Welcome to Forums Sign in | Join | Help | Forums
in Search


Duplicate Criteria

Last post 08-19-2008 7:45 PM by Juliette Sodergren. 1 replies.
Page 1 of 1 (2 items)
Sort Posts: Previous Next
  • 07-31-2008 6:42 AM

    Duplicate Criteria

    Hello,

    I am currently in the process of running a duplicate report on RE but wanted to find out what criteria everyone uses to ensure you are picking up every duplicate on the database.

    If you have the time please can you tell me the criteria you use for both Individual and Organisational searches.

    At the moment I am searching on:
    Ind - Surname=10       Org -   Org Name= 40
    First Name= 3                       Post Code= 8
    Post Code= 8

    Many Thanks

    Ryan

  • 08-19-2008 7:45 PM In reply to

    Re: Duplicate Criteria

    Hi there

    Firstly, in UK you have a longer postcode, we only have 4 digits in Australia. That being said, I would reduce your PostCode to about 3. It's a bit rule of thumb really but if you get overly specific, you're more likely to miss out on matches.

    We have in the Business Rules Section of RE:

    Individual:

    Firstname 1 (to allow for where we only have the first initial)

    Surname 7

    Postcode 3

    This is so that on data entry, we try and capture the duplicates before they are entered. The user may get in the search results a few irrelevant matches but it's better than creating duplicates.

    Organisation

    More tricky as we often have one organisation record for each branch of a bank, so the address is important for us:

    Org Name: 12

    Address: 20

    Suburb: 8

    Postcode: 3

    When querying, we ask the data entry team to only enter:

    Individual

    Surname: First few characters

    Firstname: First Initial

    Postcode: First 3 characters

    This will bring up more potential matches than if you enter the whole surname or postcode.

    De-Duplication

    Since the above steps aren't fool proof, and the RE matching is based on matching from left to right, we export our data and match it using a 3rd party tool. This picks up more than just matching the first n number of characters in each field. We use a product from DataTools called TWINS, although I'm sure there'll be others on the market in the UK. This will pick up tighter & looser matches on the data, as well as you being able to specify items such as Phones, emails, date of birth in the matching. 

    Kind regards

    Juliette
    Cancer Council NSW, Australia

     

     

    Filed under: ,
Page 1 of 1 (2 items)