Data Deduplication

Identify and Eliminate Duplicates Fast

On average, a database contains 8-10% duplicate records. These duplicates result in waste and inefficiencies and cloud your ability to get a single, accurate view of the customer.

Learn More

What We Do

Melissa is the most powerful and accurate matching and deduping solution on the market to combat the problem of duplicate records.

What sets it apart from the rest is its intelligent parsing capability to understand and parse the various components of domestic and international addresses. By combining deep domain knowledge of international address formats and advanced fuzzy matching techniques, MatchUp gives you the ability to identify and merge/purge even the most difficult-to-spot duplicate records.

Eliminate clutter and duplicates that prevent a clear view of your customers
Increase the accuracy of your database saving you time and money
Reduce postage and mailing costs by eliminating duplicates using advanced matching technology

How MatchUp Works

MatchUp employs a matchcode to determine if two records should be considered duplicates. MatchUp uses a predefined matchcode, or one that you have created using the Matchcode Editor.

The following matchcode components (data types) are available for use in identifying duplicates:

Prefix	Street Pre-Directional	ZIP+4®
First Name	Street Name	Postal Code
Middle Name	Street Suffix	Country
Last Name	Street Post-Directional	Phone/Fax
Suffix	PO Box™	Email Address
Gender	Street Secondary	Credit Card Number
First/Nickname	Address	Date
Middle/Nickname	City	Numeric
Department/Title	State/Province	Proximity
Company	ZIP9	General ID
Company Acronym	ZIP5
Street Number

Fuzzy Matching

MatchUp combines Melissa’s deep domain knowledge of contact data with over 20 fuzzy matching algorithms to match similar records and quickly dedupe your database.

MatchUp employs the following fuzzy matching algorithms to identify “non-exact matching” duplicate records:

Phonetex	Consonants Only	Smith-Waterman-Gotoh
Soundex	Alphas Only	Jaccard Similarity Coefficient
Containment	Numerics Only	Overlap Coefficient
Frequency	MD Keyboard	Longest Common Substring
Fast Near	Jaro	Double MetaPhone
Accurate Near	Jaro-Winkler	Needleman-Wunch
Frequency Near	n-Gram	Vowels Only
UTF-8 Near	Dice’s Coefficient

deterministic-probabilistic-matching-levenshtein-algorithm

Global Merge / Purge & Deduping

The World Edition of MatchUp supports 12 countries, including Canada, Germany, U.K., and Australia. MatchUp’s advanced deduping can see through diacritic equivalents to Latin characters and interpret keywords that are the same but spelled differently (i.e. Germany and DEU).

Unique Matching Scenarios

MatchUp has some unique attributes which can be employed to help identify duplicates in some interesting ways.

1. Survivorship for Golden Record Creation

Matchup can select the best elements from multiple records to survive consolidation, ideal for the creation of golden records for a single customer view. Available in Microsoft SQL Server Integration Services (SSIS) and Pentaho PDI.

2. Proximity Matching

MatchUp’s patented distance algorithm uses latitude-longitude coordinates and proximity thresholds to identify duplicate records that are geographically close together. For instance, using location attributes, MatchUp can detect matching records at different addresses (for example, a company with two different entrances) but within a specified distance to each other.

3. Householding

MatchUp can identify and consolidate records that are members of the same household to better understand customer relationships, lifecycle, and needs. You can also use MatchUp to bring together multiple business accounts into “corporate families” to build insight and better evaluate the total sales relationship. Householding can also be used to eliminate unnecessary multiple mailings to the same household to cut down on wasted print, production, and postage costs.

data-deduplication-unique-matching-scenarios-1

Three Ways to Dedupe Your Data

MatchUp offers three methods of operation (or ways to match records):

1. Read / Write Deduping

Compares records in one or more databases at once. Each unique group will have one record that receives an “output” status; the other matching records receive a “duplicate” status. Ideal for matching entire databases at one time.

2. Incremental Deduping

Enables real-time matching by comparing each record as it comes in (like from a web form or call center) against the existing master database. If the incoming record is not a duplicate, it can be added.

3. Hybrid Deduping

Provides a combination of the first two methods with the flexibility to customize the process to match an incoming record against a small cluster of potential matches. With hybrid deduping you can store the match keys in a proprietary manner. Ideal for real-time data entry or batch processing of entire lists.