Metadata is the descriptive layer behind a file: title, creator, date, keywords, rights, dimensions, and more. It is what a search engine, a DAM, or a person reads to understand what a file is without opening it. It is not the same as the file itself; it is the data about that data. Metadata comes in three types, descriptive, administrative, and structural, each serving a different job.
Why it matters
Metadata is the backbone of search and organization. Picture 500 photos in a tidy folder. In three years, will anyone remember the copyright holder, the usage rights, or the photographer, and find the one they need in seconds? With good metadata, any user can. Without it, the library is a guessing game and assets get recreated at real cost.
How it shows up in practice
When you take a photo on an iPhone, the file already carries embedded metadata: location, date, and time, plus facial grouping. A digital camera adds resolution and exposure. Inside a DAM, a marketing team layers on custom metadata that matters to them, the campaign, the approved channels, the expiration date, drawn from a controlled vocabulary so terms stay consistent. The Baltimore Ravens, for instance, tag game photos by week and play type so a designer can pull the right image during a live broadcast. Platforms like Adobe Bridge and Photo Mechanic let you inspect and edit the metadata embedded in a file directly.
Common mistakes
- Applying every field a standard offers, so most go unfilled and accuracy drops.
- Letting users free-type values instead of choosing from a controlled vocabulary.
- Tagging assets only inside the DAM so the metadata is lost on migration.
- Treating metadata as a one-time project rather than something that is audited and maintained.
The Stacks metadata guide walks through the types and best practices.