Most information professionals would agree that the actual building of a thesaurus (sometimes known as a taxonomy in DAM circles) is pretty straight forward: choose and follow one of several standards (ANSI/NISO, ISO or BS) and ensure your work reflects the given content. From relating concepts to building out hierarchical structures, the steps are well defined. What significantly adds to the complexity is working with the various, often disparate, constituents to come up with a usable product. That’s why taxonomy design should be user-centric. Users know what their information needs, goals, and preferences are. It is up to us, the information professionals, to suss out what those are and design to them. Unless we are the designer and the sole user of the taxonomy, such as Carl Linnaeus, seeking input from users and involving them at every point in the design process is key to building a usable taxonomy that will grow and change to accommodate users’ needs.
BUSINESSES NEED TAXONOMY
A taxonomy, specifically an enterprise taxonomy, offers users a consistent, company-wide, language to describe, tag, and index content. In other words, it helps users organize and find material important to their work.
(the French military leader)
(the dessert) , CC BY-SA 3.0
With a taxonomy, a user’s search results are more relevant to their need because most of the included subject terms represent simple or unitary concepts which aid in distinguishing groups of content from one another. This is one of the central advantages of a taxonomy: that it minimizes the level of effort and time to find the material users need.
USERS DON’T NEED TO BE MYTHICAL CREATURES
Consider the taxonomy’s users, use, situations, business context and interaction with the system (Saracevic as quoted by Shiri, 2012). This can be done in a variety of ways, some of which will be more appropriate to your situation than others. Here are some of the strategies I have employed successfully:
- Assemble a taxonomy development team that includes subject matter experts (SMEs), selected users from across the company and various content managers.
- Observe search behavior during UI testing, card sorting exercises, and DAM System user testing. Pay particular attention to your users’ search formulation tactics, “the process of designing or redesigning the search formulation and the ways in which information request and query elements are analyzed” (Shiri 2014, p. 66).
- Gather and analyze metrics from across your company. These can include asset usage statistics, successful/unsuccessful searches, and check in/out records from the corporate library.
- Conduct a company-wide content audit. During one-on-one interviews, carefully document each type of content, who creates it, the context in which it is created and used, and its associated workflow.
- Pay close attention to how staff organize their emails and digital files.
Be sure to hold educational seminars to define taxonomy and how it will help your constituents. These sessions are great opportunities to energize users and perhaps encourage a few to get involved in the process. Holding these presentations will also encourage stakeholders to buy-in to the idea behind a taxonomy and take ownership of it. Remember, you will need upper-management to support the initiative because you will ask them to give you time with their subordinates.
When you have a draft version of a branch of the taxonomy, have at least one SME review it. I enjoyed using Google Sheets to share each branch because it allowed the users and I to exchange comments about specific terms. Some commercially available taxonomy tools will save you time by allowing reviewers to add comments and candidate terms directly to the taxonomy, saving you from outputting your work-in-progress each time there is a significant change.
The last stage of your taxonomy design should include validation exercises. This is were staff from across your user base will test the taxonomy. I asked users to perform some specific concept searches (relating to something they were working on) and a few random searches they came up with on the spot. The exercises were followed by questions to gauge how commensurate the taxonomy was to the concepts they searched for.
Keep staff regularly informed of what you are doing, how the project is progressing, and when they can expect the project to be finished. This may seem like obvious advice. After all, staff are busy (why else would they be employed?) but the time and effort it takes to produce and present these reports may dissuade you from heeding it. Just remember, unless you keep users focused on the endgame, they will turn their attention to something that is more engaging.
Promote and advocate the benefits of the taxonomy each and every chance you get. At meetings, when there was a bit of time to spare, I demoed the taxonomy, showing what it looked like, how it would be used, and how it would improve their workflow.
Above all, know when to wrap up a taxonomy project because, if you are like me, it is easy to keep going. There are always ways to make it better and it is such fascinating work. I am a very curious person and designing a taxonomy requires that I learn about a subject and how users interact with content: two of my favorite things about taxonomy design.
What user-centric-design strategies have you employed to build a taxonomy? How well did they work for you?
Ali Shiri (2014–04–03). Powering Search: The Role of Thesauri in New Information Environments (Kindle Locations 1270–1271). Information Today, Inc.. Kindle Edition.