About matching

AD
DM

Your Network orgClosed A version number applied to Network features to ensure that they are compatible for importing configurations to target environments. Feature versions increment each time a major enhancement of the feature is released. is provided to you with Veeva-owned records pre-loaded. The Network match process allows you to load any other data into Network and match or align those records to what already exists in Network. Data sets vary greatly and records that represent the same entity rarely contain identical components.

You can tailor match rules to allow other data sets to match and then merge into existing Network entities, thereby creating one copy of each unique entity instead of numerous, disparate variations.

How does matching work?

Network offers numerous built in comparison strategies or algorithms that provide the end user with great power and flexibility. These algorithms are leveraged in the match rules to compare data and determine if they match strongly, weakly or not at all.

Each of these algorithms perform a comparison of two data points or strings but each uses a different algorithm to do that comparison.

These algorithms include but aren’t limited to the following:

Algorithm Explanation
Equal Compares two strings to confirm if they are exactly identical
nGram Breaks the contents of a string (like corporate_name__v) into parts of length “n” and then does its comparison.
JaroWinkler Determines the similarity between two strings using the statistically proven JaroWinkler comparison method
Beider Morse
Double Metaphone
NYSIIS		 Checks for phonetic equality of two strings.

Types of Network matching

Network offers two distinct match methods that are built into the application:

  • Ad hoc match - Use to perform once-off matches on a single file.

    Matches found via ad hoc match do not impact your Network org, they provide match information from Network for the incoming file, but never merge to the data in your Network instance. Ad hoc match provides guidance on whether each record strongly or weakly matched records in the Network org, or indicates if a match could not be found.

  • Regular match process - A key element of the process used to load any data set into Network.

    Match rules are set up as part of the source subscription and dictate how each incoming record should be matched and merged into existing Network records.

    Match rules are also leveraged when adding individual records, validating if the new record matches an existing record or not before simply adding it to your instance.