With ever shrinking budgets, many organizations are finding digital asset reuse a growing organization imperative. Effective use of metadata and taxonomies in the context of a DAM solution can facilitate great reuse patterns. Here's a start down that path.
Digital assets are anything that is non-textual: photos, videos, flash animations -- any kind of rich media. Wonderful as they are, they do present challenges for information managers.
One of the key problems with managing digital assets is that more often than not, the responsibility for creating assets is decentralized and siloed by channel or market. One group is working on email marketing, another on web commerce, and the eLearning team in Belgium has no visibility what the team in France is accomplishing. Further, there may be no common repository where assets are kept, meaning that content creators have to go running around chasing source files from colleagues and agencies.
To maximize impact of spend and gain efficiency in the asset creation process, digital assets need to be managed in a central hub -- a digital asset management system (DAM). However, it isn’t as simple as buying a new DAM solution, uploading all your assets and flipping the “on” switch -- there is much more to reuse than technology. And one of the key elements of any DAM project is a solid metadata and taxonomy framework.
[Editor's Note: For an introduction to DAM, see the article Digital Asset Management Defined.]
Describing Assets -- Good Metadata is Essential
Digital assets are more challenging than textual content in that they are completely dependent on their associated metadata for searchability in any system.
While search engines deal with unstructured documents by indexing their full text, we don’t have that luxury with digital assets. We can’t tell what a video is about without some sort of textual descriptor. Untagged assets are virtually invisible to search engines, making metadata and taxonomy invaluable allies in asset reuse.
DAM metadata is a tricky business; there are so many angles to consider:
- Different types of assets have different metadata requirements: Photos have orientation and resolution, whereas videos have running lengths and audio transcripts. Each type may also have associated standards.
- Assets can be decomposed into many levels: Each level may require its own metadata. A website might contain a flash animation, which in turn might contain a graph, which might contain an image of a product, each of which need to be tagged separately and differently.
- Rights are key: To reuse an asset, users need to be able to identify what license rights have been obtained, when they expire, what channels they apply to, what ways the asset can be used -- all of which can be hard to track and keep up to date.
- Everything is subjective: Images can be harder to tag consistently, given their inherent subjective nature and the ambiguity of language. Images can represent actual things or concepts, and both are highly dependent on interpretation. Does an image of a green and a red tomato represent the concept of variation? Or just the foundation of a good salsa? And what if I search for "tomato" and you search for "vegetables" -- would we both find this asset?
With all this, it can often feel as though creating DAM metadata is as much work as creating the asset itself. However, detailed metadata defines the ability to share and search assets and is worth the effort.
Leveraging Metadata & Taxonomy in Search
One of the most important aspects of developing metadata for digital assets is to determine what fields should use a controlled vocabulary -- a taxonomy. Taxonomy will not only help create consistent terminology, but it also gives you the ability to go beyond keyword search and surface assets using more sophisticated techniques.
Many DAM tools are limited to folder-based navigation, meaning that assets must be categorized in a single bucket, forcing users down a specific path to find them.
A Folder-based DAM Storage Approach. Image Courtesy of DeltaPhotography.com
Faceted search (or guided navigation) is a more effective way to search or browse for assets, as you can start or refine your search from any of the defined taxonomy branches. In this example from Getty Images – an online stock photography library -- I can search for a term, such as "diversity." I get a multitude of assets in my results, as well as options to refine my search, such as composition, subject, people, etc.
A Faceted Search Approach to DAM Storage.
Tagging components as well as finished assets with detailed metadata allows for an even finer level of access and reuse.
Metadata is also useful in search result pages to determine quickly whether an asset is appropriate for reuse. Presenting assets in a table format allows filtering/ordering on additional metadata fields. A quick hover preview can also give users key information about that asset, such as basic usage rights, availability of source files, language, etc.
When creating metadata for digital assets, you will likely need to expand your scope beyond the footprint of the DAM. In most cases, the creative process involves many systems, both upstream and downstream from the asset repository that involve the same metadata required in DAM:
- Creative planning
- Creative development
- Manufacturing (of end products)
When developing your DAM metadata, it is important to consider these connected systems. As the asset travels, its metadata should travel with it, which helps distribute the tagging process throughout the asset lifecycle.
Harmonizing metadata across tools does present a taxonomy management challenge. In complex environments, "taxonomy-as-a-service" is an excellent method for synchronizing vocabularies. In this model, the taxonomy sits in a central repository and can provide any required combination of vocabulary as an XML feed to consuming systems.
Standards are another key element of cross-system asset description and management. There are many types of standards available for digital assets, including:
Adobe's Extensible Metadata Platform (XMP) is becoming the predominant metadata standard for images, and being adopted by many vendors in the DAM space. The Metadata Working Group is a consortium of leading companies such as Canon and Sony, pursuing interoperability across applications and devices.
If this topic has piqued your interest, you can join me and two other speakers this Thursday for a free webinar. The event runs January 28th, at 2pm eastern time and is hosted by Earley & Associates. Registration is open now.
I’ll be speaking on DAM metadata and taxonomy basics, Henrik de Gyor will walk through a case study from K12 online education, and Diane Burley from Nstein will discuss best practices in reuse.