Gartner Blog Network


More on ERP and MDM – apples and oranges?

by Andrew White  |  January 28, 2009  |  7 Comments

I received a newsletter from Hub Solution Design, an MDM services vendors, and in that newsletter was an invitation for a webinar hosted by Siperian, an MDM vendors, and Hub Solutions.  The title of the webinar was, “Top 5 Reasons Not to Master Your Data in SAP ERP”.  The invitation said that the webinar will address five questions:

 

  • Why SAP ERP is not the right place to master data
  • Why a separate MDM system is required for streamlining business operations
  • How MDM and SAP ERP coexist
  • The technical attributes, strengths and weaknesses of SAP and Siperian MDM products
  • The requirements of an effective MDM system and best practices for implementation

 

Firstly, these are good questions.  I get these questions myself frequently, though the vendor names change for the MDM and Enterprise Resource Planning (ERP) solution.  I had to blog on this topic since the questions may not be as clear, or complete, as they might be.  Not surprisingly, they are intended to motivate you to attend the webinar (like any good marketer should do) so I can’t fault the vendors for that.

 

ERP might be a good place to master data.  As always, it depends on a number of things.  For some firms ERP might be the name given to the unified stack (one data model, one process model) of business applications where master data objects such as customer, product, or location are first created.  As such, ERP is the source, and may continue to be.  Perhaps IT and the business has no need or desire to disinvest in ERP; perhaps the level of consumption of this master data outside of the ERP domain is very low.  The real question for the user is this: where is the right source of master data for you?  For some it could be ERP, for others it could be MDM, for yet more, it could be a hybrid approach including multiple systems.

 

The second question is a very tricky indeed since there are many threads it can take.  My favorite for this month relates to the degree to which the data model is optimized for purpose.  ERP data models are designed to meet the needs of (mostly) transaction oriented business tasks.  Business Intelligence data models, on the other hand, are optimized for data aggregation and dimensional analysis.  MDM, independent of any and all consuming applications (such as ERP and BI) has yet another data model.  This data model is, or should be, optimized for the purposes of ensuring semantic consistency of master data across any and all other systems in the business.  As such, MDM’s data model is a compromise more than anything, such that integration between numerous spoke systems and the MDM hub are timely and effectively operated.  So is there such a thing as the “right” or “only” data model for you?  Probably not.  Should MDM be a separate system from ERP, and any other business application or data warehouse?  Maybe. 

 

I wish I could attend the webinar but I cannot due to prior commitments.  It should be good to listen, learn, and then brainstorm the issues.  Maybe you can….and then comment here. 

Category: erp  master-data-lifecycle  mdm  siperian  

Tags: erp  hub-solutions  master-data-lifecycle  siperian  

Andrew White
Research VP
8 years at Gartner
22 years IT industry

Andrew White is a Distinguished Analyst and VP. His roles include Chief of Research and Content Lead for Data and Analytics. His main research focus is data and analytics strategy, platforms, and governance. Read Full Bio


Thoughts on More on ERP and MDM – apples and oranges?


  1. Dan Power says:

    Hi Andrew,

    Great post as usual! Could you please correct the web site in the first paragraph? The company name is Hub Solution Designs, and the web site is http://www.hubdesigns.com.

    Thanks and best regards — Dan Power

  2. Andrew White says:

    Oops – sorry about that – fixed!

  3. A. Osman says:

    There is a place for MDM, especially when companies have a number of key systems that are disparate, but shares common master data such as Customers and product items. In some cases, a Master item may have a complex process such as additional info gathering, approvals, routing etc.
    I would however not see the need for an MDM when the bulk of a companies transactions actually run on one ERP (the name of the ERP is not important.)

  4. Andrew White says:

    Thanks for the comment.

    I would agree that, if the bulk of a business is able to run its transactions in one instance of ERP, it might not need a separate effort called MDM. The implication being that if all the data was centralized then MDM (as a discipline) would be supported by the processes modeled and managed in ERP. However, ERP never lived up to its promise: so few firms say that the bulk of their business is supported by one ERP. Only smaller firms are able to do this; most mid-size firms (and larger) end up with multiple systems of records since ERP was never quite “good enough” in the business applications and business intelligence arenas.

    The other complication is what do we mean by ERP? For some users, ERP is an application stack sharing one common data model and process model (not many of these around), and for others ERP is a strategy that results in, “buy everything from one vendor” which is a very different thing.

  5. […] there is an article on ERP vs. MDM (ERP and MDM) that discusses the seperate data models (and requirements) […]

  6. Amit Jha says:

    Hi Andrew ,

    This one made an interesting reading for me ! I would like to add few things besides what has already been said . I guess driver for MDM is not just ensuring that data is shared across application . What one also need to think about if that data is complete and devoid of any data quality issues and not causing any business process disruption and poor decision making . While working with a large hi-tech manufacturing company in the US i encountered 10 instances of the supplier name being maintained in the same system and often with junk and illegal values bcoz of lack of processes, to name a few . This is where MDM can possibly help even in a single ERP scenario,to ensure that data policies and management processes are in place .

  7. Krithivasan says:

    Hi Andrew,

    MDM is still in evolving stages and companies now only start looking at MDM for managing their Master Data. But implementing MDM is not a one time process, as this acts as a hub for master data. Data continously flow IN & OUT of this hub. The Incoming data needs to be cleansed to ensure Data Quality. So MDM is a long journey – please visit the following blog & give your comments: http://www.infosysblogs.com/oracle/2009/03/master_data_management_a_lifel.html



Comments are closed

Comments or opinions expressed on this blog are those of the individual contributors only, and do not necessarily represent the views of Gartner, Inc. or its management. Readers may copy and redistribute blog postings on other blogs, or otherwise for private, non-commercial or journalistic purposes, with attribution to Gartner. This content may not be used for any other purposes in any other formats or media. The content on this blog is provided on an "as-is" basis. Gartner shall not be liable for any damages whatsoever arising out of the content or use of this blog.