Last month's column emphasized the importance of presenting information that is personalized, actionable and meaningful - in short, relevant - in a customer value dashboard. But that's not all. Dashboard contents need to meet other conditions to be truly useful.


Fresh means more than just showing the latest information available. The dashboard must highlight the implications of that information. For example, new customers from different sources may be known to perform differently in terms of subsequent purchases. When the mix of customers changes, the dashboard needs to show the expected impact of the new mix on future revenues, not simply report the mix change or, worse still, just the total revenue from the current month.

In businesses with regular contract renewals, such as mobile telephones, leading indicators of renewal rates are often available well before the actual expiration date. The dashboard needs to report the expected renewal rate based on the status of those indicators, rather than waiting until final results are in.

In addition, the dashboard should show how expectations have changed based on the latest batch of data compared with the previous batch. Long-term forecasts from narrow slices of data - perhaps one month's transactions - will necessarily be volatile, so some smoothing will be required. How to do this without hiding whatever interesting trends are embedded in the data is part of the art of designing such systems. Some false alarms are inevitable, and some real problems will be missed for a time; but it's better to try to forecast and make some mistakes than to report only actuals and have no advance warning.


The dashboard should not show static or redundant information. Values that are updated monthly should not appear in a daily dashboard report, although they should be available for review at any time. Better to have a daily report for information that appears daily, such as Web marketing campaign results (which might actually be updated much more often - perhaps every 15 minutes). Provide a separate tab for monthly information and highlight this tab on the day the new month's results are released. This will encourage users to check the new monthly figures when they are at their freshest and, therefore, when users have the most possible time to react. Like old information, redundant data makes the dashboard harder to use because the viewer must look in more places to learn something new. It is better to present important information in one place and then go on to something else - always keeping the older information available for recall if a user should need it and giving the user the option to combine information in different ways when it suits her purpose.


Information in the dashboard must not only be correct, but it must be accepted as correct. This means users need to be able to relate it to other data sources they already trust. It is not a matter of presenting the same numbers as older sources; a customer value dashboard is likely to include forecasts and aggregations that are not found elsewhere. Users should be able to dig into the dashboard calculations to see how they were generated and what data they are based on. They should be able to tie this underlying data back to existing, trusted sources.


Dashboards have pretty dials and gauges and warning lights, but these aren't just for decoration. They are used to make the information more understandable and to direct users to the areas requiring attention. Users should have some control over what is included and how it is presented because different users have different styles of absorbing and working with data. The components themselves need to be more than pretty pictures. They must be connected with underlying information so users can dig into an interesting item to find out more details.

In the case of customer dashboards, it is also important to clearly distinguish actual values from projections so users can understand which items might be changed by management actions. Ideally, the dashboard would include simulation and modeling capabilities that let users explore the impacts of alternate assumptions or changes in future results - although this is arguably extending beyond the scope of the dashboard itself. 

Register or login for access to this item and much more

All Information Management content is archived after seven days.

Community members receive:
  • All recent and archived articles
  • Conference offers and updates
  • A full menu of enewsletter options
  • Web seminars, white papers, ebooks

Don't have an account? Register for Free Unlimited Access