Deduplication (Dedupe) Service

Research has shown that 70% of consumers state that getting duplicate mail annoys them intensely. Unlike many other offerings, WinPure will detect and flag duplicates in your data at three different levels of granularity, using advanced dedupe methods and algorithms.

At the lowest level, it will detect and flag duplicated individuals, at the second level it will flag family matches where it finds two or more records for the same family group (at the same address).

Finally, at level three, it will flag households, where individuals, not necessarily with the same family name, share an address.

		Significantly lowers the likelihood of the same message being sent multiple times to an individual
		When used in conjunction with your existing customer database, it can prevent recruitment offers being made to existing customers

How WinPure will dedupe your mailing lists:

The results data includes extra fields which is an Individual ID, a Family ID and a Household ID. The key to understand where potential duplicates lie is in these fields. Where the numbers are the same, this indicates a potential duplicate at that level; where they are different, this indicates that these records are different, or there are other factors in the file which mean it cannot be flagged as a duplicate (see Case IV for an example of this).


CASE I- Individuals are matched, Family and Household are matched.

Where forename/surname and address correspond, then a match will be made at all levels

CASE II- Individuals are matched, Family and Household are matched.

Where forename/initial and address correspond, then a match will be made

CASE III- Individuals are not matched, Family and Household are matched.

Gender is derived from the title, if a gender cannot be derived because the title field is null, the gender will be flagged as unisex.

Where forename exists in both records and the initials match but the forename does not this will not be matched at Individual level but will be matched at Family and Household.

CASE IV- Individuals are not matched, Family and Household are matched.

Where there is more than one possible match against two or more forenames then none of the individual matches will be made. Family and Household will be matched as normal

CASE V - Individuals are matched, Family and Household are matched.

Where the two forenames are the same a match will be made, i.e.

CASE VI- Individuals are matched, Family and Household are matched.

Where the title is unisex and forename/initials correspond, a match will be made, i.e.

A match will be made in this case provided there are not any other influencing records, such as a Miss Sandra Smith at the same address. If this were the case then the results would be like Case IV above.

CASE VII- Individuals are not matched, Family and Household are matched.

Where there is a unisex title and a possible match from each gender on initials no Individual match will be made, Family and Household will be matched.

CASE VIII- Individual - Match, Family - Partial Match, Household - Match.

Where there is a match on surname then a family match will be made. In addition, in this case we have found, another non-family individual at the address and so a household match will be made.