Should our data architect be an active participant in data management projects, such as choosing a new data quality tool or creating a data governance process? Are there any benefits to having a data architect be part of those task groups?
Your data architect should be involved in most, if not all, data management projects. It sounds as if there is only data architect in your organization, thus making his or her role in overall data management efforts even more important. If that is in fact the case, your data architect should have a “big picture” view of the data needs of the entire enterprise.
The operative word in this context is architect. You of course would want your data architect to be involved in most data management decisions, in the same way you would want an architect to be involved in major decisions about a building. For example, a building architect typically wouldn’t play a role in most day-to-day operations, such as building maintenance – but he probably would have some helpful recommendations about how the building should be maintained for an optimal life span.
Regarding the process of choosing a data quality tool, your data architect should have a good understanding of the data quality issues that the organization faces and must be able to address with the selected software.
For data governance, there definitely should be a data architect involved – one who can bridge the business and technology worlds from a data/information perspective. While a data architect typically isn’t a voting member of a data governance board, the architect will need to be depended upon to provide valuable input and raise awareness of significant issues requiring data governance and/or data stewardship attention.
Since you’re asking this question, I wonder if you have the right person as your data architect or if your organization doesn’t adequately understand the role of a data architect. If you define that role only as developing data models, the person who holds the job should really be called a data modeler, not a data architect. Of course, a data architect does do or oversee data modeling but also does a lot more.
The data architect plays a critical role in determining architectural approaches for data environments and helps ensure that the data needs of the enterprise are being met. For example, how and when should data be centralized rather than federated? For a master data management (MDM) program, should a repository or registry approach be taken? For data warehousing, what components will be required (e.g., an operational data store, an atomic data warehouse, data marts, cubes, real-time partitions)? How does it all tie together? Many enterprises are realizing that the data layer is the optimal layer for achieving integration. Your data architect should have the experience, knowledge and vision to facilitate enterprise integration via the data layer.
This was first published in June 2011