The Business Case For (Semantic) Data Management

Executive Workshop on Strategy, Governance, and Semantic Standards

The business case for data management is crystal clear. Data is your most valuable resource. It describes your processes, your customers, your supply chain participants and your market environment. It is an essential factor of input into every aspect of operations that we’ve allowed to become isolated, incongruent and inflexible. It doesn’t have to be this way. This is a problem that can be solved.

The pathway to solving the incongruence and structural rigidity dilemma is to combine the principles of data hygiene with semantic standards that allow for granular meaning to be captured and exchanged across your organization. It is a story that is both simple and elegant – even though it has taken me almost four decades as the scribe, analyst and therapist for the information industry to assemble and articulate.

Michael Atkin constructed this executive workshop to present a clear rationale for semantic data management as a business imperative. It is presented as a complete business case in concise language without technical jargon. I will present use cases where semantic standards are most valuable, balanced against the costs and obstacles associated with organizational alignment and program implementation.

This is a course for those responsible for the development of data strategy … those involved in positioning the data management value proposition to business and technology executives … and those responsible for converting data from a problem to be managed to a resource to be employed. We have designed it as an interactive workshop supported by stories and anecdotes that illustrate the benefits of information literacy in the digital world.

Mike has been the analyst and advocate for data management since 1985. His experience spans from the foundations of the information industry to the adoption of semantic technology. Mike has served as an advisor to financial institutions, global regulators, publishers, consulting firms and technology companies on the principles, practices and operational realities of data management.


Course Synopsis (and key takeaways)

Segment 1: Background

  • Data processing is not the same as data meaning
  • Information intensive industries share common data challenges
  • Data incongruence and structural rigidity are solvable
    (and overwhelmingly worth it)

Segment 2: Principles of Managing Meaning

  • Data is a representation of real things
    (obligations, commitments and research parameters)
  • The ‘prime directive’ of trusted data without reconciliation
    or manual transformation
  • Meaning can be modeled at a granular level via standards
    for content expression
  • Remember the ‘semantic trinity’ – identify, describe and express

Segment 3: Causes and Implications of Incongruence

  • Content must be precise and understood in context
  • Technology fragmentation and structural rigidity are the
    drivers of data incongruence
  • Data has been modified, transformed and renamed many times. Database proliferation makes the problem worse
  • Business and functional silos are customized for specific applications (proprietary models)
  • Incongruence and rigidity cause business frustration, make it hard to link processes, drain resources and stifle innovation (bad data tax)

Segment 4: Semantic Capabilities for Executives

  • Complex data relationships are hard to manage with
    relational technology (conventional technology does not facilitate scenario-based analysis)
  • Some of our ‘golden source’ data priorities are false and contribute to the problem
  • Semantic standards were designed specifically to handle complexity
  • The value proposition for semantic standards is overwhelming (automated quality, concept reuse, lineage, integration, automation, governance, auditability)

Segment 5: Costs of Conversion

  • The biggest barriers to adoption are cultural, not technical
  • The physical infrastructure requirements exist but are not significant
  • Foundational and domain-specific ontologies are required
  • New skills are needed to operationalize knowledge technologies

Segment 6: Core Data Management (data hygiene)

  • Baseline capabilities (i.e., policy, accountability, funding, inventory, systems-of-record, lineage, identifiers, glossaries, metadata management, mapping, etc.) are essential
  • Firms must still manage the full data ecosystem including coordination with business, technology and the control functions.
  • Semantic standards simplify, but do not replace the need
    for governance

Segment 7: Messaging

  • Education of key stakeholders is a prerequisite for success
  • Semantic champions need to speak to senior management
    in the language of business (with verifiable metrics)
  • We need to talk less about how it works and more about
    what is possible from adoption