Image may be NSFW.
Clik here to view.This blog post ties in very well with the previous post where we discussed the importance of not losing sight of the actual user that will consume the visualisation. We discussed amongst other things, the importance of designing the visualisation for a cross-section of users taking into account their analytical efficacy. This post will take a look at the concept of simplicity as related to information presentation.
Key Considerations
Design by function
While it is imperative from a BI perspective that we move away from silo thinking, I have found in my experience that visualisation built across specific themes are not only easier to manage, but also easier for end-users to understand and make use of. From an organisational perspective, As an example, I would rather have two separate workbooks, one dealing with sales and any analysis thereof, and another dealing with employee HR details rather than having a single one with visualisations that span both data sets. Creating dashboards by function will also make overall maintenance easier.
Avoid clutter
For data visualisations to be effective, they have to be designed in such a way that they are easy on the eye. Dashboards should not be about seeing how much information and how many graphs can be placed onto a single page – they should be about graphically communicating pertinent information that can result in an action or corrective measure being taken to the user at a glance. Having clutter will result in degradation of the above and in the real possibility of confusing the user.
Condense the information
A major reason for making use of data visualisation is to condense large volumes of data into compact meaningful visualisations. In general, the less space needed to communicate the message without losing context and meaning, the better. The best way of doing this is to make use of summaries and exceptions. The former includes presenting results at summary and average level whereas the latter only includes the display of those things that are outside the defined toleration band. Knowing the intended audience, and business case for the requested visual, is critical for deciding on the summarisation level. A further note on condensation of information is that it is often not necessary to use decimals and in those cases where they will not impact the decision to be made, they should be done away with.
Focus on the take-away
Understanding the users’ needs at a high level is key in designing visualisations and it is important that the data you want the user to concentrate on be made to stand out. This can be achieved using any of the ideas that we have discussed in previous blogs. Examples are using fully saturated colours, altering font and element sizes as well as placing the key data either in the top left hand corner or in the middle of the viz. All these will focus the attention of the user to the major points you want them to take away.
Conclusion
The above pointers on simplicity of design are there to help drive cognition and usage of information. If the viewer fails in either of comprehension and understanding, then regardless of how well designed the visual is, it cannot be counted as a success. The key takeaway from this should be “Keep It Simple”!
Also in this series:
- Data Analysis with Good Visualisations – Guided Analytics
- Data Analysis with Good Visualisations – Interface Design
- Data Analysis and Good Visualisation Practices – make BI actionable
- Data Analysis and Good Visualisation Practices – use comparisons
- Data Analysis and Good Visualisation Practices – consider your audience