Achieving Fully Governed Master Data

This is my third and final post in my series about what to look for in your master data management solution. The first topic was about modeling your master data, the second was mastering the data, and this one is about governing the master data.

As you have read elsewhere in the Kalido Conversations blog, my colleagues Winston Chen, Mike Wheeler and Bill Hewitt have blogged extensively about the need for data governance and how to make your data governance programs operational. MDM is one of the target areas for implementing the remediation part of governance, and it can also be a tool used to engage with the people in your organization who should be involved in that process.

If you are familiar with MDM products, you’ll know that many of them are focused on delivering capabilities for the IT department to develop the model, implement the rules and provide some sort of repository for the clean master data. Most also include some form of data validation capability, and the good ones embed a workflow engine so that the process can be executed without having to write code to link in to an external engine. This is all good for the IT team, but how do you engage with the people who have knowledge about the data, can contribute to keeping it clean, and may even be able to spot issues that the technology didn’t capture?

A key capability to look for in an MDM solution is one that can share information assets quickly and easily with business users through an intuitive consumer interface.  This is the easiest and often the most successful, way to get business users involved in the process of governing master data. Business users ought to be able to search, browse and view master data with a familiar Web-search-style interface. Properly done, this interface is easy to use because it’s just search, so your non-technical business users will know instantly how to use it.  By making it simple, adoption by business users will be faster and easier than if they had no interface at all or are asked to review spreadsheets.

But the interface for business needs to go beyond just browsing, searching and accessing clean master data. They also need to be engaged in the process of keeping master data accurate and up to date – what’s needed is a way to engage the business in the governance process.

Here’s a perfect scenario: a sales rep for a pharma company is visiting physicians and making sales calls. After one visit to a practice the rep discovers that a few doctors have moved out of the practice and are now affiliated with a new hospital, or perhaps one retired – in other words, there was a change. What the sales rep can do, using an MDM business user interface, is request a change to remove or re-associate the affected physicians from that specific practice. The change request would alert the data steward via a workflow notice, who can then make the change, route it for approval, and then publish the updated master data. Marketing now has a more accurate list and can run targeted campaigns accordingly, and the sales reps can schedule follow-up sales calls with the right customers.

Now, while it’s important to make it easy for business people to be involved in the governance process, at the same time you don’t want to rely on humans to do all the remediation work. The second key capability is the MDM technology should provide a way to automatically validate the data coming into the MDM repository by checking against the business rules defined in the model, and matching records from various multiple source systems. If it matches, it gets published, but anything that doesn’t match would need to be reviewed by a data steward or someone with context about the master data – such as the business user mentioned above.

But if you are getting people involved, how do you govern the process of cleaning the master data? This is the third key capability — supporting the business process of managing master and reference data through its lifecycle.  I’ve mentioned workflow a few times, and it is a critical component. Workflow can provide another level of governance by allowing for approvals and authorization through the remediation process. So those doctors that the sales rep wants to reassign? Someone may need to approve that. It’s also likely you may want several levels of approval on certain types of changes, or certain types of data. Any financial master data, or master data related to regulatory compliance, is going to need several levels of workflow-driven review and approvals.

Some tools offer embedded workflow but others don’t or offer only a glancing blow at the requirement. Those that interact with your favorite workflow engine will require more care and feeding to keep the MDM and workflow tools in sync, maintaining compatibility across version upgrades, etc. Embedded workflow simplifies maintenance and can allow for a high amount of flexibility in what you can do.

If you are able to govern your master data in this way, you will improve accountability and stewardship through collaboration and business involvement. The process I just described helps bring everyone involved in master data together for a common goal. The consumer interface is key for bringing business into the process and helps improve data integrity and accountability beyond what the technology alone can do when validating data.  Sometimes we humans are actually smarter than the computer!

0 replies

Leave a Reply

Want to join the discussion?
Feel free to contribute!

Leave a Reply