We are often called in to assist clients with their business intelligence solutions and the one common challenge that we find is to simplify the information so that the user can effectively action the information. We have come up with a simple and sometimes controversial approach – we don't always give users what they want.
Technorati Tags:
Business Intelligence,
Data Visulaisation,
5point9
Business Intelligence solutions are growing in importance and the need to have effective access to information and maximising the benefits of the insights are now business imperatives. Whether to meet compliance / regulatory demands in financial services, advanced analytics in anti fraud cases or performing forecast scenarios the need to enable users to simply converse with data is critical.
It is always tempting to try and offer the user a feature rich experience in order to maximise the amount of data available in a dashboard or report. Offering users the maximum from what they have is always a solid approach however it is often by saying "no" to features or data that can be more useful to users and the business. The risk and common pitfall that we see is that there is a risk of "noise" which can dilute the information and insights that are available for the user. This noise can be in the form of irrelevant or spurious data, extra features which users rarely use of complex filtering or other secondary and tertiary features which can slow reports or make it very complex. We are not suggesting that we want to limit the ability to drill or interact with the data but that these features are done in a logically and intelligent way which makes sense.
It is always positive to get the users input into dashboards, dials or reports but the issue is that the user does not always know what they want – this may sound counter intuitive or a brave assertion but the number of dials and pie charts that we see allow us to conclude that things have been done because they could rather than what they should. User feedback is important for adoption but in order to avoid the above pitfalls their should be a business decision maker who has the right to veto and is the single point of decision making.
The approach that we use is to create prototypes on paper which can then be reviewed in workshops with the various decision makers and users. This is a powerful tool and can be drawn on with marker pens to show how they can be improved. After a couple of iterations there is usually consensus and the single decision maker has come to the conclusion of what should and should not be included. Once this has been completed the initial version of reports and tools can be created and shared with the business. With web analytics, this should then be monitored to see how the users are interacting with the tools and also who effective the reports are being, we always try and tie back the reports to actionable insights and therefore business results – this helps to focus the offerings and can highlight areas of improvement or additional information requirements. This continuous improvement cycle will have a longer timeframe and require less effort once the business intelligence solution is in place as the solution is honed and streamlined to the business.
Thanks,
You can reach us at Technical@5point9.com
