The Dedupe Settings page is a central hub for managing how duplicates are detected and how records are merged. From here you can access configuration for Matching Models, Master Record Rules, and Field Merge Rules.

Accessing Dedupe Settings

Navigate to Dedupe > Settings from the left sidebar, or go directly to /dedupesettings in the URL bar.

Requirements:

  • An active DataGroomr license
  • Setup/Configuration permission

If you do not have the required permissions, an Access Denied message will be displayed. Contact your DataGroomr administrator to request access.

Settings Overview

The page presents three cards, each linking to a dedicated management area:

1. Matching

Set up rules or AI models to detect duplicate records.

Matching models are the core of DataGroomr's deduplication engine. They define which fields are compared and how similarity is evaluated to identify potential duplicates.


Two types of models are available:

  • Classic (Rule-Based) — Configure field-by-field matching rules with operators such as Exact, Fuzzy, Contains, and more. Ideal for straightforward matching scenarios.
  • AI / ML — Train a machine learning model using labeled pairs of records. The model learns patterns from your data and can detect duplicates that rule-based approaches may miss.

From this page you can:

  • Create new Classic or AI/ML models
  • Edit existing model configurations
  • Set a default model for each Salesforce object
  • Assign models to specific datasets
  • Clone models (same type or convert between Classic and AI/ML)
  • Copy models between Production and Sandbox environments
  • Delete models that are no longer needed

Models are grouped by Salesforce object (e.g., Leads, Contacts, Accounts) for easy navigation.

For detailed instructions, see: Manage Matching Models and Training Machine Learning Matching Model.

2. Master Record Rules

Control selection of the surviving master record.

When duplicate records are merged, the Master Record Rule determines which record is retained as the primary (surviving) record. This automates what would otherwise be a manual decision during every merge.


Examples of criteria you can define:

  • Keep the record with the most recent Last Modified Date
  • Keep the record with the most populated fields
  • Prefer records from a specific Record Type or owner

From this page you can:

  • Create new master record rules
  • Edit rule logic and conditions
  • Set a default master record rule
  • Assign rules to specific datasets
  • Clone or copy rules between environments
  • Delete rules

For detailed instructions, see: Merge Rules: Master Record Selection.

3. Field Merge Rules

Control how records are merged on the individual field level.

Field Merge Rules determine what happens to each field's value when two or more records are merged. They modify the data saved with the surviving master record.

Examples of field-level merge strategies:

  • Always keep the most recently updated value
  • Keep the longest or most complete value
  • Prefer values from a specific record or source
  • Concatenate values from multiple records
  • Distribute child records across surviving records

From this page you can:

  • Create new field merge rules
  • Edit field-level merge logic using the Rule Designer
  • Set a default field merge rule
  • Assign rules to specific datasets
  • Clone or copy rules between environments
  • Delete rules

For detailed instructions, see: Field Merge Rule, Rule Designer, and Merge Rules: Distribute Values Rule.

Common Actions Across All Settings

All three settings areas share a consistent set of features:

  • Search & Filter — Quickly find models or rules by name or description using the search bar in the toolbar.
  • Bulk Selection — Select multiple items using checkboxes to perform batch operations such as delete.
  • Set Default — Designate a model or rule as the default for its object type. The default is automatically applied to new datasets unless overridden.
  • Assign to Datasets — Associate a model or rule with one or more specific datasets.
  • Clone — Create a copy of an existing model or rule. For matching models, you can also convert between Classic and AI/ML types when cloning.
  • Copy Between Environments — Move models and rules between your Production and Sandbox Salesforce environments.

Related Articles