SAP MDG / MDM: One Instance Vs multiple instances – Food for thought!

[former_member600215]

Hello,

Continuing my blogging journey, here is another blog on SAP Master Data Governance (MDG). (to know more about SAP MDG, visit: https://www.sap.com/uk/products/technology-platform/master-data-governance.html)

If you have missed reading the earlier blog, here is the link: https://blogs.sap.com/2022/10/06/sap-mdg-for-retail/?preview_id=1619569

This blog will attempt to give you some food for thought to consider deployment of more than one SAP MDG instances in a client IT landscape. Recently while working at one of my engagement, we were evaluating this option and I thought why not write about it for others to think.

Ever wondered of scenario where the enterprise landscape makes you think of considering deploying more than one SAP MDG instances? Anyone practicing SAP MDG must have heard of the expression “Single version of Truth” – would you achieve that by having more than one SAP MDG instances running in the IT landscapes?

By now, you probably would be questioning - is this even a valid thought? Why would any organization want to go for more than one SAP MDG instance challenging the very foundation of SAP MDG to ensure there is ONE authoring system?



Well, let me give you some food for thought where deploying multiple SAP MDG instance may sound to be relevant or at least makes you validate as one of the options.

Business scenario:

Product master: An enterprise multiple divisions, each division looks after different product segments. Following are the few data facts:

• The definition of Product master in each division is different – assume approx. 30% deviation.


• Some divisions handle sensitive data and do not want to share this with other divisions.


• Business process differs across divisions – product lifecycle definition varies across each division.

In this scenario, one may give it a thought to see if it makes sense to have more than one SAP MDG  instance. Traditional thinking to go for one instance will of course ask for:

1. Aligning the definition of master data across the enterprise.


2. Streamlining business process to author master data across the enterprise.


3. Defining master data quality rules globally for the product master.

But what if it is not possible to do above because of the nature of the AS-IS scenario and enterprise is not ready to align on these points because of the potential impact it may have on the overall business functions.

This is the point where an enquiry / investigation can be done to validate if this problem can be solved by considering multiple SAP MDG  instance (one per division(s)) – the benefits of SAP MDG can be enjoyed per division resulting cumulative benefit at an enterprise level.

Now, I am not saying that this is a must, and such problem should always be solved with multiple SAP MDG instance. You could still go with one SAP MDG instance by:

1. Slim down product master data definition (to ensure it is acknowledged globally at an enterprise level).


2. Leave the management of difference in the product master definition in the local systems as they do not qualify to participate in the master data definition.


3. Design master data business processes around slim data model.

There are different ways / thought processes to solve such scenario and it needs to be tailored to that enterprise by ensuring they can realize benefits of SAP MDG.

Let me know by commenting what you think of this blog and if you came across such scenario and what did you do to solve it.

Follow this topic page and my profile for the similar contents on SAP Master Data Governance.

Looking forward for the feedback.