It’s funny that Forrester’s Boris Evelson recently received a slew of comments from data warehousing and business intelligence pros wondering why they hadn’t included an evaluation of Excel as a business intelligence tool. Boris goes on to explain why Excel doesn’t work as a standalone BI tool.

Here’s my take on the subject:

Like it or not, the most widely used BI tool today is the spreadsheet. If you are a BI or data warehousing manager, you need harbor no illusions that you are gathering and manually entering data into spreadsheet. Furthermore, you need to work with the logic captured in Excel spreadsheets and enable this logic to be used for analysis and decision support.

What’s the problem? Simple. The proliferation of custom-built reports with unstructured data pulled from who-knows-where using self-serving metrics brings into question the value of the information available.

The problem is not the spreadsheet, but rather the fact that the data pulled into it may come from various “unofficial” databases (i.e., not your data warehouse or data mart) and the data may be transformed without the visibility that is now so critical in a corporation. In short, the problem is not the use of Excel versus some other BI tool but rather the inconsistent and too often flawed data integration that is done to feed the spreadsheets.

This situation is a symptom of a problem, not the actual problem itself. Business users pull data in and transform it because it is either not available to them or they don’t understand that it is available. No one is going to go through the hassle of gathering and transforming data if they know it is already in your data warehouse, easily accessible from Excel. Business users aren’t going to create extra work for themselves.

What should IT do? First, if the business user wants to use Excel, why argue? You aren’t going to win. Your BI tool already connects to Excel. Some even connect in a bidirectional manner, allowing your business user to get spreadsheets refreshed with new data as it becomes available and share spreadsheet results with others. If not, consider how you can set up a data distribution capability from your BI tool and data warehouse to your user’s Excel spreadsheet.

Second, use this as an opportunity to examine what data your users are using in their spreadsheets and determine how to get it into your data warehouse, data mart or analytical cube. Often times most of the data the business needs already exists in your data warehouse, but either supplemental data is needed to be added or data from the data warehouse needs to be filtered and/or aggregated into a data mart to enable business analysis. This is an excellent opportunity to work with the business, improve its productivity and increase the value of your efforts.