Solve the puzzle 🧩 : Create a unified customer profile in Data Cloud
- Do you have customer information from different sources ( example marketing cloud, SAP, CRM ) , stored with different identifiers in the data cloud?
- Are you trying to figure out how to combine the information from different sources together without duplicates?
- Do you need to build a single profile for the customer with the aggregated information ?
Compilation doesn't have to be difficult. Data cloud's Identity Resolution can do all of the above and its way simpler than you think.
Prerequisites:-
Data models objects have been created.
Steps:-
1) Go to Identity Resolutions in data cloud
2) Click on new and select the correct data space and primary data model object. Click next, enter a ruleset name and make sure you enable the run jobs automatically option.
3) Create a matching rule example:-
Fuzzy names and Normalized email OR
Fuzzy name and Normalized phone OR
Fuzzy name and normalized Phone and Normalized email
4) The system will create unified identities and store Unified DMO's , which can be found under the Data Models tab.
At this stage each Data source has its own unique ID field which is linked to a single unified ID in the Data cloud. Example
SAP : Unified ID (1234) | Name ( ANANYA) | Unified ID ( 123ABCxyz)
Oracle : Unified ID ( ABCD) | Name ( ANANYA ) | Unified ID ( 123ABCxyz)
CRM : Unified ID ( abc12345 ) | Name ( ANANYA) | Unified ID ( 123ABCxyz)
Now next time when SAP is sending me information for Unified ID 1234 I can make sure it goes to the correct Unified Profile ID (123ABCxyz ) in the Data cloud.
Reconciliation Rules:-
If there is a conflict in a field wherein two systems are sending in different data for the same field example consent field then which systems data gets preference is defined using Reconciliation rules.
It gives you the option to select any of the following methods in case of conflict :
Last updated -Â The value from the most recently updated record must be selected.
Most frequent -Â The most frequently occurring value must be selected.
Source priority - This rule sorts your data lake objects in order of most to least preferred.Â