We spend lot of time in data cleaning. But there is no simple definition of exactly what clean data means. Therefore, I created a simple 10 item checklist. If all 10 items are correct, data is clean. Simple and easy to use.
10 rules for Clean data
Remember, here we are talking about input data – not output (reports).
- Each column must have a heading
- No blank headings
- No duplicate headings
- No formulas in headings
- No merged cells
- Each column must have ONE meaning
- Each column must have ONE type of data
- No subtotals or grand totals
- Formatting should not be used instead of data
- Data must grow vertically, not horizontally
Convert clean data to a Table
If you are using Excel, as soon as data is cleaned, convert it into a Table. Why? Here are the benefits of using Excel Tables – 13 articles.