To help ensure business intelligence system success, IT should automate data collection and standardize data references. In order to create a business intelligence system that serves the needs of the business, IT has two challenges: getting clear direction regarding what data it needs to track, and maintaining consistent data. Read these tips to ensure you can support successful BI projects. Automate data collectionTo maintain quality control over BI projects, Dan Hooper, vice president of sales and marketing for Dallas-based Microsoft Gold Certified Partner Integrated Services Inc., recommends tools such as Microsoft SharePoint that automatically track when employees enter data into systems (in the same way that an e-mail message automatically incorporates information about when it's sent). "It delivers the metadata about time, process, and workflow," he explains. This helps you see if someone is taking too long to submit information needed for analysis. The other reason to automate collection is to improve data accuracy. Caron Mooney, director of IS Partners, a Johannesburg, South Africa-based Microsoft Gold Certified Partner specializing in BI, recalls a client that wanted to track when shipments arrived. Its goal was to ensure that suppliers were adhering to their contracts regarding delivery service levels. When the company started collecting information, the data showed that every single delivery was late. But in reality, in order to process trucks waiting in line, workers were tossing shipping documentation into a basket to enter the following day. "All the IT department had to do was institute a simple bar code on the paperwork and have workers swipe it through a reader," Mooney says. The lesson: Sometimes you don't need an expensive system to collect data automatically. Establish consistent data referencesA key tenet for a business intelligence system is the ability to collect information from a variety of sources and analyze patterns across an entire company. To do that, however, you need to ensure that data references between various systems are consistent. That requires a level of standardization that only IT can impose. "You need to evaluate the terminology in different data sets to create a common language," says David Loshin, president of Knowledge Integrity, a Silver Spring, Maryland-based consulting firm. It's IT's job to harmonize the data, using data profiling tools (from third-party providers or from capabilities within Microsoft SQL Server) that analyze inconsistencies and recommend improvements. If data is inconsistent, executives will start arguing about which data is correct. "Nothing kills a BI project faster than inconsistency," Loshin says. Inconsistency is also an issue with data entry. Mooney recalls a situation where her firm was analyzing a client's help-desk calls. One of the fields the agents had to fill in was the priority of the call, but many of them were ranked zero. "Nobody could figure out what a priority of zero was," Mooney says. To avoid this type of situation, set up the system so that employees can only enter pertinent numbers or qualities. Executives may initially question the need for these requirements, but the goal of consistent and reliable data across the company should trump their concerns.
 | Silicon Valley-based freelancer Howard Baldwin writes regularly for the Microsoft Midsize Business Center. His work has also appeared on AllBusiness.com and in CIO. |
| 
