What is the best chart to compare two sets of data?

How to choose the right chart type.

Choosing the right chart to compare two datasets can feel overwhelming, but don't worry. The best chart for your needs depends on what you want to highlight and the nature of your data. This guide will walk you through common comparison scenarios, helping you choose the most effective visualisation.

Adobe Express can help you get started once you know what the right chart type is for you.

• Comparing categories (e.g., regions, products): If you want to show the magnitude of differences between categories, a bar chart is a great option. For showing proportions of a whole, a pie chart or donut chart might be suitable, but use them cautiously as they can be harder to read with many categories. If you need to visualise the rank of categories, consider a ranked bar chart.
• Comparing trends over time: A line chart is your best bet for visualising how data changes over time. If you have multiple time series, ensure the lines are clearly distinguishable. A stacked area chart can show the cumulative trend of multiple categories over time, but be mindful of potential readability issues if categories overlap significantly.
• Comparing distributions: To see how data is distributed across two groups, a density plot or histogram can be very effective. A box plot is useful for comparing the medians, quartiles, and outliers of two datasets. Violin plots combine aspects of box plots and density plots, offering a more detailed view of the distribution.
• Looking for relationships/correlation: A scatter plot is ideal for identifying correlations between two variables. Each point represents a data point, and the pattern of the points can reveal the strength and direction of the relationship. If you have many data points, consider using transparency or density contours to avoid overplotting.

1. Bar chart.

A bar chart is often the go-to choice for comparing data across categories, and for good reason. Its simplicity and clarity make it easy to understand and interpret. A bar chart displays data as rectangular bars, with the length of each bar corresponding to the value it represents. The bars are arranged along an axis, with each bar representing a specific category.

Use cases for bar charts when comparing two datasets:

• Comparing monthly revenue between two sales teams: A grouped bar chart can clearly show the revenue generated by each team for each month, allowing for easy comparison of performance.
• Evaluating differences in sales data across products: A bar chart can display the sales figures for each product, with separate bars for two different time periods (e.g., this year vs. last year) to highlight changes in sales performance.
• Analyzing website traffic sources before and after a website redesign: A bar chart can compare the number of visits from different traffic sources (e.g., organic search, social media, referrals) before and after the redesign to assess its impact on traffic acquisition.
• Comparing customer satisfaction scores for two different customer service approaches: A bar chart can display the average satisfaction scores for each approach, allowing you to quickly identify which approach leads to higher customer satisfaction.
• Evaluating the effectiveness of two different marketing campaigns across different demographics: A grouped bar chart can show the response rate for each campaign within each demographic group, helping you determine which campaign resonates best with different audiences.

2. Line charts.

Line charts are a popular choice for visualizing trends and changes in data over a continuous period, typically time. They excel at showcasing patterns, fluctuations, and overall direction.

What it looks like: A line chart displays data points connected by lines on a graph. The horizontal axis usually represents time, while the vertical axis represents the value being measured.

Use line charts to:

• Compare trends in sales between two products across months: A line chart can clearly show the sales trajectory of each product, highlighting periods of growth, decline, and relative performance.
• Track website visits vs. conversions over time: Plotting both metrics on the same line chart reveals the relationship between traffic and conversions, helping you identify potential areas for optimization.
• Analyze stock prices of two competing companies: A line chart allows you to visually compare the performance of the two stocks, identifying trends, correlations, and potential investment opportunities.
• Evaluate the effectiveness of two different marketing campaigns over time: A line chart can track key metrics like website traffic or lead generation for each campaign, allowing you to compare their performance and identify which campaign is more effective at different points in time.

3. Scatter plots.

Scatter plots are excellent for exploring the relationship between two continuous variables. They reveal patterns, clusters, and correlations that might not be apparent in other chart types.

What it looks like: Scatter plots display individual data points from two variables on an axis. Each point's position is determined by its values for the two variables, plotted against the horizontal (x) and vertical (y) axes.

Scatter plots are effective for:

• Comparing customer age vs. purchase amount: A scatter plot can reveal whether there's a relationship between a customer's age and how much they spend, potentially informing targeted marketing strategies.
• Visualizing the relationship between time spent on site and number of conversions: This can help determine if users who spend more time on your website are more likely to convert, suggesting areas for improving user engagement.
• Analyzing the correlation between marketing spend and sales revenue: A scatter plot can show whether increased marketing investment leads to higher sales, helping you optimize your marketing budget.
• Evaluating the relationship between employee training hours and performance ratings: This can help determine if investing in employee training leads to improved performance, justifying training programs.
• Comparing the price of a product vs. its sales volume: A scatter plot can reveal whether there's a relationship between price and sales, helping you determine the optimal pricing strategy.

Tips for comparing two sets of data: Mistakes to avoid.

Comparing two sets of data effectively requires careful attention to detail. While choosing the right chart is crucial, avoiding common mistakes is equally important to ensure you can organise things clearly, accurately and insightfully.

Here are some tips and mistakes to avoid when comparing two sets of data:

• Using 3D or over-animated charts: While visually appealing, 3D charts can distort the perception of data and make it difficult to accurately compare values. Similarly, excessive animation can distract from the core message. Stick to simple, 2D charts for clarity.
• Putting unrelated metrics on dual axes: Using dual axes to plot unrelated metrics on the same chart can create misleading correlations and confuse the audience. Ensure that the metrics you plot on the same chart have a logical relationship.
• Overcomplicating with too many variables: Adding too many variables to a single chart can make it cluttered and difficult to interpret. Focus on the key variables that are most relevant to your comparison and consider using multiple charts if necessary.
• Failing to label clearly or colour data consistently: Clear and consistent labelling is essential for understanding your charts. Use descriptive labels for axes, data points, and legends. Choose a colour palette that is visually appealing and easy to distinguish, and use colours consistently across all charts.
• Using misleading scales: Manipulating the scale of your axes can distort the perception of data and create a false impression. Always use a consistent and appropriate scale that accurately represents the range of your data.
• Ignoring context: Data should always be presented in context to provide meaningful insights. Consider including relevant background information, benchmarks, or comparisons to other datasets to help your audience understand the significance of your findings.
• Assuming correlation implies causation: Just because two variables are correlated doesn't mean that one causes the other. Be careful not to draw causal conclusions based solely on visual correlations.