Master Process Before Mastering Data

It is critical to consider where data is being produced and consumed, and truly understand those processes.
Many of our customers approach master data management as a continuous improvement process. The idea is to start with a goal that is both attainable and valuable. Future projects will build on top of the successes from earlier projects.
That could mean mastering an initial set of data sources from a handful of departments first, and then adding additional sources in new lines of business later. It could also mean mastering a single domain of data, such as an organization, and later mastering related entities such as customers or products.
Over time, the MDM initiative becomes more and more valuable, as the reach and richness of the master data improves.
There is a missing piece in the picture so far. To maximize the ROI of the MDM solution, related business processes must be reevaluated. It is critical to consider where data is being produced and consumed, as the master data is essentially a point in time reflection of a business process.
Let’s consider some scenarios where the business processes of the enterprise have not caught up with the master data infrastructure:
Sales and CRM
A sales representative is introduced to a new prospect. The rep follows up the meeting by searching for the prospect’s information in the CRM system. No results are returned, so the rep creates a new account in the CRM system. After the initial meeting, the prospect does not return any phone calls, and eventually this opportunity is closed in the CRM system.
All along, the rep did not realize that the prospect works for a subsidiary of a parent company already in the CRM system. All of the contacts and historical information built up about the parent company over years were not tapped into at all.
Persons of Interest
A federal investigator has identified a person of interest for a criminal case, but does not have sufficient evidence to prosecute. Several weeks later, a detective in a different department finds the same individual in the search results while pursuing a different case.
Both investigators have information that would help each other. Since there is no notification process in place, they are unable to connect the dots in two related cases.
We've Got the Data, But Not the Process
In each of these scenarios, the master data exists in the hub but the business processes around the master data have not adapted. The business analyst or data analyst is vital to identifying and resolving these issues.
How are duplicate records being introduced into the source system data? Why are records with invalid or missing attributes created? Who are the possible consumers of the master data, and how will they be consuming it? Driving resolution around these questions will ensure that the full benefits of the MDM initiative are being captured.
In my next post, I will follow up with practical suggestions to extend and enforce the right processes. We will discuss when process reengineering and training are sufficient, and when is it appropriate to consider existing application changes and even developing new applications that rely on MDM.
Leave a Response







Entries(RSS)