Master
Data Quality
The goal of master data management is to provide first-class master data for sustained cross-system data consistency. The master data quality function includes a whole array of features to ensure quality standards for master data in the following categories:
● Accuracy - Does my data adhere to defined formats and standards?
● Validity - Does my data contain incorrect information?
● Completeness - Does my data contain all the necessary information?
● Consistency - Does my data contain contradicting or duplicate information?
SAP NetWeaver MDM offers a number of functions for ensuring and improving data quality. Some of the key capabilities are described below:
Use validations and assignments to check the accuracy and consistency of your master data
With SAP NetWeaver MDM, you can use validations to help ensure that your centrally stored master data is accurate and can be used by the connected remote systems. You can define validation expressions to perform many kinds of specific quality tests. These expressions are Excel-like formulas that are evaluated by MDM and return a distinct value for each record. The expressions are used within MDM in validations, assignments and calculated fields.
The quality tests implement sophisticated business logic into your validation. Some examples of how such tests can be used:
● Make sure all required fields are filled
● Make sure non-lookup fields that do not have pick lists to enforce data integrity have a legal value (e.g. Price > 0)
● Compare the values of different fields to make sure the relative values are valid (e.g. Price > Cost)
You can perform the validations manually, automatically on data entry or automatically as a step in an MDM workflow.
More information:
Validating
RecordsMDM assignments are MDM expressions, similar to validations, except that:
Instead of returning a Boolean value indicating success or failure, they can return a data value of any type. Instead of displaying the expression result for each record in a column in the Records pane, they can assign it to the value of the specified field that can be edited by the user.
This is a useful feature for data cleansing after an import step or for central master data management of records in the MDM repository. With assignments, the fields values of selected master data records can be defined using expressions. This allows you to set default values for fields, for example.
More information:
Create more complete and meaningful master data using the MDM Enrichment Architecture
The MDM Enrichment Architecture is used for integrating third-party data enrichment providers with SAP NetWeaver MDM. A typical scenario for using external enrichment services is to complete and correct imported master data before matching and merging the duplicates.
The enrichment services typically perform operations that can be defined based on the context of a single record. Examples are address standardization, data normalization and lookups against a standard repository.
More information:
Create non-redundant and consistent master data using matching and merging capabilities
The matching and merging capabilities are used to detect and clean up duplicates and to manage the local object keys for cross-system communication. With this consolidated data, users can access the information they need to perform company-wide analyses and reporting.
The first step when matching records is to apply an optional normalization and transformation for each field of the record. The transformations basically replace information in a field, for example to expand abbreviations. In the next step, you apply matching rules and bundle these rules into matching strategies. According to a matching strategy, for example, MDM calculates scores for the probability that specific records are possible duplicates, as shown in the following figure.
A high matching score indicates that a record is likely to be a duplicate record.
You can then compare potential duplicates and merge any objects that are in fact duplicates into a new record. Merging typically involves interactive decision making.
When multiple records are merged, redundant records are deleted from the MDM repository, but the information about the remote keys of the corresponding records in the remote system is not deleted. This is done in MDM with the key mapping function. Key mapping maintains the relationship between the key of an object in the remote system (as shown in the following figure) and the corresponding master data object in MDM.
This allows you to harmonize records in connected remote systems by synchronizing local master data records with the consolidated MDM record.
More information: