Follow Datanami:
June 27, 2016

Transcending Regulatory Compliance: Creating Business Value from Data Governance

Marty Loughlin

Data governance has emerged to the forefront of financial services ever since this vertical became inundated with increasing and evolving regulations, penalties, and regulatory entities at the end of the last decade. Stringent compliance requirements not only mandate that organizations provide accountability for data, but traceability, provenance, and auditability as well.

The most common response to these drivers is to build silos for specific regulations from the labyrinth of information management systems deployed throughout an organization, which frequently results in:

  • Point solutions with limited life spans and viability
  • Different answers from different sources to the same question
  • Obfuscated data quality and lineage
  • Lack of agility for additional regulatory and business requirements

Emerging technologies like semantic technology are transforming compliance and data governance to not only circumvents these issues but increases the yields from governance to affect nearly every aspect of data-driven processes throughout the enterprise. By embedding governance in both data management and business functions, organizations can transcend mere regulatory compliance to vastly improve fraud management, product development, customer 360 views, and much more.

Implementing the uniform policies and practices of governance with semantic technologies ingrains them within business functions and supporting IT systems at such a granular level that well governed, trustworthy data becomes an implicit by-product of simply using data.

Growing Industry Standard

The model-driven, standards-based approach of semantics for data governance is rapidly becoming an industry-wide normative throughout the financial vertical. Some of the most pervasive and pioneering efforts in this vein have been orchestrated by the Enterprise Data Management Council (EDMC), an alliance of private companies, vendors, and public sector experts who are deploying semantic technologies to account for the influx of regulations and requisites for data-driven practices in this vertical.
metadata_man

The crux of the semantic approach is in leveraging models that describe data and their requirements. Those that pertain to governance include metadata, attributes, regulatory requirements, stewardship protocols, role-based access, and others. Virtually anything about data and their use can be described via these models, which also map to business glossaries and vocabularies to ensure the uniform continuity of terms and definitions that is necessary for effective governance.

The EDMC is nearing completion of the Financial Industry Business Ontology (FIBO), an industry-wide semantic model for concepts and meaning shared between organizations and regulatory entities. FIBO is intended to harmonize data across sources to streamline the regulatory reporting process so those deploying it get the same answers to the same questions every time, regardless of where their data stems. The EDMC has also created the Data Management Capability Assessment Model (DCAM), which provides specific definitions for governance facets, their criteria, and benchmarks for which semantic approaches can suitably prepare organizations.

Supplanting Silos with Smart Data Lakes

The value of a semantic method for governance not only includes harmonizing meaning across the financial industry, its organizations, and their various data management systems, but also in readily describing aspects of data and their significance in business terms. This fact is one of its most crucial advantages because it broadens the business audience for the consumption of data. Moreover, when one combines this boon with the amalgamation of semantic technologies and tools (such as semantic data lakes), governance and its advantages truly shift from regulatory compliance reactions to proactive means of expanding the business value of data. The utilization of semantic data lakes typifies these benefits that a standards-based approach to governance provides.
Enriching conventional data lake repositories (such as Hadoop) with the aforementioned semantic methods enables organizations to replace individual silos with a single repository for all their data, which are semantically tagged according to relevant factors such as governance policies, requisite access, consistent metadata, definitions and attributes. By ensuring that all data adheres to what is actually an evolving semantic model that readily expands to include new data types and sources, organizations now have a coherent way to manage them in accordance with governance protocols.

Creating Business Value from Governance

shutterstock_growth_chart_Bakhtiar Zein

(Bakhtiar Zein/Shutterstock)

The possibilities for extending the value produced by governance into areas of the enterprise beyond regulatory compliance are nearly unlimited. One of the most convincing examples involves transformation, a key facet of data preparation. Non-semantic approaches to this common prerequisite for analytics involve writing code for data. Semantic technologies can incorporate transformation processes into their underlying models to generate code automatically. Thus, governance models can directly impact operations in a way that is more expedient and streamlined than other methods are. There is a reduction in complexity in the number of steps it takes to ingest data, prepare them, and ultimately analyze or create action from them—resulting in fewer points of error and complications for governance practices.

This holistic method of managing data extends g
overnance’s chief output—quality, trustable data—into whatever use case the enterprise chooses to apply it. The evolving semantic model at the core of these technologies facilitates agility that can readily respond to future regulatory requirements as well as business ones. This proactive semantic approach to governance is also directly related to analytics, since users can ask any question of their data that they want. The underlying technologies are able to glean the relationships between data and their elements in a way that is much more nimble, and less time-consuming, than relational models are.

Audibility and Lineage

The immediacy of the value that enterprises gain from a semantic approach to data governance is heightened by the myriad regulations surrounding finance. Specifically, a standards-based environment provides the sort of provenance for data lineage purposes that is essential for auditing for regulatory compliance. Mapping to regulatory requirements and implementing role-based access to them provide a data lake_4compliance framework; the ease of data provenance in a semantic environment ensures adherence with a striking degree of specificity. Organizations can trace the use of data in a variety of different formats to assure regulatory compliance of data’s transmission or deployment.

The enterprise can link its semantic model to a wide number of forms of communication including trades, email, online activity, phone records, messages, internal systems and others to trace data lineage according to regulations. The combination of data provenance, regulations mapping, and role-based access to data based on regulations provides a holistic way of determining and evaluating adherence to regulatory compliance. Moreover, these facets of compliance are facilitated with the comprehensive approach of the semantic model that is the centerpiece of a standards-based method for data governance.

Opportunity for Advancement

Initially, data governance was viewed as a means of protecting enterprise assets from the penalties associated with non-compliance. A semantic approach to governance transcends this utility by not only including it, but also numerous other advantages stemming from the reliable data quality begat from this discipline. The harmonization of meaning that is central to the implementation of semantic technologies for governance greatly increases enterprise-wide agility. Furthermore, the capacity to describe data’s significance in business terms strengthens the connection of the business audience to the resources that give it competitive advantage. All of these factors, as well as the efforts of the EDM Council and the growing pervasiveness of a standards-based approach, attest to the reality that with semantics, governance has become an enabler of business value.

About the author: Marty Loughlin Marty Loughlin is Vice President, Financial Services at Cambridge Semantics Inc. Prior to joining Cambridge Semantics, Marty was the managing director for EMC’s consulting business in Boston. His 25- year career has focused on helping clients leverage transformative technologies to drive business results, most recently in cloud and Big Data. Prior to joining EMC in 2005, Marty was co-founder and COO of Granitar, a web consultancy that grew to 250 people in four years and served clients such as State Street Co., The New York Times and Standard & Poor’s. Marty began his career in Ireland with Digital Equipment Co. He holds a bachelor’s degree in English from Dublin City University and a high-tech MBA from Northeastern University.

 

Datanami