Manage Learn to apply best practices and optimize your operations.

Perk up your BI planning with write-back

How do you get more out of your business intelligence (BI) system than just reviewing data? Write-back allows for solid BI planning -- it proposes changes written to the system, letting you play “what if” and allowing for creative decision making and implementation.

Business intelligence (BI) can be a useful tool for businesses to spot the root causes behind problems, identify patterns and trends, and ferret out new business opportunities. In many cases, however, BI systems let you simply review data -- and little else.

Better BI systems support something called write-back, a feature that can improve your BI planning. Write-back lets you propose changes to be written into the BI system itself. Those changes affect calculations made from the data, and the BI system readjusts and displays the new results. In this way, write-back lets you play “what if” games with your BI system, helping to drive decisions and implementation throughout the enterprise.

Typically, the BI system starts with data in a data warehouse -- “real” data drawn from your production data sources. If that’s all the data you had to work with, you wouldn’t be able to create many “what if” scenarios. That’s because the data warehouse doesn’t store “if”; it stores “what really was.” Write-back simply allows data in the data warehouse to be changed -- sort of.

Of course, you don’t want to allow users to change real historical data on a whim. Instead, using Online Analytical Processing (OLAP), the write-back data doesn’t go into the data warehouse. OLAP stores it elsewhere, creating a sort of “overlay” on top of the data warehouse, but leaving its information intact. This overlay is used for further analysis.

This write-back process is often referred to as “planning support” by BI vendors; the intent is to help you plan future actions by writing changed assumptions back into the BI system. The system can then re-calculate whatever outcomes you are looking at, and you can use those new outcomes as milestones and goals.

A quality BI system will allow multiple scenarios to be written back into the system, typically allowing you to tag each scenario. For example, let's say you have a “best case” set of data and a “worst case” one; you could switch between them on-demand to see how each brought about the outcomes you’re looking at.

Note that some BI systems do allow you to write back to the real data. That’s not always a good idea, unless the system provides an explicit way to “roll back” your changes and return to the real-world data you started with.

Dig Deeper on SQL Server Business Intelligence Strategies