Match Transform Multiple Fields

[former_member783646]

Hi Experts,

While comparing Multiples fields in Address Data, I'm getting an incorrect results. Address data is now working as [ Country OR Street OR Postal code OR City ] and therefore showing 100% always, but it should be like [ Country AND Street AND Postal code AND City ] and the expectation is that to show exact percentages

First it is comparing Country if it matches showing 100% even though street and postal code and city is different.

Please find the attachments of setting which I used and please share the feedback what all the changes need to be done.

[virginia_hagen]

With the input that you are showing, I'm not sure that we can suggest how to fix it as you are doing something non standard and I'd have to set up a job to see what is returned.

Best practice is to take the standardized data from an address cleaning process and use that as input into match.

Have you run the address through an Address Cleansing process? Global Address Cleanse, URAC, or DQMmicroservices?

This will standardize your data prior to using it as input into match. There are output fields that can be returned from the upstream transform that gives you the discrete field so that you don't have to do the cross field matching and it will (if its able) correct the data so that the spelling of the street names and such are consistent.

The other recommendation that I have for you is to use the Match Wizard.

This typically sets up the transform to give you the best results and then only minor changes are required.

The other helpful starting point for matching is the blueprints that are available. This give you example data that can be used with the job so that you can more easily see how a job might be designed for each business case.

Knowledge Base Article ##1263925 - Are Data Quality blueprints available? - Data Services will lead you to the wiki page where they are located. Direct link to the wiki is here: https://wiki.scn.sap.com/wiki/x/XCW7Gg

Thanks,

Virginia