1. Avoid the use of quick filters on dashboards to improve load times. Use cross tabs, heat maps and highlight table filters….sometimes pie charts work for this purpose as well.
2. Keep dashboards to 4 panes and imagine a Z pattern of importance from upper left to lower right.
3. Don’t build grids or pivot-table-looking dashboards. Use visual analytics and bring details in on demand via filter actions, jump filters, annotation or customized tooltips.
4. Pimp your tooltips at the end. If you don’t, you will create more work for yourself.
5. Use color sparingly and never more than 2 ways in a single dashboard. Preferably one.
6. Avoid pie charts for one to many comparisons.
7. Use bullet graphs or bar charts.
8. Don’t repeat the same chart type in dashboards – its boring.
9. Leverage data blending for ad hoc analysis to add meaning and WOW factor.
10. Strive for a high data to ink ratio – eliminate color, shape, size or text that doesn’t add information or is redundant.
Finally, be aware of the three types of data and in what contexts they arise. And, to insure success….get training….preferably a 2-day onsite to learn technique….from InterWorks.