BI systems all too easily become victims of their own success. A department establishes its own data mart or data warehouse, word soon spreads, and other parts of the organization want to join in the act.

Like any software development project, if you don't design or plan ahead, your teams will inevitably find themselves constantly reinventing the wheel. Of course, if you're doing that first project, it makes sense to prototype quickly without all the requisite planning. But once you get to projects two or three, that's another story.

Consequently, once your organization has proven the worth of BI, one of the most important components to plan for is data integration. All too often that is considered a separate problem because the owners of the source data or the systems may differ from the group(s) requiring business intelligence – leading to organization, budgetary, or political roadblocks. Consequently, data integration winds up an afterthought or is repeatedly attacked as a unique problem, leading to considerable waste and quality issues.

Not surprisingly, most BI tools have mirrored the organizational stovepipes of their customers, treating data integration as a separate problem from reporting or analysis. For BI customers, the consequences are clear. At the top of the list are the well-documented complexities of integrating multiple tools, including vendor products or homegrown programs.

And, if each BI project is treated as a brand new data integration problem, there will be the need to revisit issues such as mapping, aggregating, and synchronizing data from multiple sources. Furthermore, governance issues such as consistently enforcing policies covering data access, data quality, and audit trails will have to be decided anew. Adding insult to injury, when treated as standalone tasks, the data feeding each BI system may not necessarily be in sync, as extracts and feeds may be scheduled at varying intervals.

The bottom line is that there is an economic price in wasted redundant effort, and a business price, attributable to the fact that all BI systems may not necessarily be literally reading from the same page.

Traditionally, some organizations have attempted to resolve the issue by building enterprise data warehouses. Maintaining such master sources may have sound business justification. However, the creation of an enterprise data warehouse only solves half of the problem: establishing a common source for intelligence data. It does not attack the underlying problem of coordinating the integration of source data and therefore, does not prevent duplication of effort. Only a consistent data integration platform that is an integral part of the BI system architecture can resolve that issue.

The origins of BI stem from the ability to more effectively report on how your organization is performing and why. With the regulatory environment growing more stringent, much of the same BI data used for feeding smart decisions may also wind up repurposed for building the audit trails supporting sound corporate governance, customer privacy, operational safety, or environmental stewardship.

At the end of the day, the success of these efforts drills down to a simple concept: your organization's ability to vouch for the reliability, veracity, and timeliness of its data. That elementary principle is at the root of Sarbanes-Oxley (SOX). Fortunately, if your organization has already been devising standard policies and processes for managing reporting, it may already have a large part of the battle won. In effect, your company may be already practicing data governance.

For most enterprises, compliance is consuming more time and resources than ever. Yet, if your company is already investing time and effort to ensure the veracity of financial data, it gains the added assurance that any reports or dashboards should also be consistent, timely, and accurate. That generates huge competitive benefits that stretch well beyond compliance. Although there's little that any organization can do to sugarcoat the burdens of compliance, the data governance policies that result can provide a critical silver lining in the cloud.

Etta Levine is editorial director for Information Builders' Insights and editor in chief of Information Builders Magazine. To provide feedback go to www.informationbuilders.com/insightstalkback

If you received this message in error, or wish to unsubscribe from this newsletter, please send a blank e-mail to stopinsights@informationbuilders.com.

Information Builders, Two Penn Plaza, New York, NY 10121