4 Best-Practice Tips to Improve Your KPI Dashboard

Thursday 19 May 2016
Reading Time: 4 minutes

The KPI dashboard is often the staple report sent to management figures in various departments across the organisation. Its aim? To help the viewer make informed decisions based on real time business performance at a glance.

For example, their area of responsibility could be a particular division of the organisation such as a department, or, at an even higher level, a collection of divisional reports to show the overall health of the organisation for top-level management executives.

The key phrase you should take away from this is “at a glance”. Too often these reports are cluttered with information that could be better displayed or removed when not relevant. The focus should be on creating a dashboard that quickly delivers relevant, easy to understand information regarding the KPI’s set – no more, no less.

1. Apply design principles

The first best practice tip I have for you is perhaps the most simplistic and therefore somewhat overlooked aspect of dashboard creation – styling and formatting.

Of course the most important facet of a KPI dashboard is the data included in it. That data, however, needs to be interpreted correctly and applying basic design principles goes a long way.

We humans are visual creatures. The use of colours and imagery really do liven up an otherwise “bland” looking dashboard. Consider the KPI dashboard as a medium to tell a story and, with that, use design principles such as colours and imagery as a visual aide.

I advise a selective colour palette (3-5 colours where possible) of contrasting colours for readability and well-spaced elements to ensure the dashboard is kept simple and uncluttered, providing easily digestible information to the viewer.

A report showing a typical A&E dashboard of an NHS Trust

2. Be visual with your data

Creating a good theme and layout is just half the battle, however. The remaining half lies in the use of charts and other visual elements to better display key information.

Nobody wants to look at a report that appears if it has come straight from the 80’s. This is the 21st century, it’s time to get a bit more Minority Report and a bit less Flash Gordon.

We should be making a greater effort in making our KPI dashboards visually appealing. The style of infographics lends itself to this very well and it is what I would actively encourage report builders to start thinking about this as a reference when building their reports.

Infographic report displaying Connexica's performance over the last 5 years

3. Use the right data

Remember that the data you’re using needs to be directly relevant to the KPI’s the manager is interested in. A common mistake when creating a KPI dashboard is including granular data, often because the user has tried covering all angles of what they think that particular manager will be interested in.

In reality, the manager is only interested in the high-level data that shows how well their area of responsibility is performing in relation to the KPI’s set.

As an alternative, consider including the granular data as a separate report as part of a wider report pack. In CXAIR, users are able to create “tiles” that would achieve this, allowing users to drill-down on a particular value and get taken to the underlying report data via our drag and drop interface.


CXAIR Crosstabs - drill through on your structured data!

4. Use the correct chart

Finding and displaying the right information is obviously one of the significant aspects of dashboard creation, but let’s not take away the importance of using the correct chart to showcase information pertinent to certain data sets. Depending on the type of or volume of data you are dealing with, choosing the correct chart type becomes very important.

There’s no exact science which dictates what chart you should use to showcase your data. The best advice I can give is to obtain a solid knowledge of the data you’re working with or bring in someone that can help in this regard. Other than that, apply basic principles such as when you’re working with a dataset with a large number of values it wouldn’t make sense to use a pie chart, for example. For more in-depth guidance on choosing the right chart, feel free to check out our fact sheet on using the right chart.

Using the correct image


Get in touch to discuss your requirements further