The other day, I worked with a UX designer on a project to present assets to users. During our discussion, it became clear the designer lacked experience with Digital Asset Management (DAM). Consequently, I needed to provide a clear definition of metadata.

More importantly, I needed to explain arguably the two most important top-level classifications: Derived and Applied Metadata.

The difference isn’t always clear. Therefore, I wrote this short post to clear up the confusion. It should be especially helpful for those just starting their DAM journey.

What is Metadata?

Before diving into classifications, let’s define the core term. Simply put, metadata is data about data. It provides information about other data.

According to Merriam-Webster, metadata is “data that provides information about other data.” In the context of DAM, it is the crucial information that describes your digital assets, making them usable and findable.

Derived and Applied Metadata are the two core ways we generate this description. Each complements the other. Together, they make business-critical assets easy to find in your DAM platform. At the heart of every DAM system is the ability to make assets searchable.

Generally, the more business-relevant information you have, the better. Some metadata is systematically extracted (Derived), while other metadata is context-driven and added by users (Applied).

What is Derived Metadata?

Derived Metadata is descriptive information extracted directly from an asset. The system extracts this data as you import files into a DAM. As a result, it provides users with an immediately searchable baseline of asset properties.

This helps users find assets before an asset librarian enriches them with business context. Derived metadata comes from numerous places, including:

• File Information
• Embedded data (IPTC/XMP)
• AI (Facial Detection, OCR, Speech to Text)

1. File Information

File Information is the most basic data available for every file on a computer. This information relates directly to the file itself. It includes details like the name, size, type, creation date, and modification date. Additionally, format-specific attributes provide extra details, such as image dimensions, page counts, or color space.

2. Embedded Data

Embedded Metadata is information stored in a known location within the file, typically the header or footer. Each file format acts differently. Some store information in one location. Others duplicate data in multiple places to provide redundancy.

IPTC and XMP are the most common standards for embedded metadata. These standards facilitate metadata sharing. They allow descriptive data to travel with the asset wherever it goes.

Technically, someone applied this data at some point. However, because your organization did not apply it during the current workflow, we treat it as derived. For example, stock photography houses often include keywording, photographer names, and photoshoot details in their files. GPS data in smartphone photos is another common example.

3. Artificial Intelligence (AI)

AI is becoming increasingly accessible in our daily lives. Through AI, DAM platforms can derive detailed information about the assets they manage. They analyze files and tag them with improving accuracy.

For instance, AI toolsets can complete facial recognition or identify text in images using Optical Character Recognition (OCR). They can also transcribe speech in audio and video files without human interaction. These automated processes require training to improve accuracy. As AI improves, we get better-described assets without manual effort.

What is Applied Metadata?

Applied Metadata is information added to an asset to describe what it is and what it represents. This happens within the context of your organization.

Context is king. Because context is vital, you must apply searchable details to your assets. This allows your DAM to index them effectively. Consequently, users and systems can quickly identify the “right assets.” The ability to find and reuse content quickly is imperative to marketing organizations.

Applied Metadata is not specific to the technical file. Instead, it often describes the mood or tells the asset’s story regarding its creation. It allows for categorization and association with various parts of your business. This increases findability for users.

Examples of Applied Metadata include:

• Asset Type
• Usage and Rights information
• Product and Brand details
• Campaign codes

You can often source Applied Metadata from other enterprise systems. These include Product Information Management (PIM) and Title Management systems.

Conclusion

Many ways exist to categorize asset descriptions. Derived and Applied Metadata are two major classifications. They help us understand what we can automatically extract versus what the team must manually apply.

Ultimately, the more metadata you can automatically apply, the better. This reduces the requirements expected of users importing assets. It ensures they can focus on higher-value tasks rather than manual data entry.

If you and your team need to learn more about metadata or DAM, contact us, we’re here to help.