Spreadsheets are a wild card in the master-data management equation.
On the one hand, spreadsheets are one of the primary client applications for viewing and working with master reference data from corporate data warehouses. On the other hand, they make it far too easy for users to ignore or flout the MDM controls governing data warehouse operations. Users often populate Excel spreadsheets with information that is outdated, inconsistent and full of errors, and with formulas that deviate significantly from company-approved analytical models.
Spreadsheet governance should be a core component of every organization's MDM and regulatory-compliance strategies. To ensure all users are working from a common set of master data and analytical models, a company should place its spreadsheets under strict MDM version, change and access controls. For compliance purposes, organizations must be able to determine which spreadsheet version a given user relied on at a particular time when taking a specific action. Ideally, desktop spreadsheets should be populated with current, sanctioned data from corporate-sanctioned data warehouses and other official repositories.
In the past few years, the range of commercial spreadsheet-governance tools has grown steadily, though the niche is still far from mature. Vendors in this emerging market include established business intelligence vendors, such as Actuate, and compliance-focused vendors, such as ClusterSeven, Compassoft, Mobius Management Systems and Prodiance.
The Sarbanes-Oxley Act has driven much of this growth, especially where financial master-data governance is concerned. Also, Microsoft inadvertently has fueled the growth of the spreadsheet-governance software market because its stand-alone Excel application lacks such features (though this deficiency is being addressed in the server-based Excel Services feature of the forthcoming Office 2007 suite).
Unfortunately, spreadsheet-governance tools are far from the enterprise mainstream. Today's spreadsheet-governance products are point solutions rather than integral components of comprehensive enterprise MDM and data-warehouse products. Of the dominant MDM and data-warehouse vendors - Business Objects, Hyperion, IBM, NCR/Teradata, Oracle, SAS Institute and Tibco Software - none has added spreadsheet-governance features to its suites. Furthermore, Microsoft, whose SQL Server DBMS is the platform for many data warehouses, has not presented a clear-cut MDM road map into which Excel Services someday might figure.
Another deficiency in today's spreadsheet-governance solutions is lack of integration with tools that enable life-cycle management of predictive analytics and data-mining models. Some vendors of predictive analytics software - most notably, SAS, SPSS and Teradata - provide tools for policy-driven, server-controlled governance of complex statistical models. SPSS and SAS recently released significant new model-governance feature enhancements to support life-cycle stewardship of these business-critical analytical models.
But none of these vendors has expressed interest in extending its servercentric governance environments to encompass desktop spreadsheets, which often contain an organization's most mission-critical analytical models.
It's a safe bet, however, that unified analytical-model governance - encompassing desktop spreadsheets and more sophisticated statistical models - will become a standard feature of most MDM and data-warehouse environments during the next two to three years. It's also a safe bet that many of today's spreadsheet-governance tool vendors will be acquired by the leading MDM and data-warehouse software providers, all of whom tout their ability to consolidate master reference data into environments that provide a "single version of the truth" for operational business intelligence. But as long as corporate desktop spreadsheets are free to display different, inconsistent versions of that truth, MDM's promise will remain unrealized.