I’ve set up my Data Streams and mapped them to the Individual DMO in Salesforce Data Cloud. I created an Identity Resolution rule set using "Exact Email" and "Fuzzy Name," but after running the rules, I still see multiple Unified Profiles for what should be the same person. Is there a delay in the processing, or am I missing a critical step in the "Match Rules" configuration?
3 answers
Identity Resolution isn't instantaneous; it runs on a schedule or can be triggered manually. However, the most likely culprit is your "Match Rules" hierarchy. If you have "Exact Email" but the emails have different casing (e.g., [email protected] vs [email protected]), they won't match unless you've normalized them during ingestion. I recommend using a "Calculated Insight" or a Data Transformation to lowercase all emails before they hit the DMO. Also, check your "Reconciliation Rules"—if you have multiple sources, the system needs to know which source "wins" when data conflicts occur.
Have you checked the "Consolidation Rate" in the Identity Resolution summary tab? If the rate is 1:1, it means no records were merged. Could it be that your "Party ID" or "Source System ID" are being treated as the primary key incorrectly, preventing the fuzzy logic from even looking at the name fields?
Always verify your "Anonymous vs Known" settings. If one record is flagged as anonymous and your rules only target known individuals, they will never merge into a single golden record.
Spot on, Joseph. The 'is_anonymous' flag is a frequent trap. If you're ingesting web tracking data, you have to be very intentional about when you upgrade that profile to "Known."
Richard, the consolidation rate was indeed 1:1. It turns out I had mapped my Source ID to the Individual ID field, which acted as a hard unique constraint. I followed Margaret’s advice on normalization too—cleaning the whitespace and casing in the Data Stream mapping fixed the "Exact Email" failures. Now my consolidation rate is closer to 1.4, which looks much more accurate for our database!