Establishing a strong foundation of data governance is important for a master data management (MDM) initiative to succeed, as I outlined in my last post. But it’s not enough. In addition to a governance model, we have to set the right expectations about what MDM is and what its implications are.
Often touted as a silver bullet technology that will solve your organization’s data problems, MDM is actually neither a silver bullet nor a technology. In fact, there are several myths about MDM that mislead organizations to believe that MDM starts delivering value right out of the box.
1. MDM is a tool or a technology
Let’s get rid of the notion that a vendor’s MDM “product” will solve our customer data problems. MDM is less about technology driven by IT and more about governing master data, establishing processes, clearly identifying business needs, and business engagement. The tools and technologies become most effective only when this disciplined thinking is in place, which then also results in greatly enhanced utility of your master data.
2. MDM is about data integration
Vendors often tout their product’s ability to integrate master data from different source systems into a common hub. They also brag about how their product is great at creating a “golden record” of the customer. While consolidation methods are important, MDM is less about creation and more about how that data should be used, shared or repurposed. It’s more important to define proper usage scenarios (e.g., “I need to know which of the customers I plan to target have reported product defects”) and then ensure that your integration strategy can support such a usage scenario.
3. MDM is a project
MDM should not be viewed as a “boxed” project with a specific set of deliverables, which, when completed, will give you a fully functional MDM environment. Instead, treat MDM as a long-term program whose objective is to transform the way your organization views and uses information to meet business objectives. This transformation should be about:
- Developing information to be a business asset that provides a unified view of your customers.
- Using that view to gain insight about your customers, their interrelationships with other domains and their impact on your business processes.
- Exploiting those insights to optimize your business processes.
Since businesses are in a constant state of flux, all these transformative aspects need continuing investment in people, processes and technology.
4. MDM is synonymous with the data warehouse
Though there are similarities between master data domains and data warehouse (DW) dimensions such as customer and product, MDM provides much more than just a set of dimensions to slice and dice your transactional data. MDM resides within its own hub and provides a set of services through which various operational and analytical applications (the DW included) can subscribe to master data, thus eliminating data inconsistencies and information silos.
5. MDM is just a data quality initiative
A DQ strategy is one of the first steps in your MDM program execution. Tools and activities such as data profiling, cleansing, standardization and monitoring of DQ metrics can help measurably improve data quality. But MDM is not only about data quality. The reason to execute MDM goes beyond improving the quality of your data. It is also fundamentally about data usage policies, data stewardship and metadata management.
Dispelling these MDM myths will help you set the right expectations with your sponsors by tying the program goals more closely to your business objectives and thus, improving the likelihood of the program’s success.