Data Management 2012: What's In; What's Out
With regulatory reform dominating 2011 and likely to continue dominating 2012, transparency, global legal entity identifiers, managing big data, evaluating the maturity of the data governance process, and managing systems of record were among the top forty three scenarios that will become popular in the data management arena in 2012.
So says a report just issued by the Enterprise Data Management Council, the trade group representing the largest data management experts on both sides of the pond.
Among the top forty three unpopular ones: disclosure; global location number; data overload; operational risk management and golden copies. The goal of all regulators is to eliminate systemic opacity -- the inability to view which financial entities are trading which types of securities or contracts when, for how much and with whom. Ensuring that data is transparent is the task of the newly created Office of Financial Research, which the U.S.' Dodd Frank Wall Street Reform Act has designated the organization to measure systemic risk.
Transparency and disclosure may sound synonymous, but they are quite different, explains Michael Atkin, managing director of the EDM Council in New York. Transparency is a far more sophisticated form of disclosure where regulators, investors -- and financial firms -- themselves understand the source of the data and can interpret just what it means. Doing so will require the use of operational ontologies, semantic definitions, and data lineage to reduce the potential for inconsistent, duplicative data in multiple silos and retain an appropriate audit trail. Relying on a golden copy located in a centralized repository to feed downstream applications won't be enough.
"There is really no such thing as a golden copy because that would imply there is one version of the truth," says Atkin. "In fact, there are multiple repositories of data aligned with systems of record so firms need to understand the systems where the data is located and the different versions of the same information which diverse business units consume." That data includes not only descriptions of financial instruments traded and processed but also large sets of unstructured data such as news or other important information. The category also includes tick data which firms must store and analyze in real-time to split-second -- if not faster -- investment decisions and trade executions. Firms must find ways of managing with the growing volume of data to avoid being overwhelmed by a tsunami of data.
To understand just who is executing those orders and for which customer or with which counterparty, the OFR is also promoting the need for standardized and globally accepted legal entity identifiers rather than a hodgepodge of codes to track the business entities buying and selling financial instruments or contracts. The ID codes, now under review by the International Organization for Standardization, is supposedly better than a "global location number" promoted by the GS1 organization best known for creating barcodes for the retail industry.
To ensure that firms can properly manage all of their data to comply with regulatory initiatives such as Dodd-Frank, financial firms also need to measure their current level of efficiency. Rather than tackling the problem in a siloed approach -- through each business line -- to reduce operational risk, they should adopt a data maturity model. Such a model, now being constructed by the EDM Council and Carnegie Mellon Software Engineering Institute, would define the components of effective data management so that firms could compare themselves against documented best practices. Firms would be graded on a scale of one to five depending on how well they perform in data management strategy, data operations, data platforms and technology, data quality and data sourcing. A score of five is optimal while a score of one means the firm has disorganized procedures for managing data.
Here is a rundown of the top 43 ins and outs for data management for 2012 as outlined by the EDM:
WHAT'S IN:
Transparency
Financial Stability
Market Surveillance
Operational Ontologies
Capability Measurement
Governance Implementation
Total Cost of Ownership
Links and Relationships
Regulatory Rule-Making
Legal Entity Identifier
Attribute-Based Classification
Interconnected
Big Data
Granular Reporting
Open Source Identifiers
Semantic Definition
Meaning
Natural Language
Data Visualization
Legal Facts
Flexible Architecture
Data Manufacturing
Data Comparability
Information as an Asset
Strategic Data Management
Added Value
Data Quality Rules
Data Management Maturity
Data Harmonization
Utilities
Regulatory Collaboration
Incremental Migration
Data Alignment
Data Lineage
Requirements Management
Business Alignment
Data Lineage
Requirements Management
Business Alignmnent
Data Profiling
Front-to-Back Lineages
Exception Management
System of Record
Client Benefits
Risk and Compliance Business Case
Office of Financial Research
WHAT'S OUT:
Disclosure
Systemic Risk
Market Abuse
Data Models
Performance Metrics
Governance Structure
Return on Investment
Structure
Legislative Objectives
Global Location Number
Hard Coded Classification
Aligned
Data Overload
Aggregated Reporting
Proprietary Identifiers
Schema Definition
Labels
Message Syntax
Data Analytics
Data Structures
Legacy Systems
Data Remediation
Data Mapping
Data Importance
Tactical Data Management
Competitive Advantage
Duplicate Sourcing
Operational Risk Management
Data Workflow
Shared Services
Regulatory Mandate
Rip and Replace
Cross Referencing
Data Measurement
Requirements Capture
Top Down Management
Data Cleansing
Transformed Truth
Find and Fix
Golden Copy
Value to Business
Logical Business Case
Systemic Opacity
Written by Chris Kentouris, Editor-in-chief (Chris can be reached at Chris.Kentouris@hotmail.com)
