Data that is getting garbled, mismanaged, misinterpreted, or in some cases simply lost can be very expensive. When there is no reliable source of true information related to customers, products, prices, and vendors, consequences can include: regulatory fines for systems and control failures, labor expenses incurred from extra effort required to produce accurate information for critical business decisions and legal fees spent cleaning up a public incident related to data errors.
Master Data Management
MDM concentrates on master entities that are, as defined by the business, more important than others because they are widely distributed across the enterprise. The Master Data Management enables a business to establish a "single version of the truth." It is not restricted or confined to any particular industry and can cover a broad array of corporate related data. Some examples of common master data entities are prospects, customers, products and employees.
Symptoms that your Master Data Management Strategy Needs Attention
Common symptoms include:
- Poor data quality is probably the most common and obvious symptom. The first sign of inadequate or complete lack of MDM is duplicated or inaccurate/inconsistent data. Bad data will impact both your internal and external customers, beginning at the source and impacting every system and user that interacts with that data along the way. Eventually this poor data will surface in business reports and metrics that present an inaccurate view of reality.
- Other symptoms may surface in the form of regulatory fines for systems and control failures, to labor expenses incurred from extra effort required to produce accurate information for critical business decisions, to legal fees spent cleaning up a public incident related to data errors.
- The existence of manual business processes that directly result from not having reliable and consolidated data. For example, an IT Engineer spending large amounts of time every day working on ad-hoc reporting requests that requires taking data from multiple sources, modifying and manually consolidating into a single view.
- Having to update data in multiple places to ensure it remains in sync. For example, Nothing else about that product needs to be updated, just the name. However, due to lack of MDM, that product record exists in 5 different systems, so you have to access the system, query for the record, make the update and take care to ensure you get it exactly right. If there was a master entity, you could simply make the change once to that record and have it replicate to all downstream systems.
What Makes MDM Difficult?
MDM is difficult because it typically only becomes a priority when issues and problems arise, rather than being a disciplined process from day one. Therefore, a major activity is the cleansing of the data to get it back to a pristine state which can be very difficult and time consuming.
Building a business case for MDM with quantifiable benefits can also be a challenge. MDM efforts will have multiple stakeholders so buy-in can be difficult. Departments often operate in silos and what is a priority to one may not be to another. In addition, the budget required to finance the effort may need to come from different departments throughout the enterprise. And MDM isn't a one-time project, but instead should be an on-going effort. Even if the initial project is sponsored by IT, over time the responsibility should be spread across the organization.
What Results Can You Expect to See?
Although master data entities may comprise only 5% or less of an enterprise data model, their significance from the information value and data quality perspectives is disproportionately high. Bringing master data into order often solves 60%–80% of the most critical and difficult-to-fix data quality problems.
Good master data management will make a positive impact on many of the most important enterprise challenges and risks, such as:
- Ensuring compliance with rules, regulations and contractual obligations
- Protecting sensitive and private data
- Improving customer satisfaction
- Reducing manual processes and rework
- Reducing costs
- Managing risk
MDM is an On-Going Priority
Companies often initiate MDM as the result of a crisis and once that crisis is resolved, interest in the effort tends to subside. Big data is more than just a trend in the market place, it is a reality, and so companies should never treat MDM as a one-time effort but instead should see it as an on-going priority.
It typically impacts everyone including the customers, the users, the developers and even investors. Because IT owns enterprise data, funding for an MDM effort will typically come from its budget, at least initially.
But over time, the business should take responsibility for data quality and assume joint ownership of the program.
A Single System of Record
Systems will share data, but there should be a single system of record for the master data entities. This will ensure the single version of the truth concept we discussed earlier. That's not to say data cannot be updated in other business systems, but the systems must be architected in such a way that those updates are made to the master data record so that data can then be shared with all other interfacing systems that need to consume and surface the data.
Through Tribridge’s IT excellence solutions, we develop MDM strategies and work with customers to implement the right strategies for effective master data management. To learn more about how your organization can develop and implement a Master Data Management Strategy, reach out to Tribridge for help.