When is a dashboard a dashboard

Dashboard design rules Design principles for creating meaningful dashboards

Certainly every new dashboard to be created has to meet specific requirements. However, certain principles and best practices generally apply to all dashboard designs in dashboarding. Thanks to modern self-service BI tools, graphic designers or data scientists are no longer required to create your own data visualizations, but the most important basic knowledge of design should still be available. After all, the BI dashboard is at the top of every BI project and, with the visualization of the results, acts as an interface to the clients or decision-makers. The actual data queries and analyzes are completed, but only with an appealing visualization can the data analyzes be brought to a successful conclusion.

The dashboard design offers you many options to let your creativity run free. However, you should not make full use of all options, but rather observe the most important principles and best practices. This is the only way to present information that is easy to understand and directly usable. After all, the dashboard should be user-friendly and provide valuable assistance for the upcoming decision-making processes. In the following, we introduce you to the 11 most important principles and best practices for meaningful and results-oriented dashboards.

1. Always keep an eye on your goal and target group when dashboarding

At the beginning of any design, even before you select data and design elements, you need to be clear about the goal of the data visualization and your target audience. For this it is necessary that you put yourself in the role of your target group. These questions need to be clarified, for example:

Should the visualized data be presented to a larger audience?

Do the users look at the information individually on the office computer?

Is the information required "virtually on the go" on the mobile devices of your target group?

If the dashboard is presented, make sure that all important information fits on one page. For the display on the office computer or on mobile devices, responsive online presentations should be selected that take into account the different screen sizes and resolutions of the PC monitors, tablets or smartphones.

Each dashboard must always be designed for the specific target group. The intention is to optimally support this group in their decision-making process. The information is only useful if it can be used directly for the decision-making process. Identify the key information for your target audience and leave out the irrelevant. The type of business activity to be supported is also important. It is important to distinguish between the following four basic orientations:

Strategic purposes: These data visualizations focus on long-term analysis and the presentation of trends.

Operational purposes: Operating processes are monitored and shorter periods of time or real-time data are considered.

Analytical purposes: Data visualizations for primarily analytical purposes are designed to extract the essential information and results from large and often heterogeneous amounts of data.

Planning purposes: A planning dashboard is aimed at middle management and helps to formulate growth strategies based on trends, strengths or weaknesses across different departments.

2. Don't put too much information on one page

When designing a dashboard, you very quickly fall into the mistake of trying to pack too much information and details on one page. The representations are then comprehensive, but completely overloaded with information. The essential facts are difficult to grasp. In principle, it is impossible to design a dashboard that meets all the requirements of the different target groups at the same time. Put yourself in your target group's shoes and think about what information they need. Sales managers want to see different information than, for example, the HR department, marketing specialists or purchasing experts. But even within the same department, employees at different hierarchical levels have very different information needs (e.g. junior sales manager vs. sales manager). It is ideal to design a separate dashboard for each target group. If this is not possible, work with separate dashboard tabs - one tab for each target group. By simply clicking on a tab, the respective person will find exactly the information they need and do not have to endlessly click through filters, selection lists or data series. A helpful rule of thumb in this context is: do not pack more than five to ten different data visualizations on one dashboard! More visualizations create confusion and are very difficult for the human brain to process at the same time.

3. Select the relevant KPIs

Once the target and target group of your data visualization have been determined and you have understood that only a certain amount of information fits on a dashboard page, the next step is to select the KPIs that are actually relevant. Only with the right KPIs will you be able to offer the right insights into the respective business areas. The right KPIs are a must for data visualizations that efficiently support decision-making processes. To support you in this selection process, we have put together over 250 KPI examples and KPI templates for you. There you will find suitable KPI examples for the various business areas such as management, finance, sales, marketing, human resources, service and support, purchasing or IT. One of these examples comes from the retail sector:

The KPIs in this dashboard section show the total volume of sales and the average shopping cart size over a certain period of time. These values ​​are useful for retailers. You can see at what times the demand for your products is higher or lower. It is easy to determine when measures such as advertising or discounts to increase sales make sense.

4. Present the information in a specific context

Without presenting the information in the right context, it is difficult to understand whether values ​​are good or bad or typical or atypical. Pure numbers on a dashboard are meaningless to users as long as they are not related to comparative values. Even more important is the fact that without comparative values ​​it is not clear whether a certain action is necessary.

To put the information in context, give the charts a name. Name all axes and provide comparison values. Even if you take this information for granted, it should definitely not be missing. They support the viewer's understanding. Good comparison values ​​are, for example, comparisons with set goals, with past time periods or with planned values.

5. Keep the dashboard as simple as possible

One of the most important dashboard design principles is: keep it as simple as possible! This so-called KISS principle (keep it short and simple) is used successfully in many other business areas. For the dashboard design, this means that the information must be presented in an understandable and easily accessible way. If design and charts seem too complex to the viewer, they are initially only concerned with grasping the representation and understanding the structure without being able to absorb any information at all. A good rule of thumb in this context is: the viewer should find the information relevant to him within five seconds. The important key figures must be immediately recognizable and the essential questions must be answered at first glance. The user must never be forced to first perform calculations or link the information himself in order to get the answers to his questions. In this context, as already described in the previous chapter, it is important to present the information in context and, if necessary, to provide comparative values ​​or to insert easily understandable trend indicators.

The following example shows a lot of information on just one page with a total of nine different visualization blocks, but the simple presentation makes it possible to record relevant values ​​such as sales figures or profit. Colored trend indicators and percentages show at a glance the development compared to the various comparative values.

6. Structure the layout and take into account the importance of the information

A well-structured layout and the correct placement of the individual information elements on the dashboard page should not be underestimated for the effectiveness and efficiency of a dashboard. Your presentation should be structured and visually organized to make it easy to capture the information you need. Nobody likes to work with data and information that has to be laboriously gathered from a jungle of charts and numbers. The basic rule is that important information goes up, less important information goes down. The key information should be at the top left of the dashboard page. This is due to the fact that we automatically begin to read what has been written at the top left. If the most important information is there, the viewer captures it first.

Another helpful design principle is to start with the so-called big picture and the most important information or trends. This initial overview is followed by more detailed information and supporting charts. Charts with information that are related to each other should be placed next to each other. In this way, the users do not lose their focus while they grasp the context.

In summary, it can be said that a good dashboard layout is based on the reverse pyramid logic that is used in journalism. The essential information is placed above. The details of this information then follow. In the lower area you can find more detailed background information, which enables a deeper understanding of the topic.

7. Use colors sparingly and visually clearly separate blocks of information from each other

When it comes to the use of colors, the instinct to play runs through dashboarding with many designers. But in this area too, less is more! Be economical with the use of colors and optical effects. Limit yourself to a few colors and stick to these colors consistently. Avoid the typical style of PowerPoint presentations from the 1990s. A modern dashboard is minimalist and clearly designed. The "flat design" that is common today dispenses with realistic representations of textures, three-dimensional structures, cast shadows or extravagant decorations.

A good approach is to choose the colors, fonts and logos of the corporate design and use them sparingly and consistently. But stick to two or three colors and only play with weakened gradients of these colors. Under no circumstances should too many saturated colors be used. Intense colors can be used to draw the viewer to certain information, but too many saturated colors overwhelm him and do exactly the opposite.

Information that belongs together is best presented with the same basic color and differently attenuated color saturations. In this way, the viewer recognizes the relationships based on the colors. A slightly different color saturation, for example, represents a certain quality of a variable.

Caution is advised when using the traffic light colors. Many associate the colors red or green with the properties "good" or "bad" or "stop" or "go". You should only use traffic light colors in this context. In another context, they create confusion and misunderstood statements.

Another design principle in the color area deals with the delimitation of the various information blocks and elements of a dashboard. Sufficient space between the blocks plays an important role for the visual composition and the rapid comprehensibility of the information. To delimit the information blocks, the same color must be selected on the entire page, which is clearly different from the other colors used. For example, the colors white or gray are good choices. Also make sure that the boundaries are wide enough and follow a horizontal or vertical structure.

The above KPI dashboard example from production shows a very successful and harmonious use of colors. A dark design with a few colors is chosen. The individual information blocks are clearly delimited from one another with wide gray borders.

8. Choose the correct chart types

Choosing the right chart types is critical to meaningful visualization of your data. With an incorrect or missing diagram, you can ruin all your efforts to present results in a meaningful way. It is important to understand what type of information is to be transported and what type of diagram is suitable for it. Depending on the information to be conveyed, line charts, bar charts, pie charts, sparklines (small progression charts), scatter charts or bubble charts can be used.

Line diagrams are very suitable when changes are to be illustrated over a period of time or over another course. They are compact, easy to understand, and accurate. Most people are familiar with the format of the line graphs and will be able to see what they are saying right away.

Use bar charts when you want to compare items of the same category (for example, sales broken down by customer country of origin or product category). These diagrams, too, are easy to understand, compact and precise.

Pie charts are not a good choice in most cases. They are imprecise because it is difficult for the observer to compare the sizes of the various sectors of the circle with one another exactly. Pie charts can be useful if the goal is simply to ensure that the target group immediately recognizes the largest pie sector. The smaller and more circular sectors are, the more difficult and incomprehensible this type of diagram is.

Sparklines are small diagrams that show the course curve of a series of measurements. They have no scaling, which is why users do not recognize the individual values. Sparklines are good when you have a lot of values ​​but just want to show trends. The trends in the small diagrams can be seen at first glance.

The information in scatterplots is relatively difficult to grasp. They lack precision and clarity. Only use this type of diagram with a knowledgeable audience. They can be used, for example, for interactive presentations.

Experts largely agree that bubble charts are unsuitable for a dashboard. It is a type of diagram that most users are unfamiliar with and that lacks precision and clarity. Even when presenting simple information, a high level of intellectual performance is necessary to understand it.

In order to choose the right diagram type for dashboarding, visualizations can be divided into the four categories of relationships, distributions, compositions and comparisons (Relationship, Distribution, Composition, Comparison). If you assign your presentation goal to one of these categories, you will quickly find the right diagram type. The following graphic shows which diagram type is suitable for which type of data visualization:

9. Use the possibilities of interactive elements

Interactive elements allow the user to go deeper into individual issues with a click of the mouse without overloading the dashboard. Useful interactive elements are, for example, drilldowns, click filters or time interval widgets. Drilldowns allow you to refine the information and zoom in on the existing data. Virtually any level of detail can be achieved with just a few clicks of the mouse.

Another interactive element is the click filter. With these you can assign temporary filters to the diagrams and graphics by clicking on individual elements and change the dimensioning. The following example illustrates how a click filter works:

Visualizing data over different time periods is another important goal of dashboard design. The interactive element of the time interval widget offers the right solution for this purpose. With a click you can dimension a diagram for a certain period of time and the graphic immediately shows the corresponding data, for example in annual, monthly and weekly periods or on a daily basis. The following example shows how such a time interval widget works:

10. Don't overdo it with real-time data

Another tip for a good dashboard design concerns the use of real-time data or the selected dashboard update intervals. In a few cases it may be necessary to track and display values ​​in real time. Most dashboards do not need to be continuously updated and do not require live data.It is usually sufficient to update data periodically, for example on a weekly, daily or hourly basis. Too intensive use of real-time data often only distracts from the essential content. The data should clarify the overall picture of a situation or a trend and not cause unnecessary restlessness in the presentation and statements through constant change. Ultimately, choosing the right data is more important than its current status.

11. Use the feedback received to optimize the dashboard design

A last but certainly not unimportant tip about good dashboard design is to use the feedback received to optimize your visualizations. Approach the individual target groups and collect feedback on a regular basis. By asking the right questions, you will receive valuable feedback to continuously improve the layout, features, look and feel and balance of the KPIs of your dashboard design. This is especially important as the digital business world is constantly changing. The only thing that is constant is change. So never stop developing your dashboard and react to the changes. This is the only way to ensure the continued success of your analysis efforts.

Summary & recommendations for action

In our article we explained to you how to recognize good or bad dashboard design and which best practices and design principles exist for dashboarding. As a summary it can be said: A good dashboard is simple, not overloaded with information, easy to understand, clearly structured visually and presents the relevant information with the appropriate diagram types. The target group and the key messages to be conveyed should never be left out of sight. Ultimately, the data presented serve to support the decision-making process with relevant information. Follow all of the design advice listed here, create meaningful visualizations, and increase the success of your analytical efforts. They enable everyone in the company to identify the essential information quickly and easily. This creates the best conditions for more growth and success of your business.

Here we have put together over 75 dashboard examples for good dashboard design and a complete library of dashboards for different purposes, industries and platforms.

If you would like to put your newly acquired knowledge of good dashboard design into practice and increase the success of your company, do not hesitate. Register now for a free 14-day test and start using datapine’s dashboard tool today. Create your individual dashboard with just a few clicks and impress your superiors, colleagues or partners!