There are a number of different activities that fall under the category of data cleaning, all of which are designed to confirm, reformat, update and enhance business information in order to make it more usable.  Some of most common data cleaning processes are:

Address cleaning: this is the process of smoothing out the variations in the way people fill out forms and enter address data. By matching addresses to Postal Address Files, we can standardise formats of customer details, verify or determine gender, check validity of addresses, confirm phone numbers and flag customers that meet key criteria.  This service enables our clients to significantly reduce their postal mailing waste and staff time used to communicate with unengaged customers, as well as enable building robust datasets for analysis.

De-duping: data deduplication is designed to get rid of duplicate copies of repeated data points or entries. By comparing entries, we identify customers with the same name, birthday and address (for example) – it’s very likely that these two entries are actually the same person, so they can be consolidated.  This process, like address cleaning, is helpful for companies looking to minimise wasted resource in marketing communications, as well as improve data storage utilisation.

Matching: this is a process in which we create rules to identify groups of records that ‘match’ according to specific rules – like people who live in the same household, for example.  Most of the matching we do is around name and address data.  We have the ability to match anything within a dataset (DOB, Tax number, stores transacted at, phone numbers, email addresses, gender etc.).  Matches are useful when it comes to cross-selling, or consolidating mail outs (e.g.  only sending one letter to a household).