Published at: 2026-09-17

Detailed rule description


Overview

The matching behavior of duplicate check rules is affected by multiple underlying factors, including case sensitivity policies, special character filtering, ES tokenization match ratios, and null value handling. This document describes the specific behavior of each rule to help administrators understand why certain duplicate check results are produced and configure rules precisely for real-world business scenarios.

Basic duplicate check rules

Enterprise-type duplicate check rules

Production enterprise

  • Primary attribute: Case-insensitive
  • Plain text: Case-sensitive
  • Lookup Relationship: Case-insensitive

Sandbox enterprise

  • Primary attribute:
    • Exact match: Case-sensitive
    • Fuzzy match: Case-insensitive
  • Plain text:
    • Exact match: Case-sensitive
    • Fuzzy match: Case-insensitive
  • Lookup Relationship:
    • Exact match: Case-sensitive

Special character handling

  • Primary attribute duplicate check: The system filters out all special symbols, including: ~!@#$%^&*()+=|{}':;',\[]<>/?~!@#¥%……&*()——+|{}【】《》「」';:""'。,、?
  • Punctuation marks: Punctuation cannot serve as a differentiating condition in primary attribute duplicate checks.

Phone number handling rules

  • Parentheses in phone numbers are converted to spaces.
  • The system splits the number into multiple numeric segments using spaces as delimiters.
  • If any single numeric segment matches, the system identifies the record as a duplicate.
  • Solution: Use hyphens “-“ instead of parentheses.

Duplicate check configuration

Data scope settings

  • You can customize the data scope for duplicate checks. The default scope includes inactive data. You can configure the scope to exclude inactive data from duplicate checks.
  • Only existing data that meets the specified conditions participates in duplicate checks.
  • Data that does not meet the conditions is not identified as a duplicate even if it matches.

Trigger conditions

  • The system supports multiple duplicate check rules.
  • Duplicate checks are triggered only when created or modified fields match the predefined conditions.
  • Operations that do not meet the trigger conditions do not execute the duplicate check process.

Duplicate check blocking rules

  • When a duplicate check rule includes fuzzy match, the system does not forcibly block the operation.
  • You can decide whether to ignore the duplicate check result and continue saving.

Special duplicate check scenarios

Combined duplicate check

  • Conditions are fixed and cannot be customized.
  • You can view the field mapping relationships through the interface prompts.
  • A valid duplicate check rule is required for this feature to take effect.
  • Combined duplicate checks have independent rules and are not affected by regular duplicate check settings.
  • Conditions use exact match with an OR relationship. If any single condition matches, the system identifies the record as a duplicate.

Fuzzy duplicate check rules

Matching rules vary based on the keyword length:
  1. Fewer than 6 characters: 100% match based on ES tokenization.
  2. 6 to 9 characters: ES tokenization allows up to 25% of tokens to not match.
  3. More than 9 characters: 75% token match based on ES tokenization.

Null value handling

Three configuration options are available:
  1. Match null values: When a field in the new record is empty, the system identifies it as a duplicate of existing records that also have an empty value for that field.
  2. Do not match null values: When a field in the new record is empty, the system does not identify it as a duplicate of existing records that have an empty value for that field.
  3. Skip if empty: When a field in the new record is empty, that field does not participate in the duplicate check.

FAQ

Duplicate check rule activation time

  • Rules do not take effect immediately after configuration.
  • You must wait for the system to reindex data into ES.
  • Activation time depends on the data volume.
  • You cannot modify duplicate check rules while data is being reindexed.

Special scenario notes

  • Lead to Account conversion: Duplicate check validation does not apply. Configure this separately in Setup.
  • Creating master and detail objects simultaneously: Duplicate checks are not supported for Detail Objects.
  • Creating Account and Contact simultaneously: Duplicate check rules do not take effect.

Simple duplicate check tool usage

  • A single input box matches all duplicate check rules configured in Setup.
  • Pure numeric input matches only phone and telephone-type fields.
  • Non-numeric input matches text, URL, and email-type fields.

Related topics

Submit Feedback