Estimatin

[Former Member]

Hi MDM Experts

I would like to know how MDM objects are estimated,

Suppose I have the following objects to be build...

1. Moduling Repository

2. Set up Validations

3. Set up Multi level Workflow approval process

4. Build Searches

5. Setup for MDIS

6. Setup for MDSS

What is the estimation process generally practiced.

Regards

[Former Member · Accepted solution]

Hi Vickey,

Based on my Experience and knowledge i would suggest the following estimations:

1. Moduling Repository - Depends on your understanding of the business ,if you have the Functional design ready which talks about the fields and their types as present in the ERP system then designing within MDM is not a great task( max 1 day with all properties filled and for a complex repository),

but if you haveto model the data in MDM yourself then thiscould take couple of days to months

2. Set up Validations - Creating a validation in MDM is a couple of minutes job but you need to build the logic which could take time depending on your talent and logic.Simple and few validations can be build in few hours.

3. Set up Multi level Workflow approval process - This is a complex task and can take any time from few weeks to more than that.

4. Build Searches - In few hours

5. Setup for MDIS - Building the map will take time depending on your configuration and number of fields mapped else setting up just the Mdis config is a few minutes job.

6. Setup for MDSS same goes for the syndication.

Hope It Helped

Thanks & Regards

Simona Pinto

[Former Member]

HI Vickey!

Following are the factors Influencing Timelines of MDM Implementation,one can go to granular level and estimate time requirements of each of the individual processes.

DATA MODELING

SAP provides standard repository & maps for Vendors, Customers, Materials, BP and Article. These repositories contain standard data elements. Z elements/tables defined by the client have to be added to the repository.

Identifying the complete list of Z fields/tables to be included in MDM and modeling the elements in the repository can be a time-consuming & complicated activity. Further, any modification to the repository will lead to modification of standard import maps, standard syndication maps and the standard XI configuration being used in the scenario.

COUNT OF MASTER RECORDS

The count of master records has an effect on cut-over data import time and consolidation time. If u2018consolidated master datau2019 forms a part of any project deliverable, then the number of master records will have a significant impact on the project timelines.

COUNT OF SOURCE SYSTEMS

Extracting data from source systems requires the following deliverables:

u2022Creating/Using an extraction program in source system.

u2022Configuration in XI.

u2022Import Maps and Ports in MDM.

COUNT OF DESTINATION ITEMS

Pushing data to destination system requires following deliverables:

u2022Syndication Maps and Ports in MDM

u2022Configuration in XI

SAP provides standard XI configuration and standard syndication maps for SAP R/3. For non-SAP destination systems, some or all of these deliverables might have to be created and configured.

Does data enrichment scenario exist?

Some scenarios might demand data enrichment using 3rd party application. Identify those scenarios and determine whether direct Import of data is possible. If not, custom Java development or configuring of u2018MDM Data Enrichment Adapteru2019 might be required.

Does extension scenario exist?

Extension Scenarios are common in u2018MDM for Materialsu2019 implementation. Such requirement demands modification to u2018Qualified Tablesu2019 using MDM Java or ABAP APIu2019s.

WORKFLOW CREATION TOOL

Workflows in MDM can be created using either of the below mentioned technologies:

u2022MDM WORKFLOWS: MDM Workflows reduce the development time; have limited functionality and cannot be used in complicated scenarios.

u2022GUIDED PROCEDURES: Guided Procedures enable creation of workflows that can call processes/objects from multiple systems. Custom functionality can be achieved by calling Java code from Guided Procedures.

u2022BUSINESS WORKFLOWS: Business Workflow is a proven technology for implementing workflows in SAP. Exposing and calling these workflows from MDM demands integrating Business Workflow with UWL. Development of these workflows has to be done in SAP system.

MASTER DATA Vs TRANSACTIONAL DATA

In some scenarios, the client might find it convenient to maintain transactional data in MDM repository. However, though such a requirement can be mapped, may lead to major performance bottleneck due to the constant import/modification & syndication required to keep the transactional data updated.

FINALIZING REPOSITORY STRUCTURE

Ensure that the repository structure is finalized at the end of prototyping phase. The finalized data model should be well documented and Signed-Off by the client. Any change to the structure there after will result in linear increase in the implementation time of the project. This is because; any modification to the structure might result in modification to all or some of following components:

u2022Repository structure

u2022All Import maps created on the relevant table/field

u2022All syndication maps created on the relevant table/field

u2022All Validations/Assignments created on the relevant field

The above also applies to change in display fields. Hence, the display fields for a table should also be finalized in the prototyping phase.

Acceptance of the New System by End-Users

Data modeling in MDM has a major impact on how users create new masters records in MDM. At times, the repository may have been designed to optimum use of Qualified tables and other similar feature-rich tables in MDM. However, if the End-User finds it difficult to create data as per the repository design, the whole purpose of implementing MDM might be defeated.

SIZING / BANDWIDTH

Server Sizing is of utmost importance in MDM implementation project. MDM server loads a sizable amount of data in-cache to enable various search functionalities and to minimize the probability of disk accesses. This demands availability of enough memory (RAM). Also, when importing data into repository through MDM Import server, several resource intensive steps like parsing, transformation, database write etc. are performed. Efficient & speedy import demands proper server sizing. Further, ensure that the recommended bandwidth is available.