Data Cleansing
Data Cleansing
The process of finding and fixing errors, duplicates, and inconsistent formatting to make data clean and usable.
In Simple Terms
Data Cleansing is the work of organizing data that's been entered inconsistently into a format that's easy to use. For example, imagine survey responses where some people wrote "USA" and others wrote "U.S.A." or "United States" — this process unifies them into one consistent format. It's also used to catch typos in street addresses or spot and merge duplicate entries for the same person.
Behind the Name
The name "Data Cleansing" pairs "Data" with "Cleansing" — the same word used for a facial cleanser that wipes away makeup and grime. The idea is that this process clears out the "dirt" that piles up in your data, just like a cleanser clears your skin. You'll also see it called "Data Cleaning" sometimes.
Take a Closer Look!
Data Cleansing is the work of finding errors, duplicates, and inconsistent formatting in the data stored in a database, then tidying it all up so it's ready to use.
If the data you've collected stays messy, you can't run accurate analysis or calculations on it — so this is a crucial prep step before putting data to work.
Broadly speaking, the work involves unifying inconsistent formatting and removing unnecessary information.
For example, it might mean standardizing whether phone numbers include hyphens, filling in blank fields, or merging duplicate entries for the same person into one.
This can be done by hand, but when there's a large volume of data, it's often processed automatically with dedicated software instead.
Doing this work makes the data more reliable and lets computers process the information correctly.
This careful groundwork is essential both when a company analyzes customer data to develop new products, and when feeding data into technologies like machine learning.