Enterprise Information Management (EIM) is receiving increased attention from CIOs and other business leaders.  Executives see the value of a shared partnership with the business and the importance of data stewardship and data governance.  The field of information management is maturing, and many organizations are making remarkable progress towards managing information as an enterprise asset.

Now that the leadership’s attention has been directed to data management, EIM professionals face several critical challenges that require further focus.  Without consciously addressing these challenges, EIM leaders and practitioners risk losing management’s attention and the credibility of this profession.  Following is a short list of the major issues data management professionals must resolve to make EIM a permanent, respected, and contributing function in any enterprise.

  1. EIM Leadership – EIM programs must integrate and coordinate all related data management efforts.  This is of utmost importance to the success of today’s CIO, who needs a companion responsible for this leadership.  A Chief Data Officer (CDO) might be a VP in some organizations, or a director in others, but this person should report directly to the leadership as a peer to similar executives responsible for managing the technology infrastructure and the application portfolio.  A direct reporting relationship to the leadership ensures continued attention and accountability.  CDOs must be passionate about the importance of information management and sufficiently knowledgeable to be successful in the organization.  An effective CDO will get key messages to senior management.  Most data architects, team leaders, and first-line managers never get that opportunity.
  2. Sustained Data Governance – Most data governance programs are dependent on the leadership of one or two key individuals.  These people inevitably move on to other opportunities, and the data governance initiatives they started inevitably die on the vine.  Sustainable data governance programs depend on sustained executive sponsorship, collaborative leadership, clear data stewardship roles and responsibilities  Additionally, strong data governance requires dedicated support staff to facilitate activities, productive and well planned meetings that optimize business data steward time commitments, effective communications, and repeatable data governance processes (i.e., business data definition, quality requirements definition, policy and standards review, issue tracking and resolution).
  3. Information Valuation Techniques – Very few people have been able to quantify the business value of information.  Current accounting practices consider information as intangible assets, estimating the business value of enterprise information assets within the general balance sheet category of “Goodwill.”  Project funding is dependent on estimating the business value of information access and information quality.  The more successful organizations have captured anecdotal “value stories” from respected business subject matter experts and extrapolated savings from a single circumstance based on its enterprise-wide prevalence.  Until accounting practices mature, EIM and data governance professionals need to develop their ability to find, capture, and extrapolate anecdotal business value.  If any EIM topic requires creative thinking and intellectual thought leadership, it is the need for practical techniques for data valuation.
  4. Management Metrics – There is little consensus today about what measures should be captured and what metrics should be tracked to manage data and the EIM function.  As noted above, business value measures are the most difficult to determine (what increased business value can be attributed to the EIM program due to data standardization and improved data management discipline?).  Compliance and conformance measures evaluate the level of adoption, regulatory compliance, and conformance to standards introduced by the EIM program.  The earliest measures available are activity and participation measures, indicating the level and breadth of acceptance, organizational coverage, and sustained commitment.  For a while, these measures will have to suffice – but for how long?  The EIM community would benefit from a portfolio of practical measures and graphs from which to select in building an EIM scorecard for senior management and data management professionals alike.
  5. Cost-Effective Metadata Management – Large organizations with significant resources have invested in building enterprise-wide Managed Metadata Environments (MME), but the cost of the repository software tools, support staff, skills, and process implementation has been far too restrictive for most organizations.  Easy access to high-quality, integrated metadata is essential to managing enterprise information assets, improving information quality, integrating data across applications, and supporting informed business intelligence.  Most organizations lack creative, cost-effective solutions that provide an immediate return on modest investments.
  6. Focus on Metadata Delivery – Metadata repository implementations begin by focusing on capturing and integrating metadata from multiple sources.  Continuing this focus without paying attention to how to provide business value through metadata access and delivery has led many repositories to be characterized as “roach motels” – metadata goes in, but never comes out!  Wouldn’t it be wiser to focus on leveraging an initial scope of metadata to the fullest extent possible, demonstrating business value?  Frankly, most metadata repository software is not easy to use.  Successful MMEs pursue a wide variety of access and delivery channels.  Metadata administrators must worry about meeting the real needs of business and technical users, and focus attention on enabling ease of use and publishing.  Business glossaries can be fed from the repository, or they may be web pages, reports, and documents that provide an easy-to-reference “retail” alternative to direct query and reporting against the technically intimidating metadata repository.
  7. Information Architecture Integration – While an enterprise data model is the heart and soul of any information architecture, there are other valuable components, including:
    • Information value chain analysis (CRUD matrices identifying the relationships between data and process, data and organizational roles, data and organizational units, and data and application systems)
    • Information supply chain analysis (data flow diagrams tracing information products as inputs and outputs of business processes)
    • Reference data sets – standard code values, their hierarchies (taxonomies), and associations (cross-references)
    • Semantic ontologies, integrating these closely related models with the enterprise data model, provide a consistent way of looking at both structured and unstructured data across the enterprise
    • Meta-models and the MME (metadata integration) architecture
    • Data integration architecture, including the master data management (MDM) architecture
    • Data warehousing and business intelligence/analytics architecture

These related models should be integrated and internally consistent, linking application business objects and semantic web services to the enterprise data model (itself a semantic model), and linking data with business processes and other elements of enterprise architecture.  It is important to share perspectives across factions, reconcile semantics, and speak with one clear voice.

  1. Information Architecture Usage – EIM professionals need to ensure the information architecture is used in business planning, IT planning, application and project portfolio management, project scoping, requirements analysis, and application design.  Cynics in an organization already consider the information architecture to be “shelfware” – prove them wrong.  Business and IT planners are not likely to see how architecture may help them immediately.  Data architects need to promote and demonstrate how to put information architecture to good use.
  2. Expanded Business Intelligence Support – If “it has been built” and they still have not come, it is time to provide improved support to the various user communities with an analytics strategy.  Casual users need an easy-to-use portal that accesses a library of standard periodic reports and parameter-driven real-time reports.  Power users need training, access to business metadata, and guidance from BI support specialists.  The true data-intensive professionals in a company will benefit from advanced analytics, including statistical analysis, data mining, and predictive analysis.  All these technologies require user training and continuing support from helpful BI specialists with “people skills.”

Conclusion

There may be other critical issues to resolve in an enterprise, but this list seems comprehensive for managing data and information as enterprise assets.