Empty value rules define how blank properties affect a duplicate match.
This matters because real datasets often have missing emails, phone numbers, domains, addresses, or CRM fields.
Why blanks are tricky
| First name | Last name | Company domain |
|---|---|---|
| Ana | Lopez | example.com |
| Ana | Lopez |
Should these rows match? The answer depends on how much evidence the other properties provide.
Datablist empty value rules
Datablist exposes two rules for Smart, Metaphone, and Distance comparisons:
- Value required is the default. The property does not match when either record is empty, including when both are empty.
- Match on empty value scores two empty values at 100% and one empty value at 75%. Records with every compared property empty never form a group.
The score for the property contributes to the overall similarity score across the selected properties.
⚠️ Blank fields can create false matches
Keep Value required for strong identifiers such as email, domain, LinkedIn URL, CRM ID, or SKU.
Use Match on empty value only when other properties provide enough evidence. For example, two matching emails may remain useful candidates when a secondary Last Name value is missing.
Use this setting with multi-column deduplication, smart matching, or distance matching.
See the exact scoring example in the Duplicate Finder reference, or follow the complete deduplication workflow.