Data Governance Part II: How to Create a Common Data Vocabulary
Second in a multi-part series on data governance.
I know we’re in the age of big data, but the key challenge facing most global companies I work with is the lack of a common vocabulary for doing business. Internally, companies are a proverbial tower of Babel, communicating at cross-purposes with data and metrics that are inconsistently or ambiguously defined. Internally, companies don’t speak the same language, which undermines their operational efficiency and competitiveness. In short, they lack a strong, vigorous data governance program to manage critical data as a business asset.
Data Driven Yet Suboptimized My clients are data-driven companies with substantial investments in data warehousing (DW), business intelligence (BI), and even master data management (MDM). Even without so-called big data, these companies struggle to manage their “small data”—data that emanates from multiple enterprise resource management (ERP), customer relationship management (CRM) and other run-the-business systems, both packaged and custom, either on premise or in the cloud. The problem is that these companies are fractured organizationally, with each business unit or region managing its own customers, applications, and systems. Although this empowers business units to tailor the delivery of products and services to local customers, it wreaks havoc on global data and, subsequently, the company’s ability to coordinate resources efficiently. In this state of affairs, companies optimize locally and suboptimize globally. This eventually erodes the company’s ability to compete effectively in the broader market.
Local suboptimization. W. Edwards Deming, the father of quality management who was instrumental in converting postwar Japan into a manufacturing powerhouse, advocated for the exact opposite: local suboptimization.
The obligation of any component is to contribute its best to the system, not to maximize its own production, profit, or sales nor any other competitive measure. Some components may operate at a loss to themselves in order to optimize the whole system…. Anything less than optimization of the whole system will bring eventual loss to every component in the system.
W. Edwards Deming, “The New Economics for Industry, Government, and Education,” MIT, 1993.)
To be honest, I think Deming only got it half right. At least in the world of business and data, companies should strive to optimize both local and global processes. This involves establishing global data standards that define the meaning of shared data and processes for managing them. But this requires some short-term sacrifice on the part of the business units. They need to spend time working with their counterparts across the business to hammer out a common vocabulary for the work they share in common.
But, if done properly, data standards don’t constrain business units or hamper their ability to serve customers. On the contrary, data standards foster greater collaboration with other business units with whom they share customers and processes. This make it easier for customers to do business with the company as a whole, which benefits each business unit in the long run, as Deming implies. Without data standards, companies lack a common language to communicate and coordinate effectively across organizational boundaries.
Of course, creating a common data vocabulary is easier said than done. There are two basic approaches to implementing data governance programs in companies: the Executive Fiat and the Trojan Horse. In other words, top down driven by the CEO or bottom-up driven by a strategic initiative or program.
To continue reading this article on SearchBusinessAnalytics.com, click "Continue" below.
Read the next article in this series, "Part III: Two Worlds of Data Governance: Managing Data from the Top and Bottom."