In this 3rd blog post regarding Oracle’s P6 Analytics we will begin to discuss the user experience. We’ve generally identified what P6 Analytics is and what it does in the first blog post. In the second blog post, we talked about specific product features and configuring your Primavera EPPM environment. In this 3rd blog post we will identify what the user experience looks like and how you can leverage the subject areas.
Generally speaking P6 Analytics gives you access to an already organized data warehouse and a set of canned reports. The canned reports can be configured to work for your specific system configuration, but you may want to improve upon what you see in the canned reports or even start from scratch. Analytics appears very similar to the Business Intelligence (BI) Publisher; the main difference is that Analytics comes with a few more reporting options and subject areas derived from the data warehouse. If you are familiar with BI Publisher then your user experience will seem very familiar.
When building your own reports it is important to think first about what data you want on a specific report. What questions will this set of data answer? Getting the data is essentially done by querying data through multiple subject areas. You don’t have to have the ability to write code to query the data, this can be done exclusively through drag and drop. Users are still given the ability to interject code if desired. The data from your Primavera EPPM environment is organized into 13 subject areas. Each subject area is fairly self-explanatory. Subject areas also contain fact data elements where measurements are set up to be queried as multiple dimensions such as planned labor hours by project as stored by transaction (where the transaction refers to a historical snapshot of your data)
Once the data is defined the presentation of that data should be considered. Data is put together one analysis at a time. An analysis is a report fixed to one specific set of data. However, you may include multiple presentations of that data within an analysis such as table, charts, graphs, and other functionality for representing data. You may never need to build anything more than an analysis, but if you need multiple data sets to be represented then a dashboard is a quick and easy way to do that. Just like the query of data process, dashboards can be designed by dragging and dropping analyses and / or other items. For example you may want to have 3 unique data analyses, but you may want them to all be filtered by the same parameters. In this case the completed analyses and prompt, which essentially acts as the filter, can simply be added to a dashboard page in a matter of seconds.
Analytics is a powerful tool, but requires input from the user to make it useful. Without the user then it is just organized data with potential. Analytics cannot be thought of as just a data warehouse, or just a reporting tool, or just a tool that is automated to the point of permitting ad hoc report writing. It is all those things and the ability to have everything in one package is what makes it powerful.