Identifying Duplicates 

Use

New business partner records are added to the existing records and automatically matched against them. You examine the strategies underlying the matching process and check the matching result to identify duplicate records.

Procedure

Check the import and review the applied matching strategy

...

       1.      Switch back to the MDM Data Manager and choose View ® Refresh.

In the status line at the bottom of the window you see that there are now 21702 records. The import added seven new records to the 21695 contained.

Before you check the result of the matching, you review the predefined matching strategies and rules that were applied by the automatic workflow step.

       2.      In the header toolbar change from Record Mode to Matching Mode  and choose the Rules tab.

Matching rules determine which fields the matching is based on and how important equal (or token equal) values in certain fields are. For example a token equal full address of business partners results in a high score (50), an equal first name in a low score (15).

       3.      Choose the Strategies tab.

A strategy applies matching rules, summarizes their scores and defines the thresholds for a high or low matching level.

Check the matching result and identify duplicates

...

       1.      Choose the Workflows tab.

On the Workflows tab in Matching Mode you see all the open workflows in a matching step. In our case in the Tasks pane you see the Data Import workflow that was launched when the records were imported. The workflow matched the new records based on the MDM_ORGANIZATIONS strategy.

       2.      Select the workflow step.

Now you see only the matched records and their matching results in the Records pane (upper right).

       3.      Choose the Matches tab.

When you select a record in the Records pane, the potential duplicates, their matching score (by rule and the overall), and the matching level are displayed on the Matches tab.

A High matching level indicates that the record is very likely a duplicate, while a Low matching level indicates that the record might be referring to the same business partner. Records with matching level None have some overlapping values, but not enough overlapping values to be a potential match.

Based on this information and the record data, you decide if the records are identical.

Result

You identified the duplicates and are ready to merge the duplicates to one clean record.