Data Visualization Series — 2. How to choose the best visualization
Introduction
The first post in this series explored the fundamentals of data visualization, highlighting how effective visual representation can transform complex data into understandable and accessible insights, and the importance of DataViz for companies, how it can improve decision-making and what the benefits of implementing effective visualization practices in business are. Now, the text delves into a crucial aspect that can determine the success or failure of visualizations: choosing the right type of graphic.
Choosing the right visualization is not just a matter of aesthetic preference; it’s a strategic decision that can highlight or obscure the most important information in your data. The second post in the Data Visualization Series will explore the different types of charts available and provide practical guidelines for selecting the most appropriate visualization according to the context and target audience of your analyses.
Choosing the Right Chart Type
Choosing the right graph is key to conveying your analysis effectively. Different types of graphs can highlight different aspects of the data, depending on what you want to communicate and who your audience will be. In this section we will explore the most common variations of graphs and provide guidelines for choosing the most appropriate format for your visualization needs.
1. Knowing the Different Types of Charts
Line Graphs: These are the ideal choice for showing trends and developments over time. They allow the public to clearly visualize how variables change, making it easier to identify patterns or anomalies. For example, a line graph can be used to show month-on-month sales growth or temperature fluctuations over the course of a year.
Bar and Column Graphs: These are used for numerical comparisons between different categories. Bar graphs are particularly useful for showing differences in size or quantity between items, such as sales between different products or expense categories in a budget. Column graphs, on the other hand, are effective for quick and clear comparisons.
Scatter Plots: Excellent for identifying relationships or correlations between numerical variables. They show how one variable behaves in relation to another, and can reveal hidden connections or trends. For example, a scatter plot can be used to explore the relationship between the amount of money spent on advertising and the resulting increase in sales.
Area Graphs: Similar to line graphs, area charts are used to demonstrate how a quantity develops over time, with the addition of fill under the line, highlighting the accumulated volume. They are particularly useful for visualizing the contribution of different components to the whole, such as the growth of different sectors of a company over time. This type of graph is effective for showing the composition and changes in aggregate data, where the area under the curve can help illustrate the magnitude of a trend in a visually impactful way.
Pie Charts: Used to show proportions and percentages between categories, they are ideal when you want to highlight parts of a whole. A pie chart can, for example, show the distribution of expenses in different departments of a company, making it easier to see which part consumes the most resources. However, it is important to note that correctly assigning quantitative values to areas is not a natural skill, which means that it is difficult to read pie charts when the segments are similar in size or when there are many segments. Therefore, this type of visualization is only recommended when there are up to six segments and its application should be combined with the use of data labels.
Combined Charts: Combine two or more charts types into a single view, which is useful for comparing different scales and types of data simultaneously. For example, a combined chart might use columns to represent monthly sales and a line to show the growth trend over the year.
Summary of chart types
Below is a summary table of the graphs mentioned, evaluating each one in terms of purpose, reading complexity, type of variable and practical application examples, the aim of which is to provide a quick and efficient reference to facilitate the choice of the ideal visualization based on the specific context of use.
However, it is important to note that it is often not necessary to use a graph to convey information. If the purpose of the data is well defined, you can display the information as a highlighted number on your dashboard or even as a sentence on a slide.
2. Deciding which chart to use
As seen in the first post in the Data Visualization Series, an effective graphic should communicate your data clearly, based on the objective of your analysis, the target audience that will consume the information, the complexity of the data involved and the format in which that data will be consumed. These considerations will help ensure that the graph you choose not only conveys the desired message effectively, but is also understood unambiguously by your audience.
Analysis objective: Think about the main message you want to convey. Which business tactics or strategies will your visualization guide? What business questions do you want to answer?
Target audience: Adapt the graphic to the audience’s level of familiarity with interpreting visual data. Technical audiences may appreciate more complex formats, while a broader audience may prefer simpler, more straightforward formats.
Data Complexity: Assess how detailed the data is and what you need to highlight. Simple data can be effectively represented by bar graphs, while more complex data sets may require scatter plots or multiple graphs for adequate representation.
Graph consumption: Take into account the control you have over how the information is consumed. The method you choose to present your data can have a significant impact on how it is interpreted and the level of detail that needs to be explained. Ask yourself: Will the charts be part of projected reports, consumed periodically? Will they be intended for exploratory use, allowing dynamic interactions, or will it be a more static format?
User testing: Before finalizing a choice, prioritize testing different types of graphics with a representative group of your target audience. Direct feedback can reveal problems of understanding or preferences that are not immediately obvious.
Conclusion
By selecting the most appropriate chart type, you maximize the effectiveness of your visual communication and ensure that the data not only informs, but also engages your audience. In the next post in this series, we’ll cover how to adapt data visualizations to be inclusive, focusing on adaptations for people with color blindness, ensuring that visualizations are accessible to all audiences. This step is essential for creating a truly inclusive and responsible data culture.
Profile of the Author: Mariana Brandão | With a degree in Bioprocess Engineering and Biotechnology from UFPR and a passion for numbers, I seek to transform the way decisions are made through the power of data. I’m currently a Business Analytics Consultant at DP6.
Profile of the Author: Andressa Viana | With a degree in International Relations from UFF, I work professionally as a Business Analytics Consultant at DP6. I believe that data visualization plays a key role in understanding the world and the times in which we live.
Profile of the Author: Luana Moura | I have a degree in Library Science from UDESC and a Master’s in Information Science from UFSC. I’m dedicated to collecting, analyzing, and structuring data to enable smarter decisions. I’m currently a Business Analytics Consultant at DP6.
