With the increase in data that businesses generate and consume and the advent of cloud-based analytics services, an ever-increasing number of professionals seem to be warming up to the idea of learning the basic techniques of data visualization.

Contrary to popular belief, data visualization as a practice is not novel. The use of graphs, pictures, or diagrams to represent data had started as early as the 10th century AD. The art of data visualization has developed consistently ever since across domains. In what is considered the first-ever visualization using statistical data, astronomer Michael Florent van Lagren tried to determine the distance between Toledo and Rome in 1644 using longitudes [1].

the distance between Toledo and Rome in 1644 using longitudes

With this history, we wanted to emphasize that data visualization has been around for ages. Many of us still believe that it is a specialization, but the technological advancements in recent years have made visual representations of data more accessible than ever.

The numerous commercial tools available currently make learning and implementing data visualization even more exciting and effortless. With a little desire and time, one can get reasonably comfortable with any of the available visualization tools. Having learned their ways around the new tool, what still bothers learners is the choice of appropriate visual or chart.

Choosing the right type of visual to represent your data is an essential step in data visualization. An inappropriate visual will result in your message and hard-work getting lost in translation. Through this blog, we aim to guide you in choosing the right visual based on the type of data and message you want to convey. For structure, we have classified various visuals in 3 different categories based on the ease and frequency of their usage.

Simple, Medium and Complex Chart

Here we’ve talked about a majority of simple visuals and highlighted the different scenarios when they should preferably be used.

Bar Charts

Bar charts represent categorical data when the aim is to see how each of the categories has performed relative to the others. Typical examples of bar charts would include sales of different product-categories for a retail chain or revenue figures of different geographies.

Bar charts are one of the most common types of visuals used by professionals. Anyone having worked on Excel would undoubtedly have worked on bar charts. Following is a sample bar chart comparing different countries by the number of units of an item sold in each of them.

Simple Bar Chart

Clustered Column Charts

Clustered column charts are like bar charts only in terms of look and feel. They differ in the way that one can represent the figures and numbers regarding more than one dimension.

Using a clustered column chart, we can compare the units sold for different countries and across years. To summarise, one should use a clustered column chart when there’s a need to analyze figures or numbers across more than one parameter.

Line Charts & Area Charts

Line charts are a graphical representation of a series of data points connected by a straight line and are used to represent continuous data sets. Line charts are ideal for displaying trends over periods and for listing the exact value of measures at each point during the period.

E.g., the sample chart below shows the count of open and closed tickets over a period

Count of Open and Closed Tickets

Area charts are like line charts, but the area between the X-axis and the line is filled with some color, pattern, or texture. Areas charts can also be used to show trends over time. However, they can be more useful in representing measures or figures that have a part-whole relationship with each other. A good example would be a representation of invoice and receipt amounts using an area chart. In such a visual, the relation between the two values becomes apparent just by looking at the chart.

Area Charts

Doughnut charts

Doughnut charts are like the ubiquitous pie-charts and can be used to show how a figure is composed of various parts. However, one significant advantage of a doughnut chart is that given their design, they can be used to include an extra dimension for analysis.

The example below shows how each region contributes to total sales. Please note how the trend of sales can be presented across the years as well.

Gauge Charts

Gauge charts show the progress towards a goal; i.e., it showcases the level of goals accomplished. Gauge charts represent KPIs when you know the maximum and minimum values, and there is a target to achieve, such as the yearly sales target of a company.

The figure below is an example.

Gauge Charts

In this blog, we have covered some of the standard graphs and visuals to get you started. These can be created in any visualization tool currently available in the market. There are several other types of graphs/charts that we will be covering in subsequent blogs in this series. Till then, we hope that you will try using their visuals in whichever visualization tool you prefer.

References:

  1. E.R. Tufte, Visual Explanations 1997
  2. https://ux.stackexchange.com/questions/105837/pie-chart-vs-doughnut-chart-when-to-use-each (last accessed on January 28, 2020)
1 Star2 Stars3 Stars4 Stars5 Stars (2 votes, average: 3.00 out of 5)
Loading...


Kamlesh Patil & Shivam Kumar

Posted by Kamlesh Patil & Shivam Kumar

Leave a reply

Your email address will not be published. Required fields are marked *