You've probably heard the phrase "data about data." That's metadata — information that describes a dataset rather than being the dataset itself.
Think of metadata as the label on a filing cabinet drawer. The drawer contains the actual files (the data). The label tells you what's inside, when it was last updated, and who's responsible for it.
What Metadata Tells You
Good metadata answers questions like:
- What does this dataset contain?
- Who created it, and when?
- How was the data collected?
- What do the column names mean?
- What are the units of measurement?
- How often is it updated?
- What licence applies to it?
- What geographic area does it cover?
A CSV file of municipal park locations is the data. The metadata might say: "Created by the City of Ottawa Parks Department. Updated annually. Coordinate system: WGS84. Licence: Open Government Licence – Canada." Without that metadata, you'd have a list of numbers with no idea what they mean.
Types of Metadata
Descriptive metadata helps people find and understand a dataset — title, description, keywords, creator, date.
Structural metadata describes how the data is organized — column names, data types, relationships between tables.
Administrative metadata covers management information — who owns it, access rights, retention schedule, licence.
Why Metadata Matters
Without metadata, data is hard to find, hard to understand, and easy to misuse. A dataset with no description might be used incorrectly — for example, treating a sample as if it were a complete population, or using outdated data as if it were current.
Good metadata makes data discoverable (you can find it), understandable (you know what it means), and trustworthy (you know where it came from).
For a deeper dive, see the full Metadata 101 learning path.
Learn what "good data" really means in Data Quality Fundamentals.