Best Practices in OBIEE Visualization

Best Practices in OBIEE Visualization

bigstock-Business-man-point-light-bulb--27052619_zps226ca938

In this post, I’d like to share three best-practice tips about OBIEE visualization. Fortunately, there are numerous tools that you have in order to bring the data to your users in an efficient way.  There are books written on the subject of dashboards and reports. However, I hope these simple tips might prove useful to you.

1. Always keep in mind your target audience.

If your Dashboards are primarily meant for executives you should consider following the top to bottom drill-down approach. I know it sounds like common sense; however, one would be surprised to see some of the detailed reports that were exposed to the executives on the entrance page.  Remember that this audience is usually impatient and is in a rush. Open up with some high level KPIs (such as Gauges) and implement some navigation to the more detailed reports. With the new (and long-awaited) features in OBIEE 11g, such as in-place drill down – this should not be problematic or require too much customization. Keep filters simple and at a minimum and remember, they want to know the strategic state, getting to the detailed level if necessary.

On the other hand, if your audience is more interested in a mix of analytical and operational reporting (store managers, regional supervisors, departmental heads) – you might consider entry pages consisting of a mix of high-level and granular data with more detailed dashboard prompts.

2. Keep your Dashboards lean, consistent, neat, and organized.

This is actually true for both user experience and performance considerations. Unless you are running on Exalytics, OBIEE 11G generates some heavy client-side JavaScript that might take a while to render on report-heavy pages. Also, too many reports on one page will make it more difficult for users to navigate and identify necessary info. Try to keep your Dashboard pages organized in a similar layout with consistent colors, fonts, and margins. This just implies quality and creates a positive user experience. Also, try to keep naming of the report in a similar logical manner. Always build an index page with text links leading to all material, do not assume that users will find what they need immediately. If you use borders around reports, use them everywhere. Make sure your prompts are consistent and use the same labels for filtered columns. For best results, you can utilize A/B testing with focus groups to get the Dashboards your users will like to use.

3. Use correct way to convene the message.

Avoid using inappropriate report views to deliver the message (a lot of it is common sense, i.e. using a pie chart in a time-series analysis).  Think before you use sparks or trellis charts if they are really necessary or if it’s just eye candy.

Tags:

Write a Comment

x

Contact Us Close