Automerge

The automerge functionality merges duplicate records without any approval or intervention from the data steward. Automatic processing of merge requests is critical when processing large volumes of customer data as automerge can expedite the resolution of duplicate records without manual review.

During automerge, the child entities of the duplicate records, such as contact points, relationships, classifications, and cross references, become the child entities of the master record. Note that groovy scripts on relationship objects don't run during merges. If they're critical, you can re-implement the scripts on the parent account or contact.

How Records are Selected for Automerge

Records are selected for automerge based on the following criteria:

• Score threshold: The score threshold is defined in the Match Configuration and determines if a record is included in a duplicate set.

• Automerge threshold: The automerge threshold is defined by the ZCH_AUTO_MERGE_THRESHOLD profile option and determines if the merge request for a duplicate set is processed automatically or if it must be reviewed manually.

Three possible outcomes for each record with regard to duplicate identification and merging are as follows:

• Low score below score threshold: The record isn't included in duplicate set and in the merge request for that duplicate set.

• Medium score above score threshold and below automerge threshold: The record is included in duplicate set but merge request for that duplicate set must be reviewed manually.

• High score above score threshold and above automerge threshold: The record is included in duplicate set and merge request is processed automatically.

The score for all the records in a duplicate set must be above the automerge threshold for automated processing. If one record in the duplicate set is below automerge threshold, and the other records are above the automerge threshold, the merge request must be reviewed manually.

Note: When you merge two or more records with exactly same children information under phone, email, or address the children information is merged and rolled up to the master record.

How You Configure Automerge

Enabling Automerge involves several implementation steps that must be completed by an implementor using the following tasks from the Customer Data Management offering in the Setup and Maintenance work area:

• Manage Customer Hub Profile Options: Use this task from the Customer Hub functional area to perform the following implementation steps:

  • Set Auto Merge Threshold profile option (ZCH_AUTO_MERGE_THRESHOLD) to the required value. This profile option specifies the threshold for auto merge. Merge requests with lower scores need data steward review. An exact match is 100.

  • Review the Record Size Limit of Duplicate Set (ZCH_DI_MERGEREQ_REC_SIZE). This profile option determines the maximum number of records in the duplicate set that can be merged automatically. By default, the maximum number is set to 10 records.

  • Set the Survivorship Enabled profile option (ZCH_ENABLE_SURVIVORSHIP) to Yes. This profile option enables the survivorship rules to select the master record and retain the attributes during a merge operation.

• Manage Survivorship Rules: Use this task from the Customer Hub functional area to create Set Master survivorship rules to choose the master record for merge requests created from the duplicate identification batch and set the rule to active.

  If there are no active Set Master rules or if the Set Master rules didn't trigger, the merge request must be reviewed manually, even if the ZCH_AUTO_MERGE_THRESHOLD profile option is set, the score for all records is above the threshold value, and the number of records is below the record size limit.

  Note: You can use the Set Attribute rules with Set Master rules to determine the Golden Master record. For automerge, Set Master rule is mandatory.

• Manage Enterprise Data Quality Matching Configurations: Use this task from the Data Quality Foundation functional area to perform the following implementation steps:

  • Create an active Match Configuration in Manage Enterprise Data Quality Matching Configurations task or use a predefined Match Configuration. Rebuild the keys if necessary.

  • Enable EDQ Real Time and Batch Basic Match Server in Manage Server Configurations task.

Run Automerge

This task involves the following two steps:

1. Create a duplicate identification batch and select Create Merge Request as the Automatic Processing Option.

2. Perform the task Run Request Dispatch Job to disposition the duplicate resolution sets.

The Dispatch Job processes any resolution request in Pending or Submitted status. You can run this job in two modes:

• On demand: Run Request Dispatch Job > Submit

• Per a specific schedule: Do the following steps to set up a recurring job:

  1. Click Advanced on the Run Request Dispatch Jobtask.

  2. Click Schedule tab and select the Using a Schedule radio button.

  3. Select the frequency you want and click Submit.

To see the list of dispatch jobs, and their statuses, navigate to Scheduled Processes under Tools.

Troubleshoot Automerge Issues

After you create your Duplicate Identification Batch, drill down into the completed batch to see the results. If duplicate sets have been found, and automerge is enabled, resolution requests are automatically submitted for merge.

If the resolution request wasn't submitted automatically, you can drill down to the duplicate set and compare the score for each record with the threshold in the ZCH_AUTO_MERGE_THRESHOLD profile option and the number of records with the limit in the ZCH_DI_MERGEREQ_REC_SIZE profile option. If all scores are above the threshold and the number of records is below the limit, verify that the following are true:

• Set Master rules are active and triggered to choose a master for the records in the duplicate set.

• ZCH_ENABLE_SURVIVORSHIP is set to yes.