Fuzzy matching can be unreliable for very short names, where even a single character difference can prevent a match. The Short name matching adjustment option helps address this by allowing one character mismatch when names fall below a certain length, improving match accuracy in edge cases.
It is controlled by a concept called the fuzziness floor. The floor is calculated by dividing 1 by the configured fuzziness percentage. For example, if the fuzziness is set to 15%, the floor would be 6 (1 ÷ 0.15). Any name with a character count equal to or below this value will allow one mismatch during comparison.
For example, the names John and Jonn (4 characters) would now be considered a match with this setting enabled. Even though one letter is different, the name is short enough that at least one mismatch is allowed.
This adjustment applies only to Levenshtein-based matching algorithms and comes as an optional feature which can be enabled via a checkbox setting in the Rule Parameters section titled Short name matching adjustment.