[Publisher’s Note: This is the final column from the Data Governance Professionals Organization. Thank you, DGPO, for providing content to the pages of TDAN.com over the past several years. This quarter’s column was written by Malcolm Chisholm of Data Millennium.]
End User Computing (EUC) is, in my experience, usually regarded as a sordid and unsavory activity that respectable members of the enterprise should not indulge in, and which in an ideal state would be completely abolished. It is tolerated, but not approved of, because the enterprise needs it to continue to function.
Perhaps a good analogy is banking in Middle Age Europe, where usury was regarded as sinful and disgusting, but nobody could quite get along without some form of moneylending. Or perhaps it is like smoking (tobacco that is), which is generally frowned upon, but cannot ultimately be prohibited as that would interfere with personal liberties.
What is EUC? It is a process carried out on data by the business at least partly outside of any corporate application, where a corporate application is one supported by IT. Of course, EUC’s must be run on infrastructure supplied by IT and use software packages supported by IT, but the use cases involved are not ones that are solutioned by any corporate application. An important point about EUC’s is that their outputs are used by the business, so they matter.
EUC’s Can Be Both Personal and Departmental
Very often, EUC’s are developed by individuals and run on their computers. However, many EUC’s are in fact departmental. That is, there are several stakeholders involved in their development and they may require coordination among several stakeholders in a department to run successfully. As a result, their outputs matter to more than one business stakeholder.
So, that is what EUC’s are – but who is frowning upon them?
The answer seems to be IT, which is understandable, and Data Governance, which is regrettable. Risk and compliance certainly worry about EUCs, which is to be expected. Let’s focus on Data Governance for the moment.
Data Governance has always tended to have a strong alignment to corporate systems. Many Data Governance units were born in IT. Although they have since migrated to the business, Data Governance units retain something of that legacy. Corporate systems are just easier to deal with, especially with the metadata tools available today. For instance, it is not difficult to reverse engineer the structure of a database, or trace data lineage through ETL or IT maintained SQL scripts. By contrast, there are no inventories of EUC’s, and their technical objects (like Excel) are not centralized and curated in the same way that corporate databases and ETL servers are. Given this, the attitude easily arises that if only EUC’s were like corporate systems, the job of Data Governance would be so much easier. And since EUCs are not, and nobody has a clear picture of the universe of them in an enterprise, EUCs are usually ignored in favor of corporate systems.
Data Governance is Not Aligned to EUC’s
This gives rise to a strange paradox in current Data Governance thinking. It is maintained that data is a valuable corporate resource, and that everyone in the enterprise should be empowered to work with data – but please don’t create any more EUC’s.
To be fair, alternatives are offered. Generally, these require everyone in the business to submit to some corporate technical framework that businesspeople could not possibly learn since they have no spare time after fulfilling their primary duties. Also, one-size-fits-all solutions likely will not cover the wide spectrum of use cases inherent in EUC’s.
Yet, the prospect of empowering more business users to use more data, thereby unleashing a tsunami of EUC’s is real. What should Data Governance do? The first step would seem to acknowledge just how pervasive EUC’s are and how important they are – and how necessary they are. A new approach to EUC’s is needed that will take Data Governance out of its comfort zone with corporate applications. There is a need to support and govern EUC’s, but that is not by abolishing them or somehow reducing them to corporate applications.
“Business Data Processes” is a Better Name Than “EUC’s”
One suggestion to begin with is to abandon the term “End User Computing”, which is almost a term of abuse. “Business Data Processes” or “Business Data Management” are better alternatives. They convey a feeling of empowerment of the business, of something out in the open rather than in the shadows, and something that is solid and reasonable. A vast amount of effort lies in front of Data Governance if it seriously wants to engage with the business – but changing the name would be a good start.
Malcolm Chisholm has over 25 years’ experience in data management, and has worked in a variety of sectors, including finance, pharmaceuticals, insurance, manufacturing, government, defense and intelligence, and retail. He is a consultant specializing in data governance, data quality, data privacy, master/reference data management, metadata engineering, data architecture and business rules management/execution. Malcolm is a well-known presenter at conferences in the US and Europe, writes columns in trade journals, and has authored the books: Managing Reference Data in Enterprise Databases; How to Build a Business Rules Engine; and Definitions in Information Management. He holds the prestigious DAMA International Professional Achievement Award for contributions to Master Data Management and can be contacted at mchisholm@datamillennium.com.
