In my first post, I suggested a number of questions that I would address here. The first was "What is a taxonomy?". Here, we'll talk about why taxonomy projects need to be ongoing and not something that is done once and then left alone.
One challenge to success in taxonomy and metadata standards is getting people in the organization to understand that a taxonomy is a living breathing entity. Products change, markets change, customer needs, competition, technologies, solutions, approaches, and so on, are all evolving. We use new terms all the time. Old terms (remember "information superhighway"?) are continually going out of style or simply no longer applicable or descriptive enough.
When we work with organizations to develop taxonomies, after we go through the process, they suddenly realize that getting the organization to use the result of the effort is ongoing in and of itself. So we need to develop a change management, update, and roll out process. We need to think about ongoing governance and how managing the taxonomy fits in with other enterprise standards and change processes.
Asking when the taxonomy will be complete is akin to asking when the organization will be done with sales. Or marketing. Or product development. Or manufacturing. Nothing stays static in business. Why would the taxonomy be static?
The question becomes how changes will be incorporated and how often the taxonomy needs to be refreshed.
Changes can come from a number of places. The following are typical change triggers:
- Development of new offerings,
- Expansion to new markets,
- Introduction of new types of content,
- Identification of new terms/concepts needed for tagging,
- Countries splitting or merging,
- Changes in organizational structure,
- Changes in standard taxonomies (e.g. NAICS),
- Proliferation of leaf level terms requiring new grouping categories
- Identification of new frequent terms in search logs,
- Identification of new useful access points or aspects to be used for navigation, personalization, or customization,
- Integration of new consuming systems.
Those are the triggers. What are the sources for new terms?
- Monitoring standard taxonomies
- Search-log and click-trail analysis
- Tagging needs
- User research and usability studies
- Consuming system requirements
Suggestions and requests can be submitted by:
- e-mail sent to a distribution list set up for this purpose
- Opening a request in a request management tool
- Filling in a suggestion/feedback Web form.
One of the most interesting areas of research that we are exploring is the integration of social tagging (or folksonomies) with structured taxonomies. Basically, this is a way of harvesting terms so that they can be automatically nominated as taxonomy candidate terms and then reviewed by a taxonomist.
Bottom line is that maintaining the taxonomy is as important as deriving and validating the taxonomy and needs to be performed by someone who understands taxonomy and the implications of changes to the structures used to classify and organize content.
