Data Quality Fundamentals

Data quality is one of the most important — and most overlooked — aspects of working with data. Poor quality data leads to wrong decisions, wasted effort, and lost trust.

But what does "good quality" actually mean? Data quality is typically measured across six dimensions.

The Six Dimensions of Data Quality

1. Accuracy

Does the data correctly represent reality? An address is accurate if it matches the actual physical location. A sales figure is accurate if it matches the actual transaction amount.

2. Completeness

Does the data have all the values it should have? Missing postal codes, blank phone numbers, and empty required fields are completeness problems.

3. Consistency

Is the data formatted and represented the same way throughout? "Toronto," "toronto," and "TORONTO" are inconsistent representations of the same city. Inconsistency makes data hard to analyze.

4. Timeliness

Is the data current enough for its intended use? A customer address from three years ago may no longer be valid. Timeliness depends on how quickly the real world changes and how often the data is updated.

5. Validity

Does the data conform to defined rules and formats? A postal code field that contains "ABCDEF" is invalid. A date field that contains "February 30" is invalid. Validity checks catch data that doesn't follow expected patterns.

6. Uniqueness

Are there duplicate records? A customer database with the same person listed twice under slightly different names has a uniqueness problem. Duplicates inflate counts and cause confusion.

Measuring Data Quality

Data quality can be measured by profiling a dataset — examining it systematically to identify issues. Tools like spreadsheet formulas, SQL queries, or dedicated data quality software can help count missing values, find duplicates, and flag invalid entries.

Improving Data Quality

Improving data quality involves both fixing existing problems (data cleaning) and preventing new ones (better collection processes, validation rules, and governance). For a practical guide to fixing data problems, see the Data Cleaning 101 learning path.