Data visualisation is the graphical presentation of information in a way that makes it easier to understand and process. It is usually interactive and allows for the user to draw their own conclusions about the data. Some common data visualisation formats are tables, bar charts, line charts, scatterplots and maps.
An infographic is very similar to a data visualization since it too is a graphical presentation of information that is presented in a way that is easy to digest and understand. Infographics are more subjective though by telling a complete and premeditated story which guides the user to a particular conclusion. Refer to the Equinor Brand Center for guidelines on creating infographics.
A dashboard is a combination of data visualisations arranged on a single screen which can be monitored at-a-glance. This is usually a summary of the most important data that is needed for users to take action or make decisions. A dashboard should effectively communicate a story using data.
Data accuracy is the top priority. Data must be displayed correctly and in a way that does not mislead the user.
- Be careful how missing, null and zero data are displayed.
- Use visual techniques to effectively highlight and draw the user’s attention to the most important data.
- Focus on presenting the information needed in the most simple and clear way.
- Provide adequate context in the form of labels, data values or units to aide in interpretation of the data.
- Add supplemental data when necessary to provide the correct context to understand the data. Examples are targets, trends, budgets, YTD, averages, etc.
- Enable users to drill-down with context to perform further analysis away from the dashboard.
- Sort data for easier comparisons.
- Minimise use of decimals where they are not relevant or don’t add value.
To assist the user and simplify the process of finding information, the information should be positioned corresponding to how the users are trained to read. Western cultures read left to right, meaning the most frequently viewed information should be placed in the upper left-hand corner. The least frequently viewed information should therefore be placed in the bottom right-hand corner.
Put thought into how the information is grouped on the dashboard based on the relationships of the data and the desired flow for the user. Keep in mind that users naturally assume that elements located near each other are related. They will also perceive that elements that visually look similar (size, shape, colour) are related even if they are not located near each other.
- Use the Equinor typeface when possible, otherwise use Arial.
- Use a screen size ratio of 16:9.
- The dashboard should be confined to a single screen without scrolling.
- Distribute content evenly over the width and height of the dashboard. This enables the user to see the whole picture at once.
- Tabs to other relevant views or pages should be placed along the top below the title. Don’t have more than 6 tabs.
- Always use common navigation design patterns that users are familiar with such as filter panels and tabs across the top.
- There should always be an overall dashboard title located at the top.
- Use whitespace to help group information and to declutter the page.
- For example, add 16px margin around the entire dashboard, 16px padding around content in a visual, and 16px between visualisations.
- Global filters should be grouped and placed together either on the top, left or right side of the dashboard. Filters applying to single visualisations should be placed close to the visualisation in which they apply to.
Use colour appropriately. Colour can be used to draw the user’s attention to something important, and it is also an easy way to help the user identify patterns or trends.
It is not recommended to rely on colour alone, since some users can have trouble distinguishing certain colours (colour blindness).
- Follow the guidelines given on EDS.
- Keep bright and saturated colours to a minimum, only using them to highlight data requiring attention.
- Make sure that colours used on a dashboard consist of one palette of colours, do not mix from different palettes.
- Visuals can off a white or light background colour but use a non-white background colour like Light or Medium grey for the overall dashboard.
Data visualisation types
It is important to understand the different types of visuals available in order to choose the right visual for the intended purpose.
Visualisations should include the data and context necessary to communicate the intended information in a clear and effective way.
- Be consistent with labelling and data formatting. Always label axes, show units and add a title.
- Be aware of the scaling used and that it conveys the data correctly.
- The vertical (y-axis) for a bar chart should usually start at zero. The starting point in other chart types should be reasonable for the data.
- Avoid 3D, background images, animations or other unnecessary effects that reduce comprehension.
- Ensure the user is not being misled. For example, parts of a whole should always add up to 100.
- Use visual cues like arrows, colour or text to help users interpret the data.
- Enhance the data pixels and reduce unnecessary non-data pixels (gridlines, borders, certain fill colours, etc).
- Add a means for the user to interact with the data such as filters, slicers, sorting and drill-down possibilities.
- Avoid pie and donut charts as they can be misleading and difficult to interpret.
Recommend further reading
Explore the books, blogs and articles written by data visualisation experts: Edward R. Tufte, Colin Ware, Wayne W. Eckerson and Stephen Few.