Database Hygiene and De-duping Service

Now Available from MDSS, Inc.

Can you tell these two respondents apart?

        Elizabeth Breitburg        Betsy Brightberg
        7813 Green Land            7813 Greenlane Place
        Cincy                              Cincinnati, OH  45237

In a side-by-side comparison, it's obvious these are duplicate records.  But, your computer can't tell the difference.  Look more closely, both the first and last names are different; there are slight differences in addresses; one entry is missing state and zip code info.  Even the cities are spelled differently.

Duplicate respondents and "dirty" records can create havoc with your qualitative database.  Bad addresses lead to returned mail.   Phoning the same respondent repeatedly leads to bad PR and reduced cooperation.  And database updates that get posted to only one record lead to recruiting nightmares.  The biggest danger is using a respondent who doesn't qualify for the study based on past participation.

Now MDSS has the answer – Database Hygiene and De-duping Service.  Our multi-stage process cleans up and de-dupes your entire qualitative database (Research Tracker 97, Research Tracker II, or Research Tracker Medical).  Processing can be scheduled to accommodate your downtime so your database is ready when you need it.  And best of all, it's MDSS – database experts and a name you can trust in marketing research.

Here's how it works:

Step 1:  Customer uploads database to MDSS secure FTP site
Step 2:  MDSS performs Database Hygiene and De-duping Services
Step 3:  Database is uploaded to FTP site; customer downloads and reinstalls database

Here's a complete rundown of the Database Hygiene and De-duping process

    1. Eliminate blank and "garbage" records in your respondent file

    2. Address hygiene:

      a. Standardize addresses using USPS protocol
      b. Add zip code and zip code + 4
      c. Correct misspelled street, city, and state names
      d. Identify addresses that don't match the USPS database of valid US addresses.
      e. Review and correct repairable addresses
      f. Report un-repairable addresses for future review by customer

    3. Identify duplicate sets:

      a. Identify exact matches based on last name, first name, and home phone number
      b. Identify suspected matches based on sophisticated logic including phonetic spelling of last name, first name variations, and gender matching.

    4. Intelligent data merge:

      a. Identify the "survivor record" in the duplicate set based on which record has the most recently entered or updated information
      b. Merge all respondent information into the survivor record
      c. Merge all participation information into the survivor record (plus eliminating any duplicate participation information)

    5. Delete "non-survivor" records:

      a. Delete non-survivor records for eligible respondents
      b. Mark as "alias" records for all "ineligible" respondents

    6. Generate reports of:

      a. Records identified as duplicates and deleted
      b. Records identified as duplicates and marked as aliases
      c. Records with un-repairable addresses

Want to learn more?

For more information, and our special introductory price offer, call or email

Kathy Pellman  kathy@mdssworld.com

Back to What's New with Research Tracker
Back to Data Conversion Services