Blog / Foundations, Metadata

Developing a Metadata Strategy

Peter Krogh
Thu Aug 06 2020

Last month we looked at the technology that underpins the creation and use of metadata. This month, we’ll examine how to make use of all that juicy metadata to add intelligence to your collection, make media discoverable and create workflows. Of course, the first item is to articulate the goals of the collection and the DAM project.

Your use of metadata should be a direct reflection of what’s important to you. The media in your collection, and the subject matter that interests you should drive the tags you make, your workflow, and the software and services you manage the collection with. This month, we’ll take a purpose-driven dive into the creation and use of metadata to serve your goals.

There are a nearly infinite number of ways to classify the subject matter and use of media objects. The tags you choose to invest in need to add value to your collection. These valuable tags sit at the intersection of your interests and the subject matter in your collection. Just because a tag could be applied to an image does not mean it’s helpful to you or others. It’s helpful to start by making a list of what’s important to you..

  • For individual image collections, it’s probably the people, places and events that are shown. If you have a particular interest like cars, dance or botany, then your tags should reflect this.
  • If you are managing a professional collection, then individual assignments, recurring subject matter and licensing information may be most important.
  • Institutional collections will have their own priorities, which will reflect the mission, structure and priorities of the organization, along with the intended uses of the collection.

Your tags should be ones that are relevant to the media library content, as well as to you and other stakeholders.

The need for metadata structure increases with volume

As collections grow from small to large, it becomes more important to build some intentional structure around your metadata. A handful of haphazardly-created keywords may be fine for a small collection, but will be insufficient as the collection grows.

Likewise, as the number of people using a collection grows, the need to create structured metadata grows as well. A single user collection may be quite usable with very unstructured metadata because the photographer is intimately familiar with the media that is included. When new users who are unfamiliar with the content of the collection are added, the same set of tags may be insufficient for effective use.

While it can be satisfying to create a well-tagged collection, most people will not want to drink the ocean in one gulp. It’s much better to approach the problem incrementally and add tags organically, according to the lifecycle of the image and the collection.

Gradually adding intelligence to the collection

Whenever you search for or make use of a media file in some way, you have an opportunity to add intelligence and value to the collection. The act of selecting images for use on a social media platform, for instance, carries some inherently valuable implications. First off, you have selected the image for use, so we can presume it’s good for communicating something. Additionally, by uploading to a social media platform, you convey some rights to the service, and it’s helpful to keep a record of that. And the usage is often helpful when looking for a file (e.g., “Can you please get me a copy of that photo we used for Instagram/website/brochure?”).

A place to store knowledge

The metadata attached to your media collection is not just helpful for finding the right photo or video. The media + metadata can become one of the primary ways to store information about what’s pictured, and not only about the picture itself. An enduring media collection can tell you a tremendous amount about the subjects depicted in the collection. And the very nature of a media archive makes it a highly suitable place to record and preserve historical information. In a lot of cases, there is no more suitable place to centralize the information about a family, company, region or field of study.

Of course, this is not new at all. The family photo album, for instance, has long been one of the richest documents of people, events, relationships, culture and history. Visual media can store and communicate a vast amount of information. You can learn a thousand subtle things from clothing, body language, architecture, physical proximity and more. When you add annotations to the visual information, it becomes even richer.

If keeping a record of history is an important part of your media goals, then it should inform your metadata strategy. You’ll want to think about the tagging process more expansively, and with this additional goal in mind.

  • Make tags for what you know about an image, especially if it’s not obvious from visible inspection. This could include the name of events, people and places.
  • Sometimes things that are entirely obvious to you are not obvious to others. The photographer may vividly remember the circumstances of a photo, which may be essential to understanding its meaning. I’m not suggesting that you need to dump everything you know about an image into the metadata. Instead, if preserving history is one of your goals, you should consider how structured tagging can help to attach essential information that is not universally known.
  • Get knowledgeable people to help with tags when appropriate. It’s likely that other people can help add to the knowledge base of people and events.
  • As we are on the verge of really useful computational tagging, it’s probably better to spend time adding background knowledge and context instead of adding tags for visually identifiable elements.

Take the opportunity to add knowledge to your media collection by having people tell you what they know about the subject of the images. The collection manager can ask questions and enter tags.

Is Mediagraph right for your organization?

Let’s find out together.

Book Your demo today