deduplicating

[TheToeOfPooh]

There are mnay postings on the Web but none that I have managed to condense into a simple format.

We currently have a database of UK names and addresses which gets ammended and updated as required.

Is there a relatively simple method to quickly identify duplicates
e.g. Mr J Jones 12 High Street Derby DE21 1QP
is the same as
Mr John Jones 12 Hihg Street Derby DE21 1QP
Full FirstName but "typo" in Address.

My understanding especially when incorporating external data, is to run Filters\Sorts (using Excel?), before deciding what is "clean" as opposed to "dirty" data?

We will then be running this against Post Codes to determine matches
e.g. How many are in Post Code area GL20?

Any advice will be much appreciated as always.

Bernard

 

[mdmilner]

Try running the Find Duplicates wizard to build a query.

Mike

 

[Norie]

Mike

That was what I was going to suggest at first, and it may still be useful.

But it wouldn't pick up the type of example given in the OP.

I don't think there is a 'simple' way to do this sort of thing.

 

[Andrew Fergus]

I agree this isn't simple but given the UK postcode identifies the street, I would try doing a match with the postcode, street number, surname and the individual's initial and possibly the salutation. It may not be 100% accurate but may be close enough. It also depends on how the data has been stored.....

HTH, Andrew.

 

[TheToeOfPooh]

My thanks for these pointers.

A number of associated De Duplicating Queries got us down into the data as far as we need.

Cheers.

Bernard