Unify customer fields for data unification
In this step of the unification process, choose and exclude attributes to merge within your unified profile entity. For example, if three entities had email data, you may want to keep all three separate email fields or merge them into a single email field for the unified profile. Some attributes are automatically combined by the system. You can create stable and unique customer IDs and group related profiles into a cluster.
Review and update the customer fields
Review the list of fields that will be unified under the Customer fields tab of the table. Make any changes if applicable.
Optionally, generate the customer ID configuration.
Optionally, group profiles into households or clusters.
Edit a merged field
Select a merged field and choose Edit. The Combine fields pane displays.
Specify how to combine or merge the fields from one of three options:
Importance: Identifies the winner value based on importance rank specified for the participating fields. It's the default merge option. Select Move up/down to set the importance ranking.
Customer Insights uses the first non-null value. For example, given entities A, B, and C ranked in that order, if A.Name and B.Name are null, then the value from C.Name is used.
Most recent: Identifies the winner value based on the most recency. Requires a date or a numeric field for every participating entity in the merge fields scope to define the recency.
Least recent: Identifies the winner value based on the least recency. Requires a date or a numeric field for every participating entity in the merge fields scope to define the recency.
You can add more fields to participate in the merge process.
You can rename the merged field.
Select Done to apply your changes.
Change the display name of merged or separate fields. You can't change the name of the output entity.
Select the field and choose Rename.
Enter the new display name.
Separate merged fields
To separate merged fields, find the attribute in the table. Separated fields show as individual data points on the unified customer profile.
Select the merged field and choose Separate fields.
Confirm the separation.
Exclude a merged or separate field from the unified customer profile. If the field is used in other processes, for example in a segment, remove it from these processes before excluding it from the customer profile.
Select a field and choose Exclude.
Confirm the exclusion.
To see the list of all excluded fields, select Excluded fields. If necessary, you can readd the excluded field.
Change the order of fields
Some entities contain more details than others. If an entity includes the latest data about a field, you can prioritize it over other entities when merging values.
Select the field.
Choose Move up/down to set the order or drag and drop them in the desired position.
Combine fields manually
Combine separated fields to create a merged attribute.
Select Combine > Fields. The Combine fields pane displays.
Specify the merge winner policy in the Combine fields by dropdown.
Select Add field to combine more fields.
Provide a Name and an Output field name.
Select Done to apply the changes.
Combine a group of fields
Treat a group of fields as a single unit. For example, if our records contain the fields Address1, Address2, City, State, and Zip, we don't want to merge in a different record’s Address2, thinking it would make our data more complete.
Select Combine > Group of fields.
Specify the merge winner policy in the Rank groups by dropdown.
Select Add and choose if you want to add more fields or groups to the fields.
Provide a Name and an Output name for every combined field.
Provide a Name for the group of fields.
Select Done to apply the changes.
Configure customer ID generation
Define how to generate customer ID values, the unique customer profile identifiers. The unify fields step in the data unification process generates the unique customer profile identifier. The identifier is the CustomerId in the Customer entity that results from the data unification process.
The CustomerId is based on a hash of the first value of the non-null winner primary keys. These keys come from the entities used in data unification and are influenced by the match order. So the generated customer ID can change when a primary key value changes in the primary entity of the match order. The primary key value might not always represent the same customer.
Configuring a stable customer ID enables you to avoid that behavior.
Select the Keys tab.
Hover on the CustomerId row and select Configure.
Select up to five fields that will comprise a unique customer ID and are more stable. Records that don’t match your configuration use a system-configured ID instead.
Group profiles into households or clusters
You can define rules to group related profiles into a cluster. There are currently two types of clusters available – household and custom clusters. The system automatically chooses a household with predefined rules if the Customer entity contains the semantic fields Person.LastName and Location.Address. You can also create a cluster with your own rules and conditions, similar to match rules.
Select Advanced > Create cluster.
Choose between a Household or a Custom cluster. If the semantic fields Person.LastName and Location.Address exist in the Customer entity, household is automatically selected.
Provide a name for the cluster and select Done.
Select the Clusters tab to find the cluster you created.
Specify the rules and conditions to define your cluster.
Select Done. The cluster is created when the unification process is complete. The cluster identifiers are added as new fields to the Customer entity.
Submit and view feedback for