Friday, 16 April 2010

What is Data Cleansing?

Data Cleansing is the process for ensuring that data sets are accurate and consistent. The term Data Cleansing covers a very broad range of data processing stages and involves various other databases and data sets to establish the accuracy of data.

One of the first processes that takes place in data cleansing is to check and update the addresses to ensure that they are Postcode Address File (PAF) compliant. Once we have updated all the addresses to make them as accurate as possible we now load the file into a our de-duplication system which finds any duplicated records and removes them into a separate file.

Now we have a unique file with an up-to-date and accurate set of addresses for each of our records we can begin to check the names against the addresses to ensure that the person still resides at the given address. If it is found that the person has moved addresses we will overwrite the old address with the new address and no change is made if the person is living at the address stated.

Now we have confirmed that the people in the dataset reside at the appropriate address we also need to ensure that they have not died recently. This is a very important stage of Data Cleansing as contacting the relatives of people who have just died is not only upsetting to the recipient but is also brand damaging. To check for deceased the file is checked against a national database registry of the deceased.

So to recap so far we have checked the addresses are correct, that the person in each record resides at the given address and ensured that they are still living.

So what's next, well that all depends on how you keep in touch with your clients. If you send out regular mailings then the above is adequate for your needs, but if you plan on contacting them by telephone we need to ensure that the telephone numbers are accurate. In order to add or update telephone numbers in the dataset we need to check them against a database that contains all of the phone numbers within the UK. The numbers are matched by name and address and in most cases only the surname is required.

I hope that gives you some insight into how data cleansing work, please keep checking back on this post as it will be updated regularly.

You can read more here: What is Data Cleansing


CNM Team

No comments:

Post a Comment