Taxonomies: Connecting Users to Content

by:   |  Posted on

Taxonomies may be thought of as hierarchies of categories to group and organize information to be found when browsing, or as a structured set of terms used to tag content so that it can be retrieved efficiently and accurately. Sometimes the same taxonomy may serve both purposes, and sometimes two different taxonomies are used, one for each purpose, for the same content or site.

Taxonomies are not new, in fact  there has been a lot written about them, including an informative series of six articles here in Boxes and Arrows by Grace Lau in 2015. An area that still needs to be better understood is exactly how  taxonomies should be designed and implemented to be most effective.

Suiting Users Needs

The previous series of articles on taxonomy by Lau addresses many important points about taxonomies including building the business case for a taxonomy, planning a taxonomy, and taxonomy governance. In the first article of the series, “Planning a Taxonomy Project,” she states: “Understanding the users and their tasks and needs is a foundation for all things UX. Taxonomy building is not any different. …Who are the users? What are they trying to do? How do they currently tackle this problem? What works and what doesn’t? Watch, observe, and listen to their experience.”

In this article, I will explain the role of a taxonomy as a tool that connects users to content.

Image of the relationship between content, taxonomy, and the users. The taxonomy sits between users and content, and information flows in both directions.
The taxonomy sits between users and content, and information flows in both directions.

Understanding the users is of central importance, so let’s consider specifically two techniques we can use to make a taxonomy more suitable for its users: (1) adapting the names or labels of the taxonomy concepts (terms) to the language of the users, and (2) adapting the categorization hierarchy to the expectation of the users. The complexity is to do this for multiple different users with the same taxonomy for the same content.

Different Options for Concept Labels

Different users will call the same thing by different names, whether it’s simple synonyms, such as Doctors vs. Physicians or Cars vs. Automobiles, or words or phrases that are not exact synonyms, but close enough,such as Computer security, Cybersecurity, Information security, IT security.

Different labels for the same concept: Computer security, cybersecurity, information security, IT security.
Different labels for the same concept: Computer security, Information security, Cybersecurity, and IT security

Taxonomies, in contrast to mere navigation labels, make use of such “alternative labels” for each concept, also known as non-preferred terms in thesauri. These are colloquially referred to as synonyms, but they are not exactly synonyms; they are labels for concepts that are sufficiently equivalent for the context of the content and the taxonomy. Thus, users searching on any various alternative labels will retrieve the same concept and it’s associated content.

It is a design choice whether the alternative labels are displayed before redirecting to the concept with the preferred label, or if the redirect is without a display and the user is taken directly to the tagged content set. Displaying alternative labels is educational for repeat visitors, whereas no display of alternative labels to end-users provides a clean, quick user experience. Users may not be aware that their chosen name was actually “alternative” and not “preferred.”

When a taxonomy is displayed for hierarchical browsing, only the preferred labels for each concept can be displayed. Designation of a preferred synonym as the label should reflect the wording preference of the majority of users.  

If there are two distinct sets of users, such as employees and customers, where a number of preferred labels vary, it is possible to create two display versions of the taxonomy. This can be a little more complicated to implement because commercial taxonomy management software typically supports just one preferred (i.e. display) label per concept by default. You may need to create two separate taxonomies and link them at equivalent concepts.

Different Options for Categorization

Different users may categorize differently and will look for the same thing in different places. Lau’s articles gave the example of different users of a kitchen wanting to group different ingredients differently. This would certainly be a challenge in sharing the same physical space. 

Fortunately, taxonomies are used to describe digital space so there is flexibility. While a physical object can exist in only one place in a kitchen, a library shelf, or a store shelf, the same taxonomy concept representing an idea may exist in more than one place in a taxonomy hierarchy.

In another example, some people might categorize Financing agreements under Financial documents and some might put the category under Contracts.Thus, we can have the taxonomy concepts of Financing agreements appear as both a narrower concepts of Financial documents and as a narrower concept of Contracts, and all the same tagged documents will be found in both locations. This is what taxonomists call “polyhierarchy.”

Taxonomy excerpt of the concept Financial documents in a polyhierarchy, appearing under both Contracts and Financial documents.
Taxonomy excerpt of the concept Financial documents in a polyhierarchy, appearing under both Contracts and Financial documents.

One thing to keep in mind is that polyhierarchy is appropriate for hierarchical taxonomies, not for faceted taxonomies of attributes or filters (such as ecommerce facets of Size, Color, Material, and Style), where the same concept should exist in only one facet.

Methods of Obtaining User Input

The methods to develop a taxonomy involving users have some similarities and some differences compared to other UX methods. Card sorting can be used to gather user input for taxonomies, but it is effective only for 2-3 levels of a hierarchical taxonomy and is not as effective for designing facets, where the challenge is to identify ways to describe not ways to categorize. Some hierarchical taxonomies have many more levels, so card sorting is most practical for just the top levels, or else it would become too time-consuming for the multiple hierarchies at each level.Taxonomies are more extensive than just the navigation structure of a website.

Users of a taxonomy include both those who are looking for information and those who would be using the taxonomy to tag content. Representatives of these two different user groups should be interviewed with different questions. For example, those who need to retrieve content may be asked questions around the challenges in finding content and search terms; those who tag content may be asked questions about challenges in finding appropriate terms for tagging. Similarly, user testing of the draft taxonomy should also involve both uses of tagging and uses of retrieval.

Content management users, especially those dealing with particular subject domains, may be asked to submit lists of suggested terms that fall into deeper levels of the taxonomy. Those submitting suggested terms should be provided with clear guidelines, that the terms are for tagging content, so that they do not suggest terms that are too specific and not reflected in the actual content. These terms then should be reviewed and discussed with the taxonomist to make sure that they are suitable for the taxonomy.

Another method to gather user input indirectly for a taxonomy is to analyze search logs to identify what words and phrases users have been entering into a search box to find content. These words and phrases should be considered for alternative labels (synonyms) for taxonomy concepts, and possibly for additional concepts in the taxonomy, if warranted by the content.

Conclusions

While UX research is a formal job role, taxonomy research is not, although there are standard practices. Taxonomy research is rolled into the overall taxonomy design and creation job. Because taxonomies are based on the content they are tagged to, taxonomy creators may fall into the trap of exclusively focusing on making the taxonomy reflect the content without also considering the need of making the taxonomy suitable for its users. Taxonomy user research  may not be as formal or extensive as other UX research, but it is critical to the success of a taxonomy.


Heather Hedden will be conducting a workshop on this topic at the 2021 Information Architecture Conference. Participants will learn taxonomy creation principles and how to address the issues of designing a taxonomy to serve users.


Featured photo by KOBU Agency on Unsplash.

Additional images by Linda Ramirez and Heather Hedden.

Focus on Usage Maturity: Part III

by:   |  Posted on

Using a Usage Maturity Matrix to Make Design and Strategy Decisions

Early Foundations

You may recall from earlier installments in this series, that usage maturity is a measure of users’ comfort and familiarity with, and degree of use of, a product, process or place.  

During our master’s capstone research at Kent State, my project partner and I explored the varied levels of usage maturity of participants using Apple’s voice assistant Siri and found usage maturity did not coincide with participants’ tech savviness. 

To illustrate this, think of your own tech savviness and score yourself with one to five stars, five being high. 

Now, think of an app or program that confounds you, that you struggle to learn or use well. 

Do the challenges with that digital experience compel you to lower your tech savviness rating? Probably not, you retain your 3 or 5 or however-many savviness stars, but your usage maturity with that program may be low.

Continue reading Focus on Usage Maturity: Part III

A Digital Response to the Pandemic

How a Small Team used Citizen Centered Design to make World-First Covid Apps by:   |  Posted on

In March 2020, the pandemic stretched across Europe to the UK and Ireland. As the governments of Northern Ireland and the Irish Republic began reporting cases of a deadly new coronavirus strain, both health services began to organize expert teams to take on what would be the biggest threat to public health in generations.

At Big Motive, we have enjoyed the privilege of working with the health service as part of an integrated and multi-disciplined stakeholder team, charged with combatting the virus, breaking chains of transmission, and ultimately saving lives. This is our story about design in response to a pandemic over six months in 2020.

Continue reading A Digital Response to the Pandemic

Usability Testing Guide

The process of conducting a usability test by:   |  Posted on

Usability testing is a core component of User Centered Design and can be used at any stage in the process. It provides valuable insight into the mind of the user, giving us a better understanding of users’ mental models, and it helps to highlight issues that might negatively impact the experience, while also pointing to solutions. If you are new to Usability Testing and want to learn more or just interested in how someone else approaches it, this article gives an overview of how to set-up and run a usability test, and provides a checklist of things to do to complete a usability testing project.

Preparation 

Here is a brief outline of the different stages involved in setting up a Usability Testing session. I will go into each in greater detail and explain what it is and what you need to do.

  • Briefing Meeting
  • Participant Specification
  • Recruitment
  • Discussion Guide
  • Consent Form
  • Setting Up the Session
Continue reading Usability Testing Guide

Collaborative Journey Maps

Moving Agile Product Teams from Magic to Mission by:   |  Posted on

In many organizations, the design team does some research then retreats to their tower to conjure deep magics that turn note filled notebooks into a customer journey map. At least that’s what it looks like to their peers.

IMG scribble, magic, journey map

Journey diagrams capture tons of detailed info about users, processes, and systems. The best teams share the same understanding of the user’s journey. Instead of having your team wonder where you got this information or how you came to these conclusions, have them build the journey map with you.

When you map the user journey with your team, everyone understands what it says and why. When you collaborate with your team, the journey map transforms from the designer’s magic to the team’s mission, representing the journey you shepherd for your users.

Continue reading Collaborative Journey Maps

Taking Research out of the Lab

by:   |  Posted on

To date, usability testing has been largely confined to usability labs. This ensures a controlled environment where users can interact with products or designs and researchers can field questions. The downside of this is not being able to get the context of use of what you are testing. But a recent project for a life science organization cemented the idea that taking user research out of the usability lab yields the best results.

Continue reading Taking Research out of the Lab

Focus on Usage Maturity: Part II

by:   |  Posted on

Meet Users Where They Are, Draw Them Deeper In

If we want users to remain our users, we ought to entice them deeper into our design ecosystem.

Attempts to extend or expand users’ usage, frequently results in designs complicated by added features, and functions. My user experience research has informed digital and physical designs often with an emphasis on correcting the usability of such complexities. Users interact with the things we design at varying levels of usage maturity. Usage maturity is a measure of users’ comfort and familiarity with, and degree of use of a product, process, or place. 

Continue reading Focus on Usage Maturity: Part II

Don’t Send Personal Messages Through LinkedIn (Unless You Want People to Hate You)

by:   |  Posted on

Looking through my parents’ storage boxes, I found letters that my mother and father sent one another prior to my existence. This unfathomable world was decades away from mobile phones, public internet connectivity, and social networking. Along with explanations of humorous or ordinary everyday episodes and proclamations of love, the letters included doodles, crossed out words, and long postscriptums. I don’t know if my mother or father ever dabbed some perfume on their letters hoping to evoke butterflies in the other’s stomach; it would not have been out of place.

Continue reading Don’t Send Personal Messages Through LinkedIn (Unless You Want People to Hate You)

The Lessons Learned Running User Research Interviews

by:   |  Posted on

In the world of user research, no idea is a bad idea.

If you have an idea for a great piece of research, act on it. In fact, your first epiphany is the seed from which all great things will grow. Your idea will eventually shape your hypothesis—your very best idea. This is your proposed explanation based on your current and limited evidence, paving the way from your starting point.

The investigation to follow is where your user research comes in.

Continue reading The Lessons Learned Running User Research Interviews

No! We’re Not All Just the Same

by:   |  Posted on

In May and November of 2018, I traveled to Norway to do user research. I don’t have any depth of experience with Norwegian culture. What follows is my outsider’s view and interpretation. I doubt it’s the whole story.

I tried hard to understand my surprising findings by chatting with Scandinavian friends and by researching cultural norms, but there are always limitations in how much an outsider can truly understand.

I still have more questions than answers.

Continue reading No! We’re Not All Just the Same