What is Data Cleansing

Definition

Data cleansing is the process of detecting and correcting errors and inconsistencies in data, including removing duplicates, fixing typos, standardising formats, and handling missing values, so the dataset becomes accurate and reliable for downstream use.

« Back to Glossary Index

Removes duplicate and conflicting records that would otherwise distort counts and metrics
Standardises inconsistent formatting such as dates, names, and addresses into one convention
Handles missing values deliberately rather than letting gaps silently skew analysis
Raises overall accuracy so reports and models work from a trustworthy base

Real World Example

A nonprofit cleanses its donor database by merging duplicate contacts, standardising address formats, and correcting misspelled names, which lifts mail-campaign deliverability and prevents double-counting donations.

FAQs

What problems does data cleansing fix?

Duplicates, inconsistent formats, typos, invalid values, and missing fields are the typical issues addressed during cleansing.

Is cleansing a one-time task?

No, because new errors arrive continuously, so cleansing is most effective as an ongoing, automated step in the pipeline.

How is cleansing different from validation?

Validation detects and flags bad data, while cleansing goes further by correcting, standardising, or removing it.