Dashboard Overview is the presentation of data analysis results. It is used to convey effective data information and business insights to readers, making the business landscape clearer and more intuitive, helping readers identify key metrics, and accelerating decision-making.
Given the inherent value of dashboards, building a clear and usable data dashboard is crucial. A poorly designed, information-overloaded dashboard not only fails to convey useful information, but may also make the data even harder to understand than before.

This article will guide you step by step on how to build a "good-looking" data dashboard.
However, before that, let's discuss a question: what does "good-looking" mean? ❓ ❓ ❓

"Good-looking" can be understood from two perspectives. The first is easy to read, which means presenting a clear and accurate overview of metrics so that readers can understand the data at first glance without spending time and effort to interpret it. The second is visually appealing, that is, using a well-organized and harmonious page to attract readers' attention and keep them engaged with the dashboard as long as possible.

Therefore, the key to creating a successful dashboard mainly lies in the following three aspects:

1.1 Tell a good story #

A good data dashboard should be able to connect data to business and answer the viewer's questions. The viewer should be able to accurately and quickly receive and understand the business meaning of the data and gain insights into the current business situation.
Therefore, the first key to a successful dashboard is whether it has a good storyline.

1.2 Present convincing data #

A data dashboard usually contains ten or more charts, each of which includes a large amount of data. Deciding on the appropriate format for each chart and the unit for each data point are all important considerations when presenting convincing data.
Using a line chart to display sales revenue fluctuations over time makes the data more vivid, and presenting large sales revenue figures with unit suffixes is more intuitive than showing raw numbers. Therefore, arranging data properly is crucial for improving the readability of a data dashboard and enhancing the efficiency of information delivery.

1.3 Deliver a good design #

As the saying goes: "Clothes make the person, and a saddle makes the horse." To make a dashboard stand out, good design is of utmost importance!
In data visualization projects, design principles are highly valued. With good design, a data dashboard can skillfully use layout and typography to highlight the core storyline, quickly convey effective business information to readers, and bring data and business insights to life.

Suppose you are an analyst for an education product. The customer success manager contacts you today, hoping you can create a dashboard to help them understand the key metrics of all online courses, in order to assess current user satisfaction and identify potential issues.
After receiving this task, how should you design this dashboard?
The reader's role + the information the reader wants to see = the story the dashboard needs to tell
Specifically, this can be divided into the following two steps.

2.1 Identify the dashboard reader's role (who to present) #

Users in different roles have different requirements for data dashboards. For example, corporate executives focus more on key metrics, the gap between targets and current status, and the main business risks. Business leaders, on the other hand, pay more attention to profit growth, the sales funnel, abnormal business performance, and hope to monitor the core KPIs of department employees.
Different roles have different data needs within the dashboard. Only by clarifying who the target audience is can the data content of the dashboard be further determined.

2.2 Determine the data content to present (what to present) #

After clarifying the roles of readers, we can identify data requirements through research or even interviews (for example: which metrics are needed for which dimensions?). This work is important not only at the initial stage of dashboard construction, but also during dashboard maintenance. Understanding what data everyone wants to see, and continuously updating the dashboard as reader needs change, is crucial.

2.3 Refining deep data details (reader stratification) #

Each dashboard may have multiple readers, and different users have varying requirements for data granularity. To support readers in more flexible data exploration, present information at different levels, and enable users to independently extract deeper insights, dashboard creators should appropriately embed interactions such as drill up and drill down, multi-table linkage, chart navigation, and intelligent attribution to enhance the reader experience.

To learn more about interactive features, you can click to view
Drill Up and Drill Down
Chart Linkage
Chart Redirect
Intelligent Attribution

Now you have determined the charts within the dashboard, but the dashboard still appears dull, and the data in the charts remains difficult to interpret.
Therefore, you may find that the issue lies in chart selection and data presentation. But how can you make the business significance represented by the data vivid and three-dimensional, leaping off the page?

3.1 Choose the right chart type #

As the saying goes, "A picture is worth a thousand words." The presentation of a data analysis dashboard relies on charts as the medium, and choosing the right chart enables you to convey accurate information to readers more quickly. However, BI products offer a wide variety of chart types, each suited to different scenarios. How should we select charts to display our data?
Typically, based on the structure of the data to be displayed, charts can be categorized into four main types: comparison, relationship, composition, and distribution/geography. On this basis, taking data variables into account, determine which chart type is most appropriate.

Here we have prepared a list of chart types supported by the product, along with examples of their common application scenarios, for your reference.

3.2 Making effective use of data configuration #

Selecting the appropriate chart type does not mean the work is complete. Chart titles, units, legends, and labels all greatly assist users in interpreting data, and the data format in the plotting area, in particular, contains many subtleties. The product provides the following data configuration options. Making flexible use of these features is very helpful for accurately presenting data characteristics and highlighting data hierarchy.
Data format
Data configuration within a chart is a key factor affecting how readers understand chart information. Dashboard configurators should take the following aspects into account when configuring data to make it more intuitive and prominent.

• Field name: For field names that are easily confused, potentially unclear, or system-generated derived fields, it is recommended to set aliases or field descriptions to improve readability and reduce dashboard maintenance costs.
  For example, change 'date' to 'shipment date' to clarify the action associated with the date; change the system-generated field display name 'sales amount difference percentage-1M' to 'sales amount month-over-month', making the name more consistent with readers' expectations.
  For the detailed user manual on alias management, see: Set Field Value Aliases.
• Field value: Setting the field format is equally critical. For example, '500,000,000' is clearly easier to understand and less prone to misreading than '5,000,123,345'. In chart design, units, formats, and prefixes or suffixes should be set according to the business meaning of the data itself. For example, for sales amount data, add the unit 'CNY' or 'USD'; for growth rates, proportions, and similar data, set the data format to percentage.

Conditional formatting
Icon sets, color scales, and data bars are common types of conditional formatting that can be used to enhance the visual effect of the data. Using conditional formatting makes it easier to compare metric values within a specified area simultaneously.

• Icon sets: Icon sets provide a more intuitive way to categorize and display data. Different icons are displayed based on threshold ranges, allowing readers to quickly group data by size.
• Color scales: Color scales help readers better understand data distribution. After setting color scales, values will be displayed with gradient colors from highest to lowest. For scenarios with outliers or where customized color scales are needed, the maximum and minimum values can also be configured manually.
• Data bars: Data bars help readers compare metric values and identify larger or smaller values. The longer the data bar, the larger the value; the shorter the data bar, the smaller the value.

Analytical features:

• Sorting and TopN settings: Specify which data to sort by default. Use the sorting component to highlight data that is of high interest to readers. The sorting feature provided by the product supports sorting the horizontal axis, vertical axis, and dimension fields in charts.
• Statistical item settings (such as totals or percentages): Specify which data to aggregate further. Display statistical items to compare detailed data with aggregated data, helping readers understand data characteristics and gain business insights.
• Reference line settings: To make data charts clearer and easier to understand, it is often necessary to add reference lines such as target lines, control lines, or average lines to assist with analysis.

Table styles
Tables are popular among report creators because they clearly present a large number of dimensions and metric summaries. However, due to the large amount of data in tables, optimizing table configuration to highlight key information while displaying large volumes of data is crucial. This section briefly introduces several tips for designing table styles.

• There must be sufficient contrast between the background color and the content of a cell: it is recommended to use text colors that stand out against the dashboard color. To assist users in reading, zebra striping can be used to improve browsing speed.
• Avoid using dark or thick borders to prevent the borders from overpowering the content.
• Highlight key data, which can be combined with conditional formatting features.

3.3 More interactive configurations #

How can the dashboard be made more dynamic to meet the needs of different users for viewing data at different levels of granularity? Make good use of both filters and dynamic dimension metrics to enhance the analytical flexibility for dashboard viewers.
Filters
The filtering feature helps readers quickly filter charts and view the required data. For example, in a retail sales data dashboard, store managers in different cities may want to quickly focus on their own region or even their own store to view the performance data for that part of the business.
When configuring filters for the dashboard, pay special attention to the following points:

• Filter item settings: that is, which dimensions need to be added as filters and from which perspectives readers typically want to query the data. Setting filter items requires a high level of business understanding. If you are unsure, consider consulting several business stakeholders, as they may have the answers.
• Default query time: set the default query time for the dashboard so that viewers see the latest data as soon as they access it.
• Other edge cases to consider: removing extreme data values, limiting the time range, handling cases where the data is empty, and more. Such data needs to be addressed or transformed according to the business context.
• Learn more about filters Dimension Filter, Metric Filter, and Filtering.

Dynamic dimension metrics
Want to analyze the same metric freely from multiple dimensions? Want to see the performance of multiple metrics under a single dimension? Different users may care about different metrics, and creating a separate chart for each is inefficient and takes up space.
At this point, you can try the dynamic dimension/dynamic metric feature, using filters to switch analysis dimensions and metrics to meet the analytical needs of different viewers for different dimensions and metrics.
For more information about dynamic dimensions and metricsBest Practices for Building Dashboards,Filtering

In the field of design, there is a concept called cognitive load, which refers to the cost required to effectively convey information within a work to the reader. An excellent visualization work can quickly eliminate redundant chart information and deliver sufficient information to the reader's brain through visual channels in a short time, thus completing data transmission.
This section will focus on common optimization approaches for data analysis, including design, layout, and chart color schemes.

4.1 Basic design principles #

(1) Contrast:
Highlight differences through contrast.
Common types of contrast include shape contrast, size contrast, and color contrast.

(2) Repetition:
Achieve consistency among dashboard components through repetition and enhance visual effects.
Common repetitions include repeated icons within a page, repeated charts within a page, and repeated structures across pages.

(3) Alignment:
Alignment ensures that elements are unified, connected, and have clearer relationships with each other.
Common alignment methods include left alignment, center alignment, and right alignment. It is generally not recommended to mix multiple alignment methods on the same page.

(4) Proximity:
By grouping chart components, placing similar elements close together, and separating different elements, users are visually guided to perceive differences in proximity between elements.
By applying the principle of proximity, simply grouping elements of the same hierarchy and separating them using color, background, or configuration will result in automatic visual grouping.

4.2 Common layout designs #

As a specific technique in graphic design, layout essentially deals with the relationships among elements on the canvas.

Elements on the canvas include composition, color, font, and so on.
Relationship refers to the distance between elements, their size and area, color combinations, and so on.

A successful layout design fully leverages design principles, balances the relationships between elements, reduces the reader's cognitive load, and enhances the readability of the work. The following two layout types are commonly used in data analysis dashboards.
(1)Container layout method
Group related content elements together, add a "container" beneath the elements, and adjust the formatting.
The container layout method first requires clarifying the correlation between elements, then, following the principle of proximity and separation, places related elements together and keeps unrelated elements apart. By segmenting and separating the layout, the page becomes more intuitive and clear. Dividing the page into sections with a clear structure makes the entire layout appear flexible and elegant.

When editing a dashboard, you can use a text component and lower the background color's opacity to achieve the mask effect shown above.

(2)Split-screen layout method
Use color blocks to divide the background into N sub-screens, each of which acts as a container and can carry a type of information.
Split-screen layouts serve two purposes:

(1) They modularize information expression in form, reducing the audience's cognitive load.
(2) They significantly increase the visual density of the layout and enhance the visual effect of the page.

Split-screen layouts are often combined with container layouts, using background color blocks to achieve the split-screen effect. Additionally, users can achieve split screens through dividing lines, color contrast, and product features such as tab pages or sheet pages.

Tab pages allow for more layered collapsing and previewing of data. For details, click Operations in the Multi-Page Editing Area at the Bottom of the Dashboard.

4.3 Icons and colors #

Whether a dashboard looks good and high-end is often determined right here!
Exquisite illustrations make dashboards more refined, and outstanding color schemes make dashboards more three-dimensional and rich.
Icons
A dashboard typically contains various types of charts, a large amount of numerical and textual information, and the sheer volume of information can easily cause customers to lose focus. Brightly colored icons serve as the finishing touch that draws customers' attention. Proper use of icons can significantly reduce users' cognitive load and improve reading efficiency.
For analysts who pursue excellence, icons are also essential to highlight the dashboard's strengths in subtle ways. Try adding your company's logo and your department's slogan to the template to create a unique and impressive dashboard!
For more resources, you can click this link to view.
Color
The colors of the charts should be closely related to the overall color scheme. When the dashboard uses a dark background, charts with high color saturation will stand out even more. In addition, many charts can use gradients in the details to make the visuals richer, and text shadow details can also enhance the overall visual quality.
A coordinated and well-chosen color scheme can help the dashboard tell its story more effectively. For example, when analyzing year-over-year and month-over-month changes in sales revenue, you can use a consistent red and green color scheme to represent increases and decreases in sales revenue, respectively.
Basic knowledge of color schemes:


General color scheme:
Primary color + secondary color + accent color Primary color (usually accounts for 75%)

• Primary color (usually accounts for 50%)
  • The main color of the entire work, which affects the overall visual impression of the dashboard;
• Secondary color (usually accounts for 20%)
  • Helps the primary color establish a more complete image, making the dashboard harmonious or lively;
• Accent color (usually accounts for 5%)
  • Guides and highlights specific content, emphasizing key points.

Color scheme steps:
Determine the color scheme based on the content; select the primary, secondary, and accent colors according to the theme; try not to use more than four colors.

• Primary color presentation methods:
  • Large area with high purity; although the area may be slightly smaller, it is the most prominent color visible at a glance in the visual; dual primary colors are also possible.
• Use of secondary colors:
  • You can use analogous colors of the primary color to make the visuals unified and harmonious. You can also use a contrasting color to the primary color to make the visual more lively and stable.
• Characteristics of accent colors:
  • Frequent appearance; vibrant colors; strong contrast with other colors, which can highlight key points.

For an excellent article on color matching, click to refer.

Attractive appearances are common, but interesting souls are rare. The creation of dashboards for data analysis is similar; telling a good story is the core priority. Therefore, after creating a dashboard, remember to collect feedback from frontline readers and iterate continuously.
To quickly learn how to build a clear and visually appealing dashboard, join our learning course and make progress together.
60-minute quick start to intelligent data insight products

• 30-day free trial opportunity, click to activate, Free Trial.
• Complete the course to receive a ByteDance-exclusive certificate and a premium gift package
• Systematic knowledge point review, learn to get started with BI tools from scratch
• Detailed instructional guidance, hands-on teaching to help you complete the following business sales analysis report

Registration address: Volcano Engine Intelligent Platform (VeDI) Learning Center.