Detecting Duplicates

You can set the system to insert a record into the Common Customer Directory, either through a Batch or Lookup match, and if Duplicate Detection is enabled, the system will not insert duplicate records. The duplicate detection parameters are set from the Insert – Duplicate Detection of the Batch or Lookup Workflow.

NOTE: Depending on the duplicate detection settings, Duplicate Detection affects the performance of your system.

The minimum requirements for a matching record to be considered a Duplicate are:

The Confidence Code is at least 7. The value can only be 7 to 10.
The Country Codes match.
The States match (Match Grade of A).
The Company Name Match Grade is either A or B.
If the values of any selected Client Key match, the record is considered a duplicate (D&B records do not have Client Keys).
The levels of Sensitivity are Low, Medium, High, and Very High. The following table displays typical Match Grade values required in each category to reach the definition for Duplicate. Street Number, Street Name and P.O. Box Number are included as part of Address. State is always compared.

Field	Very High	High	Medium	Low
Company Name	100+	93+	90+	50+
Street Number (Address)	100+	90+	90+	50+
Street Name (Address)	100+	97+	90+	50+
City	100+	90+	50+	Ignored
State/Province	90-100	90-100	90-100	90-100
PO Box Number (Address)	90+	90+	50+	50+
Phone	100+	90+	70+	50+
Zip/Postal	Exact	Prefix compare	Prefix compare	Ignored

Both the Confidence Code and Sensitivity Settings must pass for a record to be considered a Duplicate of one already in the local database. If a field is empty or contains invalid data, that does not rule out the record being a duplicate.
An Exact comparison compares non-normalized strings of characters, with leading and trailing spaces removed and multiple spaces collapsed. A Prefix comparison is similar to an exact comparison, and one string may be shorter than the other, but no less than half the length.

Loading

Detecting Duplicates