insights by infactory in-factory GmbH Use Tolerant Match when identifying duplicate customer data records (CRM)

Well-maintained address lists and customer master data are the basis for successful marketing and sales campaigns. In addition, an error-free address database is crucial for meaningful analyses and statistics and fundamental for targeted campaign management. One of our clients needed to improve the quality of customer data in order to better implement targeted marketing campaigns. That is why it is important to run a duplicate check regularly and to clean up addresses.

The tool Tolerant Match from Tolerant Software was designed for this purpose and can rate the degree of similarity of data records by means of a highly configurable score.

Identification types

.

Static identification

.
When customer data is loaded, it is checked against the inventory to see if the customers can be identified.

Dynamic identification

.
Within the current data delivery, checks are made for duplicates or similar records.

Identification via Tolerant Match

.
Tolerant Match is a GUI tool from Tolerant Software that can connect to databases and help identify record duplicates or similar records. The cleaned data can be written back to the database.

Tolerant Match enables fuzzy identification criteria. This means that it provides solutions to enable identification even if there are spelling problems or incomplete information in the identification data provided.

Tolerant Match finds address data and customers even when there is only fragmentary information. The special technology in Tolerant Match also allows different options to be combined for searching.

A contact can be identified using different combinations of fields. In the example, this is attempted using FIRSTNAME, LASTNAME and address.

Creating rules for reference data

.
The rule set for tolerant matching is managed in custom profiles that are put down in XML files. The rules are fairly complex and probably not very volatile. They are usually easiest to customize via the GUI and store in an XML profile.

Due to the complexity of the XML profiles, it is recommended to manage them in a version control tool, so that any changes can be easily tracked.
An example of a Tolerant Match profile:
The degree of agreement between two records is indicated by a score, the calculation of which can be influenced by the rules:

Tolerant Match Score

Tolerant Match parses the data for matches (here a sample log):

Tolerant Match Log

In the example, the condition for a match is that the score exceeds a given minimum value: If the score for the match reaches the specified value, like here (in the example in column 2, the score is 95)…

Tolerant Match Example

…the contact is identified and can be written back into the database with the found ID:

Tolerant Match Identification

Author: Mehdi Koupaei

ANY QUESTIONS? WE HAVE THE ANSWERS!

Write us. We are looking forward to your message!

MAIL TO