Digital Asset Management: Lessons from the Field #5: The Core Foundation - Taxonomy & Metadata
Now for the fun part – actually starting the implementation for your selected and contracted DAM platform (see previous Lesson from the Field #4). There are many steps involved in this process, with none likely to be more important than creating the taxonomy that will be leveraged to identify, tag, and locate specific assets and groups of assets after they are uploaded.
What exactly is a taxonomy? In its purest and original form, it is the branch of science concerned with classification, especially of organisms. For DAM, it is a way to organize and categorize data that will be ‘attached’ to assets, and always be available when they are viewed. The size and complexity of a taxonomy for any given organization’s assets varies considerably. Think of it like the building blocks of a house – its foundation, layout, its type, and level of design, all of its features, and every piece of detail that is important to its success, as a structure, and as a place to dwell for those who live in it.
The taxonomy’s core elements are known as metaproperties, and they can be of numerous types and formats, and some can be embedded in the asset digital file itself. They are made up of an unlimited number of options, or selections that can be made within a metaproperty, and those can also lead to sub-categories of metaproperties that are related to the higher-level metaproperty. Taxonomies can be as modest as a couple of dozen metaproperties, and that may work for many smaller organizations. For most organizations, the number of metaproperties will be considerably larger to encompass all desired attributes of their products, services, categories, divisions, brands, and the like (this author’s largest to date had over 125 metaproperties just for the DAM, other related systems encompassed many more).
A core philosophy we adhere to for almost any of our work is to ‘do the job right the first time.’ For creation of a taxonomy that will stand the test of time and serve as a great foundation for any organization, these are several important rules to follow and tips to keep in mind:
First, think about the various types of metaproperties that are necessary versus ‘nice to have.’ Be sure to include sufficient discussion of those metaproperties that are dependent on selections within other metaproperties.
Consider the specific format for each metaproperty, according to what is offered by the specific platform – single selections, multiple, short or long text entry, etc.
Invest as much time as is needed to ensure that all desired brands and products/services are included.
Know in advance that Asset Type and Asset Sub-Type will probably take longer to draft, vet, and approve than any other segment of metaproperties. These two metaproperties also tend to define what types of assets should go into your DAM, and what should not. (In other words, the DAM is not necessarily ‘THE single source of truth’ for all digital assets, but a primary source of truth for many types of assets, per the taxonomy.)
Involve the required stakeholders from all divisions/departments that should have a say in how the taxonomy is structured.
Consider asset naming standards may be beneficial for consistency, and how they will work in conjunction with metaproperties.
Consider what types of metadata can be driven by AI, and which metadata elements are embedded within the asset files themselves and are therefore automatically generated.
Review existing assets to be initially imported into the DAM to determine what can be gleaned from already existing asset best practices (this may also lead to an existing asset ‘clean-up’ effort prior to import).
Identify as many unique asset use cases as the organization requires, to ensure how the taxonomy must be structured/adjusted to accommodate them.
Vet the drafted taxonomy with a good cross-section of targeted users in the field, especially those who will have permission to upload and/or edit assets
Require formal approval from at least one representative from each key stakeholder group – this will limit questioning the initial taxonomy when launched (though it will continue to evolve).
Generally, no taxonomy is complete without a further discussion on related profiles of users, as not all of them will need to have access to all metaproperties. Work through this as part of the process.
Do not underestimate the value of thorough field testing once your initial taxonomy is complete. This will surface challenging areas that may need immediate tweaking, and this will save time and headache during full roll-out. Any test assets used can always be deleted if they are truly test assets. Some organizations also set up formal testing procedures. While these may not be necessary, you do want to test all options/selections within all metaproperties to be sure each and every one works as it should.
It is also highly recommended that, shortly after launch of your initial taxonomy, the organization should set up a governance committee or group that will regularly review the taxonomy, including statistics on how it is working, what is not being used as much as it should be, or correctly, and feedback from the field including suggested changes. Ideally, such a group should meet at least quarterly, although many organizations do not hold their first formal review for at least six months, depending on the volume of new assets being uploaded, number of key users, etc. It helps to also conduct regular and timely feedback sessions or surveys on how the taxonomy is working for all users. This can be part of a broader session or survey instrument gaining insights into actual use of the DAM as a whole.
#digitalassetmanagement #digitalasset #DAMstrategy #DAMbestpractice #DAMadvice #dam
#DAMvendorselection #damtaxonomy #dammetaproperties