Excluded Matches

Duplicate Searches can match edge cases like common placeholders or duplicates with unusual amounts of records that shouldn't be immediately merged. Reviewing these edge cases individually, right in the middle of a deduplication project, adds unnecessary manual work and slows down the process. Excluded Matches automatically keep these edge cases out of active Duplicate Searches without sending them to Manual Review, leaving only actionable duplicate sets to work with.

Excluded Matches are on by default for every object, with a limit of 20 records per match and poor or example data excluded. A fourth rule appears on HubSpot companies when a Salesforce sync is detected: it holds back merges that would break that sync.

Dedupely uses three types of Safety Settings to exclude matches:

• Records per match: Excludes any match that contains more than a specific number of records. (Default is 20 records)
• Poor or example data: excludes matches built on placeholder, example or low-quality data. On by default.
• Very short values: excludes matches made on very short values, such as a single digit or a single letter. Off by default — turn it on if short values are matching things they should not.

How to edit Safety Settings

1. Go to the "Safety Settings" menu.
2. Update the maximum records per match limit, or toggle "Poor or example data" and "Very short values" on or off.
3. Save changes. Updated safety settings apply immediately across all duplicate searches for that object.

Mentioned in this article: Duplicate Searches