Using the Decomposition Tree
A Decomposition Tree is a PerformancePoint analytics tool that you can use to perform root-cause analysis by viewing how individual members in a group contribute to the whole. The Decomposition Tree enables users to decompose, or break down, a group to see its individual members and how they can be ranked according to a selected measure, such as by sales amounts. Decomposition Trees are only available on views that use SQL Server Analysis Services data.
The Decomposition Tree automatically sorts results and applies and inline Pareto chart to the data, so you can quickly see the highest contributors to a particular report value. You can also see trends across individual members that contribute to an overall value.
Note: You must have Microsoft Silverlight 3 installed on your computer to use the Decomposition Tree. Depending on how a scorecard or an analytic view is configured, you might not be able to open the Decomposition Tree.
How do I open and use a Decomposition Tree?
To use a Decomposition Tree, use the following procedure:
-
In the dashboard, select a view that uses SQL Server Analysis Services data. Examples include analytic charts and grids, and some kinds of scorecards.
-
Right-click on a value in the view, and then select Decomposition Tree. The value can be a cell in a scorecard or a grid, or it can be an amount in a chart. The Decomposition Tree opens in a browser window.
Use the following procedure to conduct root-cause analysis:
-
Click a decomposition node. A list of dimensions and hierarchies displays.
-
Select the dimension and hierarchy that you want to use. The view automatically refreshes to display the next level of detail for the item that you selected.
-
Repeat Steps 1 and 2 to continue drilling to detail in the view.
-
To view member properties for a dimension member, browse to an individual dimension member. Click its node to open a dialog box that contains additional options. Click Show Properties.
The following image shows a Decomposition Tree showing sales unit costs across product categories, product subcategories, and individual products.
Using the Decomposition Tree, you can also view member properties for a particular dimension member
Root cause analysis
ReplyDeleteRoot cause analysis (RCA) is the process of discovering the root causes of problems in order to identify appropriate solutions. RCA assumes that it is much more effective to systematically prevent and solve for underlying issues rather than just treating ad hoc symptoms and putting out fires. Root cause analysis can be performed with a collection of principles, techniques, and methodologies that can all be leveraged to identify the root causes of an event or trend. Looking beyond superficial cause and effect, RCA can show where processes or systems failed or caused an issue in the first place.Root cause analysis
Core principles
There are a few core principles that guide effective root cause analysis, some of which should already be apparent. Not only will these help the analysis quality, these will also help the analyst gain trust and buy-in from stakeholders, clients, or patients.
Focus on correcting and remedying root causes rather than just symptoms.
Don’t ignore the importance of treating symptoms for short term relief.
Realize there can be, and often are, multiple root causes.
Focus on HOW and WHY something happened, not WHO was responsible.
Be methodical and find concrete cause-effect evidence to back up root cause claims.
Provide enough information to inform a corrective course of action.
Consider how a root cause can be prevented (or replicated) in the future.Root cause analysis
As the above principles illustrate: when we analyze deep issues and causes, it’s important to take a comprehensive and holistic approach. In addition to discovering the root cause, we should strive to provide context and information that will result in an action or a decision. Remember: good analysis is actionable analysis.