Data Management as a Science

By   /  September 23, 2014  /  1 Comment

Data Management as a Science x300by Jelani Harper

The overall purpose of an organization’s Data Management strategy is sure to vary according to industry and organizational objectives, business requirements and operations procedures.

Despite these variables, there is a common goal for virtually every Data Management initiative regardless of industry, business focus or even enterprise size and resource allocation—to make data trustworthy and dependable enough on which to stake the enterprise.

Doing so requires planning, tactical considerations, governance, and a comprehensive vision for the role that data plays in an organization to single-handedly tie together business interest and IT involvement, upper level management and adherence by even the humblest worker bee.

And, as recently noted by the CMMI® Institute’s Data Management Maturity (DMM)™ model program director Melanie Mecca, it requires a comprehensive, top-down strategy that touches virtually all of the myriad aspects of Data Management: “We feel that organizations don’t have enough in the realm of their overall vision, objectives, and goals. They don’t have an integrative, sequenced plan.”

DMM™

Such foresight, planning, and purpose for a Data Management strategy is just one of the numerous advantages of deploying DMM – a comprehensive, interactive evaluation of a company’s Data Management prowess and efficiency. Facilitated in three weeks by a series of workshops, meetings, and interviews with personnel from the CMMI Institute (which includes Mecca personally on some occasions), the outcome of this assessment provides quantitative scoring in 6 categories and 25 process areas as well as a report of findings, recommendations, and best practices specifically tailored to an organization’s particular results.

Categories and process areas include some of the finer aspects of governance, data operations, data quality, and lifecycle management. Organizations typically require 4-6 weeks of preparation and are instructed in how to improve their management and use of data, while having the option to join a growing partner network of international entities. Most importantly, a briefing of the major points of emphasis in the framework provides insight into the most prominent areas of Data Management itself.

Governance

There are numerous points of emphasis in DMM that reflects the immense magnitude that Data Governance plays in the role of Data Management. One of the six major categories is dedicated to governance—which includes separate stratifications for Metadata Management and business glossaries and definitions—while numerous other categories and sub-categories are related to subjects that are frequently included in governance programs and councils such as data profiling, data lineage, data cleansing, and more.

Outside of the emphasis on Data Management strategy (which is the first of the six major categories), the most pervasive theme throughout this framework is its rigid inclusion of virtually all things related to governance. The implications of these facts are that governance is a necessity for producing the well groomed, trustworthy data that is the objective of Data Management. According to Mecca:

“Data governance is shot all throughout this model. We have context statements as well as specific evaluation statements that tie in Data Governance to everything from data quality to data requirements to managing your providers. But we also have a vertical that can be evaluated stand alone called governance management. That is essentially the activities you perform to make collaborative decisions about the data asset and how you promulgate that through the organization and structure it for your own specific organizational needs.”

Cross-Organizational Collaboration

Another of the six major categories (data operations) is dedicated to the facilitation of cross-organizational collaboration in Data Management. This aspect of the model specifically targets the tenuous relationship between IT departments and the business, which is essential to redress and refine in order to stress that the business owns the data and IT’s role is to provision data for the latter.

In helping to achieve this objective there are three pivotal process areas for this category, including definitions for data requirements and data lifecycle management. It is important to maintain definitions for data requirements in order effectively management various applications. The goal of lifecycle management is to facilitate transparent data lineage; another critical component of lifecycle management is provider management, which focuses on managing and governing data sources.

Big Data

Big Data initiatives effectively represent one of the sternest tests of the organization and efficacy of an enterprise’s Data Management, simply due to the immense volumes and variations of data that are ingested at high speeds. Such ingestion is useless, of course, without effective analytics and applications that can derive and optimize action from Big Data sources. Thus, all of the various aspects of data strategy, quality, governance, and metadata must be firmly in place and appropriately updated to maintain such initiatives and leverage the value of Big Data technologies. Mecca revealed that:

“At various firms I’ve asked Big Data experts, ‘why don’t you run me through the model and tell me if there are any fundamental practices that we’ve talked about that have any gaps’. So far the answer has been no because Big Data is a specific type of implementation. What they have said is if you do all of these things [in the model] well, a Big Data implementation is going to be that much more successful with more accurate data, with good data standards, with agreed upon business meanings, etcetera.”

Back to Governance

With so much of DMM focused on governance and well-defined roles and responsibilities pertaining to data and their applications, there are a couple of salient caveats that organizations can take from DMM regarding Data Governance that can readily assist them.

  • Bi-Directional Communication: Effective governance hinges on communication from both sides: the business and IT, Data Stewards and end users, upper level management and worker bees. It cannot be one-sided; the need to refine and update governance procedures based on real-life usage and applications of data requires such communication, which serves as the backbone of governance.
  • Subject Area Approach: One of the best ways to avoid application silos facilitated by tool selection for specific departments are to take a subject area approach to governance and tool selection. Doing so reduces the risk of individual departments utilizing their own tools and maintaining their own data separate from other departments, especially when that data can influence or assist those departments.
  • Data Quality: Data quality is perhaps the ultimate boon of proper governance. Nonetheless, it involves several different considerations including its own strategy, data cleansing, and data profiling. The key is to keep data profiling measures consistent and well regulated while eschewing ad-hoc profiling, which has the tendency to defy standards.

“This is an evaluation instrument for the strengths, gaps and integrity of the actual behavior of the organization: the practices that are being performed and the ones that are not, from discovery, and the work products that support those practices,” Mecca disclosed. “We have 300 practice statements in the model and over 300 work products assigned to these practices and process areas by level of capability.”

Solidifying Data Management

DMM was created by various members of the CMMI Institute who have decades of experience in Data Management; the various components of the model had to pass a rigorous peer review process comprised of approximately 70 reviewers. Those who utilize the framework will have options for certification in different areas of Data Management.

Most importantly, DMM represents a cohesive attempt to quantify and measure an organization’s efficacy in Data Management, and provides solutions and recommendations to help it do so. Such an effort represents a formalization of Data Management as an objective field, and one which reflects the growing trend towards data-influenced practices in the modern world.

 

You might also like...

Case Study: Unilog Leverages Data Preparation to Monetize Upstream Data

Read More →
We use technologies such as cookies to understand how you use our site and to provide a better user experience. This includes personalizing content, using analytics and improving site operations. We may share your information about your use of our site with third parties in accordance with our Privacy Policy. You can change your cookie settings as described here at any time, but parts of our site may not function correctly without them. By continuing to use our site, you agree that we can save cookies on your device, unless you have disabled cookies.
I Accept