or tour operators, online or offline , end up being cancelled, while direct reservations have a much lower cancellation rate.

‍

‍

With this brief presentation and demonstration, I hope to contribute to the necessary awakening to the use of data visualization in your analysis, especially for those who do not consider this an essential analysis tool. Therefore, if you do any type of analysis work, please use and abuse the representation of data to better explore it and not just to communicate it.

‍

‍