A guide to the lifeblood of DAM:

Key concepts and best practices for using metadata in digital asset management systems.
By John Horodyski. Sponsored by Widen Enterprises and DigitalAssetManagement.com.

Scan for PDF

Copyright 2011 Widen Enterprises, Inc.

A guide to the lifeblood of DAM: Key concepts and best practices for using metadata in digital asset management systems.

Do you know what digital assets you have and how to identify, organize, and describe them? This should not be rushed, as this is critical to the impact your DAM systems use will have on your overall efficiency and, ultimately, your bottom line. Getting this wrong could damn not only your DAM, but your broader workflows and processes as well. Metadata is an asset unto itselfand an important one, at that. It provides the structure and information needed to make your assets more accessible and, therefore, more valuable. In other words: it makes them smart assets. Simply digitizing video and audio files only scratches the surface of their value as digital assets. Their full potential is realized by their use and the relevance of the associated metadata. After all, how much value does an asset have if you cant find it? This white paper will show you the essential building blocks and best practices of metadata for your digital asset management system.

What is metadata and what does it mean to DAM?

What is metadata? Metadata is, simply put, data about data. It refers to the descriptive elements that define and describe an asset. The National Information Standards Organization breaks metadata down into three main categories:

Descriptive metadata describes a resource for purposes such as discovery and identification (i.e., information you would use in a search). It can include elements such as title, abstract, author, and keywords. Structural metadata indicates how compound objects are put together, for example, how pages are ordered to form chapters (e.g., file format, file dimension, file length, etc.) Administrative metadata provides information that helps manage an asset, such as when and how it was created, file format and who can access it. There are several subsets of administrative data. Two that are sometimes listed as separate metadata types are rights management metadata (which deals with intellectual proper ty rights) and preservation metadata (which contains information needed to archive and preser ve a resource).

Here are some other key concepts to understand, especially if youre starting your metadata analysis: Taxonomy: The science of naming and organizing things into groups or classes that share similar characteristics. It can also refer to any scheme for such an organization of informationin the case of DAM, for the purpose of classifying and identifying digital assets. Taxonomy through metadata - The categories, sub-categories and terms that make up a taxonomy often manifest themselves as metadata. Metadata therefore enables more precise search results and personalization. Controlled vocabulary: Controlled vocabularies contain preferred and variant terms with defined relationshipshierarchical and/or associative. Examples of controlled vocabularies include glossaries, specialized dictionaries, standard terminology lists, synonym rings, reference data, authority files, domain-specific taxonomies, thesauri and ontologies. Thesaurus: A tool that controls synonyms and identifies the relationships among terms. It usually has a preferred term and can be hierarchical but doesnt have to be. For example, dog, pooch, puppy, mutt and dog is the preferred term.

Copyright 2011 Widen Enterprises, Inc. All rights reserved.

A guide to the lifeblood of DAM: Key concepts and best practices for using metadata in digital asset management systems.

Authority files: Typically used for lists of people, organizations etc. e.g. list of public companies, industry segments, geographic locations. This could be a taxonomy.

Building a metadata strategy: key issues

Now that the foundation has been set with definitions and key concepts, you can get to work on building an effective metadata strategy. The three key questions you need to answer are:

1. What problems do you need to solve? 2. Who is going to use the metadata, and for what? 3. What kinds of metadata are important for those purposes?
It is important to consider how much metadata you need. Metadata is expensive; it takes valuable time to create the structure and ensure that it serves your needs. If it does not, then time and money are wasted not finding assets due to inadequate metadata. Building, testing, inputting and maintaining metadata and taxonomies come with costs. Implementing metadata may require UI changes and/or back-end system changes. Every metadata field costs money and time to implement and adjust to. You need to make your model extensible and avoid the common mistake of buying tools first, then figuring out the metadata strategy later. Ensure that you account for business goals and how metadata should contribute to reaching those goals. To help get that going, there are some critical components of a metadata strategy that need consideration: Building the right team: Name a team of DAM stakeholders to take the lead in identifying goals and designing a metadata strategy to meet those goals. Naming your requirements: Before getting deeply involved with any vendors, you should be able to ar ticulate and enumerate (both to the vendor and your own organization) those things you absolutely need a DAM system to do for your organization. Making the business case: Identify all costs, benefits and risks of creating and maintaining rich metadata. When making ROI calculations, you should account for the resources required to add, maintain, test, and update metadata and taxonomies. Metadata specifications: These are always subject to change, but you should have some sense of what your metadata model will look like, including any controlled vocabularies and keywords. Ongoing workflow: Where will metadata come from? Know who will be responsible for maintaining and adding metadata, along with what processes theyll be following. Q/A & Testing: Have a method of measuring the effectiveness of your metadata model and protocols. Detailed metrics go a long way when it comes time to evaluate and make improvements. There is a considerable effort behind this, but careful observation of these components will help you start your work and move you in the right direction.

Copyright 2011 Widen Enterprises, Inc. All rights reserved.

A guide to the lifeblood of DAM: Key concepts and best practices for using metadata in digital asset management systems.

How Does it All Get There?

One cannot exaggerate the importance of understanding that most of the benefits of DAM software cant be realized without good metadata. You need to sell the vision of what the company will gain by having good metadata in your DAM system. Implementing metadata may require UI changes and/or back-end system changes. Metadata powers efficiency in DAM which is what allows administrators control and end-users the ability to find what is needed on a moments notice. Furthermore, every metadata field costs money and time to implement and adjust to. There is no benefit unless the tagged content cuts costs or improves revenues; you need to demonstrate bottom-line and top-line benefitsalthough bottom-line ones are easiest to prove early on. It is difficult to analyze how much operations cost today and how much would be saved. Therefore, focus on the productivity gains. There are some key metadata fields that you should focus on:

Basic metadata Creator Creation Date Owner Publication Date Title

Retrieval Creator Title Description Subject Publisher

Rights Embargo Date Expiration Date Location Restrictions Usage Restrictions Pricing

Knowledge maintenance People Places Organizations Financial metadata Harvesting in-line markup

Consistency is important when applying metadata. Consider the following tags:

President Barack Obama Barack Obama President Obama Obama

Each tag could point to a different topic. Yet, fundamentally, its the same principal element of the subject of President Barack Obama that is relevant. Having a principal DAM administrator and/or metadata specialist on your team will be highly valuable. In fact, depending on the size of the organization, there may well be multiple administrators in various locations responsible for tagging and asset ingestion (i.e. insertion into the DAM library). If this is the case, it is even more important to ensure metadata consistency. Last, there is metadata in headers, file systems, naming conventions and query logs that could be extracted automatically. While automatic classification tools exist and produce results that are more consistent than humangenerated ones, humans are more accurate and better at recognizing nuance. Semi-automated or hybrid approaches are often the best way to go, generally with human involvement for distributed manual review and correction.

Copyright 2011 Widen Enterprises, Inc. All rights reserved.

A guide to the lifeblood of DAM: Key concepts and best practices for using metadata in digital asset management systems.

What is your metadata model?

Time and time again, people feel the need, and rightly so, to describe their assets in multiple ways (i.e., from the perspectives of multiple users). More often than not, these exercises can lead to well over 75 metadata fields for describing assets. Sometimes, this number can rise north of 100information overkill in all but the rarest cases. What you are looking for here is a manageable set of fields with which you are able to discern the most critical characteristics (administrative, descriptive and technical) of your assets. There is no magic number of metadata fields, but you might want to shoot for a sweet sixteen: the sixteen descriptors that you need to identify, organize, and describe each of your assets. Ultimately, this will be the data your users search against.

What is your taxonomy?

Once you have identified your assets and have a manageable metadata model, its time to consider how this will be organized in the DAM system, from on the back end and front end. End users generally search for assets by a variety of means: Faceted classification systems - searching for assets based upon more than one value or dimension (e.g. Shop by Materialgold, silver, diamond, etc.) Well-defined folder browsing A structured vocabulary from the corporate system feeding the DAM assisting in search & retrieval. Think of your users and how theyll want to navigate your asset library and search for files. While there might not be a simple one-size-fits-all solution, any good DAM software should be configurable enough to meet your needs.

What are the industry standards and which are right for me?
Standards should be reviewed during your strategy development. Standards are created by industry members to meet the specific needs of that industry. It is wise to use an industry standard if you can find one that applies and extend it as needed. You should pick standards that are extensible so that you can add your own namespace (or other accepted extension). Sometimes content owners require vendors to offer some level of collaboration to enable automated content interchange and interoperability between software tools. It is important to remember that standards are valuable for efficient, precise, federated search and retrieval across repositories, as well as automating workflows, distribution, and integration with other business systems. Indeed, standards adoption results in huge cost savings due to the efficiencies created. Examples of metadata standards to consider are Dublin Core, PRISM, (PRISM DIM2), METS, ONIX, XMP MARC, , IPTC Headers, GILS, SCORM, IMS and JDF which one(s) you use should depend on your businessobjectives.

Copyright 2011 Widen Enterprises, Inc. All rights reserved.

A guide to the lifeblood of DAM: Key concepts and best practices for using metadata in digital asset management systems.

Benefits of metadata
Some people waste more than 40 percent of their time searching for existing assets and recreating them when they arent found. This lost productivity, and redundancy can get very expensive. The key to avoiding these unnecessary costs is good metadata to aid and assist in search and retrieval. Other benefits of metadata include:

Higher ROI based on increased sales through improved product find-ability, par tner cross-sells and up-to-the minute updates to adver tising Cost-cutting through resulting from fewer customer calls (due to substantially improved website self-ser vice) and more efficient CSR responses Improved regulatory compliance (i.e. avoidance of penalties for breaches or regulations) Reduction in redundancies in work and data storage. More effective rights enforcement resulting in less loss of revenue due to piracy

Metadata & taxonomy governance

The best way to plan for future change is to apply an effective layer of metadata governance for your DAM system. There is more to maintaining the metadata than just maintaining the taxonomy and metadata specifications. Vocabularies must change over time to stay relevant. This goes for new terminology being added to assets as well as synonyms and/or slang terms. The DAM softwares user interface (UI) might need a refresh or redux according to user needs and demands. One great way to make sure the system and metadata schemes are meeting the needs of their users is to offer them a standard change request form and a procedure for accepting or denying the change request. And any updates should take place at published times, only, to provide lots of notice and not affect the users in the DAM. Measuring metadata and taxonomy quality While often neglected after implementation, metadata demands ongoing monitoring and management. Dont forget Q/A and testing as they are critical to your success. Testing should begin very early in the process. In fact, start testing as soon as the first assets are loaded into DAM library. Its important to note that metadata & taxonomy development is an iterative process and you will need to solicit ongoing feedback from your users. Use both qualitative and quantitative measures, and remain flexible throughout. You should be gauging things such as: consistency appropriateness of tags time to complete tasks reaction to search results usefulness of training materials user satisfaction

Copyright 2011 Widen Enterprises, Inc. All rights reserved.

A guide to the lifeblood of DAM: Key concepts and best practices for using metadata in digital asset management systems.

Best practices
Using metadata in a DAM system takes work, but once you get going, it will be your greatest asset. Some best practices to adopt include:

Star t with a few metadata fields that are relevant to all assets and gradually move on to groups of less universally applicable fields (those that are specific to cer tain file formats, products, divisions of your organization, and so on). Avoid overloading your users with metadata fields. Have a subject matter expert analyze your content to inform decisions regarding categories and tags. Have a midpoint check-in with stakeholders to ensure youre on the right track and build ongoing consensus (e.g. every three months). Be prepared to adjust metadata and taxonomies as your business needs evolve.

Practical metadata rules

Here are some practical metadata rules to follow: Develop an incremental, extensible process that identifies and empowers users, and engages stakeholders with feedback loops, user testing and evaluations. Do a quick implementation that provides measurable results as quickly as possible and record them. Repurpose assets as often as possible. Accept that it wont be perfect; all metadata schemes can be improved. Just ensure that what you have in place meets your needs, and make adjustments when thats not the case. Implement good governance policies.

Content is no longer king. The user is.

If you have great content and no one can find it, the value of the content is diminished. You need to understand how your users and customers want to interact with assets before designing your metadata schemes. If you carry those user needs through to the back-end data structure, youll empower users with the categories and content attributes they need to filter and find what they want. Metadata shouldnt be rushed. Take the time to leverage best practices like usability testing to determine needs and validate your metadata and taxonomies. Remember that metadata is a snapshot in timekeep it up to date and let it evolve. Keep your eyes on your companys goals, as the best metadata design is the one that increases the revenue of the company by harnessing the power of your data about your data.

Copyright 2011 Widen Enterprises, Inc. All rights reserved.

A guide to the lifeblood of DAM: Key concepts and best practices for using metadata in digital asset management systems.

About the Author

John Horodyski is Principal, DAM Education (http://www.dameducation.com), a DAM consulting agency focusing on DAM education & training. John is also the Manager, Digital Programming, Product Development at the CBC (Canadian Broadcasting Corporation). John also serves as Director of Marketing & Business Development for Wrinkled Pants, an educational software studio focused on the development of education and literacy based apps for the iPad. John teaches a graduate course at San Jos State University, School of Library & Information Science in Digital Asset Management. John spent many years at Electronic Arts where he managed their global digital asset management system as well as being a producer within the EA Sports and online divisions. He has published professional articles and presented at numerous conferences on digital media, metadata in video games and taxonomy design and continues to offer DAM training & consulting. John holds a Masters Archival Studies and Masters Library and Information Science from the University of British Columbia and is the Managing Editor to the Journal of Digital Asset Management.

Copyright 2011 Widen Enterprises, Inc. All rights reserved.