Small Steps (Phases) ~ Best Practice for SAP MDM Implementation
After the Business case is prepared, this is the second most difficult part of defining MDM solution Implementation strategy. We have the responsibility of suggesting the best way to implement MDM solution in the organisation. I am not sure that how many organisations are ready for big bank approach in terms of master data management since it brings changes in the palpating of organisations most important asset i.e. master data.
Once the Business is able to visualize the benefits of implementing MDM solution, they ask that how long will it take to implement the solution. Definitely every organisation requires to get immediate benefits from the solution its implementing and do not want to wait for couple of years to see the benefits. However this might not be the best way to implement CMDM (central master data management) directly as well due to various reasons like
1. Organisation readiness
2. Data governance mechanism not in place
3. reaping MDM benefits without disturbing the current landscape
4. reaping MDM benefits without disturbing current data handling mechanism
etc
Thus the suggested approach of implementing the MDM solutions is to start with small steps (Phases). The organization will be able to get the benefits from the first step itself and then move to next step to achieve nirvana state of CMDM. They will be able to get the benefits of MDM in very short span of time say 3-6 months for one phase. The suggested approach is:
1. Consolidation
a. From source systems
b. From BI system
2. Harmonisation
3. CMDM (central master data management)
Consolidation could provide the immediate benefits for analytics requirements. This is very first step of implementing MDM. There can be two possible ways to consolidate master data from Source systems or consolidate master data from BI system (if it has all the required attributes) directly as per requirements.
Harmonisation is the next step and we will be able to achieve *single version of truth* of master data across landscape.
CMDM is the nirvana stage. The master data will me created, changed and managed centrally in MDM system and distributes to all the systems.
However, master data management doesn’t end on CMDM. MDM is not one time project but it is a program. However no more investment will be necessary with continous strong data goverance.
This example fite very well into this case. Suppose we have an illness, we go to doctor and he gives us the medicine. We dont take the full medicine but we take small doses over a period of time till the root cause of illness is solved. Its similar to implementing master data management to solve your master data problems.
Moreover, Since there may be many entities for which organisation require MDM solution like Customers, Vendors, Materials etc.. The suggested approach is again to take small step by taking 1 entity and then tackle other entities and also more diff systems in phases for MDM solution.
To conclude, the implementation steps depends on the organisation requirement. Thus it required to understand an organisations landscape to suggest the best possible method of implementing MDM Solution.
Company : SATYAM Computer Services Limited
Role: Projects Lead / Project Manager
Above shared views are my personal views and might not synchronize with my company views.
