1 Learn the concept
Data must be prepared before it can be analyzed. Raw data is usually messy: it has duplicates, missing values, inconsistent formats and errors. Preparing it means cleaning, combining and organizing it so that analysis is accurate. This lesson explains the main steps in data preparation and why they matter.
2 See it in action
- Collect: bring data together from different sources.
- Clean: remove duplicates, fix errors and handle missing values.
- Standardize: use consistent formats for dates, names and units.
- Document what you changed so others can trust the results.
Worked example
Cleaning a customer list
Problem: "Ottawa", "ottawa" and "OTTAWA" appear as three cities.
Fix: convert to one format.
Problem: the same customer appears twice with different emails.
Fix: match on name and phone, then merge.
Check: count the rows before and after, and note what changed so others can repeat it.
3 Study an example
Example scenario
A marketer combines sales data from two systems and finds the same customer listed under three spellings. After cleaning, her customer count drops and her averages become accurate.
4 Apply it and download
- Take a small messy data set.
- Clean it step by step.
- Document your changes with the data cleaning checklist.
Worksheet Data cleaning checklist and report templateOpen, print or save as PDF