How to choose the right chart for data and dashboard

“Which chart should I choose for my dashboard?” This is a common question of many data analysts when first practicing building dashboards. One data source can be represented in many different charts, which can make you confused. In this article, I will guide you through choosing the right chart for your data when building a dashboard and introduce common chart types so you know when to use them.

How to choose the right chart

In my opinion, to choose the right chart for data to add to the dashboard, we need to consider three factors: the chart’s purpose, the audience viewing the dashboard, and the limitations of the dashboard creation platform. If a chart meets these three elements, it is the right chart.

Purpose of the chart: Each should answer at least one question, such as “How are the company’s profits growing?” Choosing a chart with a Time series will be appropriate because it shows profit growth over time.
Audience viewing the dashboard: Although it will have many charts to represent, it will have the same purpose. Choose a chart so that the audience viewing the dashboard can immediately understand its meaning. For example, “What is the age distribution of customers?” Using a Boxplot chart is correct, but it will be difficult for people without expertise in charts to understand, so a Bar or Histogram chart will help them understand more easily.
Limitations of the dashboard creation platform: This is an obstacle that dashboard creators find difficult to avoid. Even though you know there’s a chart perfect for answering a question, the company’s dashboard platform doesn’t have one. For example, there is no Sankey chart available on Tableau Desktop. You must follow dozens of steps on expert websites to create a chart like Sankey.

Types of charts on a dashboard

There are too many charts in data analysis and data science. I only list charts commonly found on dashboards and supported by dashboard-building platforms. I will explain the meaning of each chart so you know how to choose the right one for your data.

Scorecard chart

Choose the Scorecard chart to provide an easy-to-understand overview of a metric or project’s performance and progress against established goals.

Usage scenarios: Enterprise performance management, financial reporting, etc.
Suitable area: The top of a dashboard.

Gauge chart

Choose a Gauge chart to represent the progress of a KPI and compare it with the target at the most general level.

Usage scenarios: KPI performance tracking, financial management, system and network monitoring, health and fitness tracking, etc.
Suitable area: The top of a dashboard.

Line chart (Time series)

Choose a line chart to identify trends and fluctuations of indices over time easily. This chart is used in most dashboards.

Use cases: Analyzing trends over time, continuous performance monitoring, comparing data across multiple categories, financial analysis, web and application performance analysis, production and operations management, etc.
Suitable area: The head and body of a dashboard.
Biến thể: Area.

Biểu đồ Line (Time series) — Line chart (Time series)

Area chart

Choose an Area chart to represent the change of an index over time, like a Line chart, but still create the feeling of volume, like a Bar chart. Area charts, such as stacked bars, can also be used in an overlapping style.

Bar Chart

Choose a Bar chart to compare data between different categories. If we choose the X-axis as time, bar charts can also be used to see the change of an index over time.

Usage scenarios: Compare quantities across categories, represent data over time, compare performance, etc.
Good place: The body of a dashboard.
Variations: Countplot, Horizontal bar, Stacked bar.

Pie chart (Doughnut)

Choose a Pie chart (Doughnut) to represent the percentage of components as a whole, helping viewers easily recognize the data distribution. However, it should be noted that the Pie chart can become confusing and ineffective when the number of parts is too large, or the ratio between parts is too close.

Usage scenarios: Percentage representation, structure representation, classification representation, market share representation, proportional distribution representation, frequency representation, etc.
Suitable area: The body and base of a dashboard.

Biểu đồ Pie (Doughnut) — Pie Chart (Doughnut)

Map chart

Choose a Map chart to help users understand the relationship between data and geography, supporting decisions and planning in many fields.

Usage scenarios: Representing geographic distribution data, market analysis, management and operations, representing economic and social geography, etc.
Suitable area: The head and body of a dashboard.
Variations: Bubble map, Filled map, heat map, Line map, Connection map, Combo map, Geo chart.

Scatter charts

Choose a Scatter chart to analyze and understand relationships between variables in data. It allows for the degree of correlation between variables and helps make observations about relationships and trends in the data. Each point on the chart represents a pair of values of two variables.

Usage scenarios: Representing relationships, analyzing correlations, detecting trends and patterns, assessing impacts, analyzing noisy data, representing multivariate data, forecasting and forecasting, etc.
Good place: The body of a dashboard.
Variation: Bubble.

Bubble Chart

Choose a Bubble chart to represent complex relationships between three or more variables and help visualize diversity in the data. They provide a comprehensive view of relationships and trends in data and allow for intuitive categorization and analysis of data.

Usage scenarios: Three-variable representation, three-way correlation analysis, multi-dimensional data representation, etc.
Good place: The body of a dashboard.

Bullet Chart

Choose a Bullet chart to compare the performance or progress of certain indicators with set goals, helping users quickly recognize progress and evaluate the situation.

Use cases: Performance measurement, project management, training progress evaluation, financial management, etc.
Suitable area: The head and body of a dashboard.

Treemap chart

Select Chart Treemap to represent hierarchical data in rectangular format. In this chart, the large rectangles represent the entire data set and are divided into smaller sub-rectangles, representing parts of the data set. Each subrectangle’s size often reflects a variable’s value, such as sales, profit, or volume. Color is often used to differentiate the value of another variable.

Usage scenarios: Representing hierarchical data structures, representing data distribution, etc.
Suitable area: The head and body of a dashboard.

Sankey charts

Choose a Sankey chart to visualize the flow of elements from source to destination. This will provide a visual overview of the system’s distribution and transformation.

Usage scenarios: Analyze user or customer behavior in a specific process, analyze departmental budgets, identify points of loss or waste, etc.
Suitable area: The body and legs of a dashboard.

Waterfall Chart

Choose a Waterfall chart to track how individual changes contribute to the overall change of a metric from start to finish, making it easier for viewers to see the key factors affecting the end result. cùng.

Use case: Performance analytics helps identify factors contributing to the success or failure of financial analysis, etc.
Suitable area: The head and body of a dashboard.

Timeline chart

Choose a Timeline chart to present information in chronological order. This helps viewers easily see important milestones and understand the overall progress of an event or project. It provides a visual and easy-to-understand view of how events or phases develop.

Usage scenarios: History and events, project management, product development, legal proceedings, user journeys, technology development roadmaps, personal events.
Suitable area: The top of the dashboard.

Boxplot charts

Choose a Boxplot chart to provide a visual view of the distribution and variation of data. This will help you identify important characteristics and detect unusual values in the data set. A boxplot is a statistical tool that displays data distribution based on summary values such as medians, quartiles, and maximum values.

Usage scenarios: Analyze data distribution, detect outliers, compare groups, summarize statistics, analyze multivariate data and test hypotheses.
Suitable area: The body and base of a dashboard.

Candlestick charts

Candlestick charts are used for technical analysis and financial trading, such as stocks, to provide detailed and intuitive information about price movements and market sentiment.

Usage scenarios: Used for technical analysis, price fluctuation trends, market psychology,…
Suitable area: The head and body of a dashboard.

Biểu đồ Candlestick — Candlestick charts

Histogram chart

Choose a Histogram chart to represent the distribution of a continuous data set through vertical columns. Each column on the Histogram chart represents a range of values (bin), and the height of the column represents the number of observations that fall into that range.

Usage scenarios: Representing data distribution, comparing group distribution, detecting outliers, etc.
Suitable area: The body and base of a dashboard.

Heatmap chart

Choose a Chart Heatmap to represent data in a table or matrix through an array of colored cells. The color of the cell often reflects its value. Heatmaps are used to explore and understand the relationships, distribution, and data patterns within a data set.

Usage scenarios: Relationship representation, frequency display, classification representation, matrix display, time data representation, …
Suitable area: The body and base of a dashboard.

Gantt charts

Choose a Gantt chart to represent the project schedule and related tasks. This will help managers and work teams understand the timing and correlation between activities. It will also help make decisions and adjust project plans when necessary to ensure project schedules are adhered to.

Usage scenarios: Project management, production planning, personnel roadmap management, construction project management, event management, …
Suitable area: The head and body of a dashboard.

Table (Pivot table)

Choose a Table (Pivot table) to summarize, analyze, and compare data in many dimensions. Pivot tables can also view data distribution by creating a percentage pivot table.

Usage scenarios: Represent data more specifically according to dimensions, reinterpret another chart in more detail, etc.
Suitable area: The body and base of a dashboard.

Conclusion

I hope this article helps you choose the right chart for your dashboard. If you have questions about any chart, don’t hesitate to leave a comment below to interact with me.