The Building Blocks of a High-Performing Dashboard: Best Practices for Effective Data Visualization

Hello! In this article we will take you on a journey to build your own high-performing and user-friendly dashboard, with a step-by-step process of improving our own. Sounds interesting? I certainly hope so 🙂

After digging through various articles and blogs online, we felt that there was no one “go-to” resource on the topic, and we decided to create our own, both for usage within our team, as well as for our dear readers, like yourself.

In this article, we will explore key principles of creating a good dashboard including best practices and common pitfalls with a focus on user experience. From understanding the psychology of color, data visualization techniques to creating a clear navigation structure, interactive elements and user-centered design principles, we will cover (almost) everything you need to know to create a dashboard that is clear, relevant, and last but not least, actionable.

We think this is useful whether you’re a data scientist, business intelligence analyst, company owner, marketer or general business user.

Introducing: our baseline dashboard. Let’s see what it takes to improve this

Before you start: Three Essential Considerations!

Get to know your audience

One of the most important factors to think about when building a dashboard is your audience. It’s essential to understand your audience’s needs, expectations, and technical abilities. If the dashboard is too complicated, people will spend more time trying to figure it out than they would without it. So, you want to make sure everything they need is on the charts, and they don’t need to calculate anything themselves. We cannot stress enough the importance of putting yourself in your audience’s shoes.

Understand the nature of each chart type

Choosing the right type of chart to present your data is super important. If you pick the wrong one, people might have a hard time understanding your information, which can lead to mistakes. Each chart has a different purpose, so it’s essential to understand what you want to convey with your data. For example:

• Bar charts are ideal for comparing data
• Line charts are useful for showing trends over time
• Scatterplots are effective in illustrating relationships between two variables
• Pie charts are best for displaying proportions and percentages

It’s important to remember to pick the right chart type and present your data in a clear and simple way so that everyone can understand it. You can even categorize the purpose into four groups: comparison, composition, distribution, and relationship. When you consider these two things carefully, you’re well on your way to designing a great dashboard. Now that these overall considerations are clear, let’s move to our Ground Rules

1. Ground Rules

1.1 Make sure your positive and negative values are plotted in the right direction

When creating charts or dashboards, it’s essential to use the correct plotting directions based on positive and negative values. Doing so ensures that the visual cues in the chart match the actual values of the data. Failure to do so can result in confusion and misinterpretation for viewers who expect the chart to match their expectations based on the values they’re familiar with.

To avoid this, when using horizontal bars, it’s crucial to plot negative values on the left side and positive values on the right side of the baseline. This placement aligns with our eyes’ natural reading direction from left to right. If negative and positive values are plotted on the same side, it can be difficult for viewers to interpret the data accurately.

1.2 Always start a (bar) chart at 0 baseline

Starting a bar chart at a baseline of zero will provide a consistent and precise scale. This helps to avoid misinterpretation and present the data in an honest way. Starting the chart at a different baseline can exaggerate the differences in the data and lead to inaccurate conclusions. For example, if you’re comparing sales of different products, starting the chart at 50 instead of zero can make the differences between sales look bigger than they really are.

However, for line charts, limiting the y-axis scale to start at zero can flatten the chart. To represent the trend accurately, it’s important to adapt the scale based on the data set and keep the line occupying two-thirds of the y-axis range. So, if you start a chart at a number other than zero, don’t forget to label the chart clearly so that people can understand what the chart is showing

1.3 Legibility > Aesthetics

It is important to keep in mind that the primary objective of a dashboard is to communicate data and insights to its users effectively. If the information is not presented in a clear and easily readable format, the dashboard will not be able to fulfill its intended purpose.

While aesthetics are important, they should not take priority over legibility in dashboard design. A well-designed dashboard should strike a balance between aesthetics and legibility, ensuring that the user is able to effectively understand and use the information presented.

1.4 Make sure your labels and data formatting are consistent throughout your chart

Make sure that all your labels have the same font, size, and style to create a visually pleasing and cohesive look. If you don’t use consistent formatting, your audience might get distracted by the differences and miss out on the important information you’re trying to convey. So, By keeping your dashboard consistently formatted, your audience can focus on the data presented and make more informed decisions. This can save time and effort, and ultimately lead to a more positive user experience.

2. Color

Color plays a crucial role in effective dashboard design, as it has the power to impact the user’s experience, understanding, and engagement. A well-designed color palette can enhance the aesthetics of the dashboard, draw attention to important information, and convey meaning through the use of contrasting colors.

However, using too much color or too much contrast between colors will cause the audience to miss the point, so if you’re ready, let’s see how color works in the dashboard.

2.1 Choose a color palette that matches with your data type

As you know that color is an integral part of effective data visualization. To design effectively, it’s helpful to categorize the color palette used in map visualizations into three types :

• Qualitative color palette : good for showing categorical variables (not a number) that are not related to each other such as Country in Europe, Biggest export category in each country, Land use data etc, often used with colors that stand out from each other.
• Sequential color palette : good for showing numerical variables that have an order, and is created using variations in hue, lightness, or both. A good example is the World population or Number of product sales in each country.
• Divergent color palette : (or a performance color palette) a combination of 2 sequential palettes centered around a value (often 0) to show positive/negative values. For instance, Global temperature or Change in U.S. population change.

2.2 Design with accessibility in mind

Just relying on different colors alone isn’t enough for the colorblind person. Our suggestion is to add some variation in saturation and luminance, like in the picture below. For everyone to more easily see the changes, a small adjustment of saturation can make a big difference.

Another suggestion is changing your data visualization into black and white to check the contrast and readability. You can click here to view a color palette guide and see how each hue would appear to a colorblind person.

2.3 Don’t get too crazy with the colors, choose a few and stick to them

Consistency is key, so it’s best to stick with your company’s brand colors or choose a few universally recognized colors like green or red. It’s also important to differentiate each color using gradients to avoid confusion. Avoid using too many different colors or high-saturation hues as this can overwhelm and confuse your audience.

It’s important to find a balance between visual appeal and clarity to ensure that your audience can understand and interpret the data effectively. So, while it’s essential to keep it simple, it’s also important to tailor your color choices to the data and visualization type for optimal results.

If we added adjustments to our dashboard using the ground rules and color principles, here is our Dashboard v2.0! …

3. Text & Label

In a dashboard, text and labels are important components that help viewers understand and interpret the data presented. These components can take many forms, including titles, subtitles, headings, descriptions, captions, and annotations. Their primary role is to provide information about the data source, date range, units of measurement, definitions of terms, and any assumptions or limitations of the data. Without this information, viewers may not fully understand the data and how it was collected or analyzed.

In addition, text and labels can enhance the visual design of a dashboard. Using appropriate font sizes, colors, and styles can help draw attention to important information and create a visually appealing dashboard. Keep in mind that it’s important to use these design elements sparingly and consistently to avoid overwhelming or confusing viewers. So, let’s start exploring more about text and labels!

3.1 Use a horizontal bar chart instead of rotating the labels

Rotating labels may seem like a good idea when you have limited space, but it can lead to readability issues. For example, if the labels are too long, they may become too small or unreadable when rotated. Using a horizontal bar chart allows you to present more text while ensuring that the labels remain easy to read, especially when the data labels are longer.

3.2 Label directly on the chart

Putting labels directly on the chart can avoid confusion by making it clear which data point the label is describing. This method eliminates the need for a separate legend and color which means users don’t have to search for information in a different location. This technique helps to improve the overall appearance and clarity of the dashboard.

However, each data visualization tool has various capabilities; some of the tools we’ve been using at our agency, such as Looker Studio (Google Data Studio), cannot do this. Click on this link to view an article comparing several visualization tools for your decision.

3.3 Avoid confusing dual axis

Using dual axis charts may seem like a good way to save space on your dashboard, they can often be confusing and difficult to interpret when the scales and units are different. It’s better to avoid using them, or simplify them as much as possible, to ensure that your audience can understand the data you are presenting. If you must use a dual axis chart, it’s important to label both axes clearly and provide clear explanations to help your audience make sense of the data.

4. Navigation

4.1 Content grouping

Grouping related content together makes it easier for the user to find the information they need. Consider grouping related charts, tables, and other data elements together, and using clear headings to identify each section.

In addition, you can also use visual cues such as adding background colors or borders to separate the groups of content. This makes it easier for the user to quickly identify the different sections and find the information that users are looking for.

4.2 Don’t overlook spacing

Proper spacing is essential for making a dashboard easy to navigate and understand. Make sure there is enough space between each chart, table, or other data element, and use clear headings and labels to help guide the user. Don’t put too much stuff on the dashboard or it might be hard for people to find what they need.

4.3 Applying the Z-Pattern and Fitts’ Law

The time it takes to reach a target depends on how far away the target is and how big it is. This is called Fitts’ Law, and it means that targets that are closer and larger are easier to reach. To make it easier for users to find important information on a dashboard or webpage, it’s recommended to organize objects in a Z-pattern, which follows the natural eye movement of users from left to right and top to bottom.

5. Interactivity

5.1 Add dashboard filtering and sorting capabilities for a more user-friendly experience

Dashboard filtering and sorting capabilities allow users to easily navigate and explore the data presented in a dashboard. This feature allows users to focus on specific data points or segments of the dashboard, making it more user-friendly and intuitive. For example, if you have a dashboard showing sales data by region, you could add a filter that allows users to view the data for a specific region, such as the West Coast.

By including filtering and sorting capabilities, you can enhance the interactivity of your dashboard and enable users to easily find the insights they need. This feature also allows users to interact with the data in a way that suits their needs, making your dashboard more flexible and customizable.