You must have seen a lot of data visualization examples. But what do you mean by data visualization? A graph? A chart? Or a dashboard? Today, I collected everything you want to know about data visualization, including the definition, importance, basic types. Also, I will introduce the top 16 types of chart in data visualization and analyze their application scenarios to help you quickly select the type of chart. In the end, you will have an idea of how to design good data visualization.

1. What is Data Visualization?
2. Why is Data Visualization Important?
3. Five data Visualization Types Commonly Used
4. Choose the right data visualization types and design it
5. How to Design Good Data Visualization?

1. What is Data Visualization?

In short, data visualization is presenting structured or unstructured data graphically to present information hidden in the data directly to people.

But there is a catch:

It is not merely using data visualization tools to turn data into graphs. Instead, it is looking at the world from a data point of view. In other words, the object of data visualization is data, and what we really want is to use data as a tool, using visualization as a means to explore the world.

2. Why is Data Visualization Important?

Let’s start with a game.

Look:

The largest importance of data visualization is that it helps people understand data faster. Finding connections between mountains of information aren’t easy, but graphs and charts can transform the invisible information into the visible graph symbol, express it directly and clearly, help you discover key points quickly.

That’s not all…..

Other importance of data visualization

Researches show that people remember around 80 percent of things they see but only 20 percent of what they have read. And the brain can remember images a million times faster than abstract words.

Therefore, visualizing data can deepen people’s memory of information.

The ability to display big data is another importance of data visualization. For instance, the dashboards built by FineReport can integrate big data from different resources, reflect real-time data, and display it on a large screen. Therefore, people can build connections between big data from different departments and monitor business performance. It opens up new avenues for business. You might find something unexpected in big data through visualizing data.

3. Five Data Visualization Types Commonly Used and Examples

3.1. Data visualization by area & size

Differentiate the length, height or area of the same type of graphics (such as columns, rings, spiders, etc.) to clearly express the contrast between the index values corresponding to different indicators. This approach will allow viewers to see the data and the comparison between them at a glance. When making such data visualization graphics, mathematical formulas are used to express accurate scales. 

Examples：

a: Check-in for City Class

This Check-in City Class histogram clearly shows the proportion of students in different regions. From the above picture, we can strongly perceive the absolute proportion of students in Beijing, Guangzhou, and Shanghai at first glance.

b: Federal Budget Map

As shown in the above figure, in the US federal budget map, the flow of funds is clearly expressed in different currency flows, and the proportion of each amount.

c: Enterprise Ability Model Spider Diagram

As shown above, through the spider map, you can see that the company’s ability in profitability and risk control capability is nearly 100 points, which can be said to be outstanding.

3.2 Data visualization by color

It is a common method of data visualization design to express the strength and size of the index value by the depth of the color. The user can see at a glance which part of the indicator data value is more prominent.

Examples：

a: Click Heatmap

As shown below, via tracking mouse movements and creating the heat map from the mouse log, the user behavior is visualized by it.

b. Earthquake heatmap

The heat map below shows the seismic intensity of each place on the map and the distribution of the seismic intensity.

3.3 Data visualization by image

By using images and icons that have real meaning, can display data and charts more realistically, and can easily convey the meaning of the data. For example, the graph below shows the proportions of each with a male and female icon as a background. At a glance, you can identify the men or women 

3.4 Data visualization by the concept

By translating abstract indicator data into familiar, easy-to-perceive data, which is easier for users to understand the meaning of the graphics.

Examples：

a: What is Unstructured Data? Everyone knows that the iceberg suspended in the sea is just the tip of the iceberg. The iceberg below the sea is the vast majority of the iceberg. Explaining the amount of data of unstructured data and structural data and describing the characteristics of unstructured data through the form of conceptual transformation is very vivid and makes it easier to understand unknown and difficult concepts.

b: Infographics Infographics are the most extensive representation of concept visualization. If you are interested in making infographics, you can refer to 7 Data Visualization Tools to Create Infographics.

3.5 Data visualization by graphs or charts

When we design indicators and data, using graphics with corresponding actual meanings to combine presentations will make the data charts more vividly displayed, making it easier for users to understand the topics to be expressed by the charts. It is the most commonly used data visualization type.

4. Choose the right data visualization types and design it

I can’t emphasize this enough:

Before choosing the data visualization types, it is important to know your data, especially data relationships are shown below.

Now, I assume that you understand your data and have the right data visualization tool. Let’s move on, choosing the very best types of data visualization for your project, and designing it impressively.

1) Bar Chart

When to use bar charts?

Bar charts are mostly used for: Data changes over time, comparison of different data, and the relationship between parts and whole.

Vertical bar charts：best for showing chronological data

Stacked bar charts: can be used when comparing relationships between parts and whole, and can be applied to discrete data and continuous data.

Horizontal bar charts: you can use it when there are many data categories and the label is long

100% Stacked: use it when only paying attention to the partial-whole relationship while the total value of each group is not important

How to design it?

1.Use horizontal text labels: Do not use horizontal or vertical lines of text to make sure the labels are easy to read.

2. Column spacing should be appropriate：Column spacing should be 1/2 the width of the column.

3. The numerical value of the Y coordinate should start from 0: If set the origin of coordinates beyond zero, you can not express the whole value accurately.

4. Keep the color scheme consistent: It is better to use the same color. If you need to emphasize the data, you can use another striking color to highlight it.

5. Arrange the category properly: Sort by initial of the word.

2) Pie Chart

When to use pie charts?

Pie charts can easily express the relationship between parts and the whole, which is suitable for discrete data and continuous data. This approach is most attractive and understandable when the amount of data is small.

Pie chart:

Ring pie chart: you can put the most important element in the middle.

How to design it?

1.Position the slices correctly： there are two design ideas.

Plan A: Place the largest part at 12 o ‘clock in a clockwise direction. Then, place the second largest at 12 o ‘clock in an anticlockwise direction. Arrange the rest counterclockwise as shown above.

Plan B: Place the largest part at 12 o ‘clock in a clockwise direction. Arrange the rest clockwise as shown above.

2. Better no more than five categories：It is difficult to distinguish regions on the chart when the data percentage is too small. If there are too many categories, put the unimportant ones in ‘other’ category.

3. Don’t use multiple pie charts to show comparison relationship: Use bar charts rather than pie charts to compare data.

4. Make sure the percentages add up to 100%

3) Line Chart

When to use line charts?

Line charts are used to show time series relationships and persistent data. It’s a good indicator of trends, accumulations, decreases, and changes.

How to design it?

1.No more than four lines: 

2. Only use full lines: dotted lines are distracting.

3. Coordinate axes should include zero reference lines: Although the line chart does not need to start with a zero baseline, the chart should try to include it. If some small ranges are meaningful, you can shorten the ratio to highlight them.

4. Display text labels at the end of the line directly：

5. The height of the line should be similar to the scale of the chart: The maximum height of the line chart should be 2/3 of the Y-axis.

4) Area Chart

When to use area charts?

The area charts can show the time series relation of the data, and different from the line charts, the area charts can show the quantity clearly.

Stacked area chart:It is used to visualize the relationship between the part and the whole, and to show the contribution of the part to the total amount.

100% stacked: It is used to show the relationship between parts and whole. especially when the specific value of the whole quantity is not important.

How to design it?

1.Be readable: In stacked area charts, put the most variable data at the top, the least variable data at the bottom.

2. The Y-axis starts at 0：The data would be more precise.

3. Don’t display discrete data：Display the stable data like temperature rather than unstable data.

4. Don’t show more than 4 groups of data categories：Too many data categories will make charts difficult to read.

5. Use transparent colors flexibly: Try to make sure you don’t use overlap. If the overlap is unavoidable, you can use transparent colors

5) Scatter Plot

When to use scatter plots?

A scatter plot shows the relationship between two sets of variables. Correlation can be shown when data volume is large.

How to design it?

1.The Y-axis starts at 0: 

2. Include multiple groups of variables: Use size and color to add variable

3. Use trend line: The trendline can display the trend and correlation.

4. No more than two trendlines:

6) Bubble Chart

When to use bubble charts?

When you want to show comparisons and rankings.

Bubble scatter plot: to display additional variables

Bubble map: to visualize the regional data

How to design it?

1.Ensure the text label clear :

2. The bubble size should be appropriate:

3. Don’t use strange shapes

7) Heat Map

When to use heat maps?

Heat maps can display classified data, using a strong sense of color contrast to represent geographic areas or data lists.

How to design it?

1.Use simple map outline: Distinct outline is distracting.

2. Select the appropriate data range：The data range should be between 3 and 5 groups. The data that out of range is denoted by +/-.

3. The pattern should be simple :

4. Use appropriate color: Intense color will lead to vision burden. Use monochrome, and adjust the shades to distinguish the regions is better.

These are the entry-levl data visualization types. You can express your data correctly and meaningfuly with these basic charts.

However, when your data gets big, and data relationships get complicated, you need more advanced data visualization types. But keep in mind, the same thing is that you need to understand your data.

5. How to design good data visualization?

Now, you know how to choose the right type of data visualization.

But how to design good data visualization?

For beginners, you can follow the design guide step by step. But no matter you’re a beginner or an expert, there are some tips to keep in mind when using it.

Know your audience

Your audiences usually have different backgrounds. If your data visualization aimed at professional audiences, you can interpret the data in more appropriate ways and technical terms. On the other hand, the general audience may need a clearer interpretation of the same data.

It is also important to know the audience’s expectations of the data. What are the key points they want? You need to be clearly present in the data.

Understand your data

After knowing your audience, you need to ensure the correctness of your data and understand your data clearly.  If you do not understand your data, you cannot deliver your information to your audience

Storytelling with data

Your design should also convey a story, not the data itself but the information behind it. Using a story often means that the audience gets more insight from the data. It helps the audience deeply understand new information.

In fact, data visualization is a great storytelling tool because pictures tell a thousand stories and you should use it as your advantage. Telling a story through a data set is not difficult because you can use colors, fonts, and statements as part of your storytelling approach.

Keep it simple

There are many data visualization techniques, but it doesn’t mean you need to include too many different techniques for it. You need to keep your data visualization techniques simple. Firstly, compare the data visualization tool and choose one based on your needs. After that, pick the right data visualization type and the perfect color combination for your design.

Also, keep your visualization effects simple. Too many elements can actually corrupt the design and skew the data.

You need to remember:

The benefit of data visualization is to present large amounts of data intuitively. If your displays seem difficult, you’ll need to go back and see if you’re using the wrong data presentation or including too much verbatim information.  

You might also be interested in…

How Can Beginners Design Cool Data Visualizations?

Top 20 Best Data Visualization Software in 2020: Free and Commercial

5 Most Common Mistakes in Data Visualization