UX in Libraries with a Case Study at NYU Libraries

By April Ibarra-Siqueiros

The field of User Experience (UX) deals with meeting exact user needs and focuses on facilitating a seamless customer or patron experience. It takes account of the emotional aspects of how users interact with a service or product and measures the quality of the experience. Much UX work is done for online interfaces, such as websites and mobile apps, but holistic UX includes other aspects of a service, such as physical space and marketing. As a broader concept, UX encompasses various approaches, such as usability testing (determining ease of use), information architecture (categorizing information and creating a meaningful structure for it), and interaction design (imagining and continually re-evaluating the interaction between a customer and a product or service).

The process of applying UX methods depends on the type of project and available resources. One model that covers various ways to meet user needs is the User-Centered Design (UCD) Process, which includes various stages of UX design. Some key steps in this process are:

Research: this can mean doing user research to get to know the target audience better; this can be done in the form of a survey or a focus group.

Design: determining website requirements and content, some examples of this might be conducting a card sort exercise where users group things into categories that make sense to them or defining the information architecture of a site and creating a site mock-up or prototype.

Evaluate: user testing, Examples of this are formal usability testing, guerrilla testing, and individual interviews.

The process may also move from researching to evaluating and then designing if a website is already published. One thing to remember is that rather than being strictly linear, this process can be more of a loop because after evaluating, the design may need to be refined, which in turn may need to be tested again. This repetition is referred to as iteration. Above all, it is important to maintain empathy for users. In UX this is key because it helps the designer to better understand the human side of site users and subsequently design with that in mind.

While UX design is rapidly expanding among established businesses and startup companies, libraries are starting to think more about how it fits into the context of the world of library and information science. The field already deals with the categorization, structure, and organization of information, and, in libraries user needs are a primary focus. For this reason, UX can be an indispensable approach to meeting patron needs. It is also important to note that UX design is quite flexible in its application and thus accessible to all kinds of libraries

This past spring, I was in a team with 5 other Pratt Institute UX graduate students that worked on projects for online library services at New York University (NYU). We were supervised by Nadaleen Templeman-Kluit, Head of UX at NYU Libraries. The projects included an information architecture redesign of the library website, an evaluation of the EBSCO Discovery Service, library access permissions page, and LibGuides. I was a part of the group that worked on LibGuides, which is a popular cloud-based content management system from Springshare used to create various types of guides, and publish university department websites. The resources hosted on the LibGuides CMS are sometimes known as research guides, subject guides, and course guides. NYU Libraries uses LibGuides to host over 550 guides and was looking into upgrading to Version 2, which offers sidebar navigation tabs and responsive layouts. Virtually all of NYU Libraries’ guides are different from each other because they vary in layout and content. Without one definitive template our team faced the problem of testing over 500 interfaces. We had access to analytics data but this information only told us what guides were being used and how many times.

When the LibGuides group, consisting of my classmates Allison Hall, Carolyn Li-Madeo, and myself, approached this project we decided to start by defining the goals. We wanted to know two things: what NYU student research behavior is like and what students know or don’t know about the guides on the library website. Our testing goals were to gather feedback about the current LibGuides interface and its navigational elements. We determined that a focus group and guerrilla usability testing were the most appropriate ways to gather data about user needs and what those users encounter when using guides.

To recruit students for the focus group we used a list of emails that had been accumulated from user testing for other projects, and put up flyers around the library. Two incentives to participate were offered (pizza and a $25 iTunes card) for participation in the one-hour session, which was divided into four stages: research behavior; general questions about various types of guides; LibGuides interface; closing remarks. The focus group started with questions that would help us understand how students do research online. We progressed to questions about what their knowledge and general awareness of guides, and finally explored a selection of guides with them on the library website.

Figure1-FocusGroup Figure 1: LibGuides focus group recruitment flyer

After the focus group concluded, we set up a table and two laptops in the Bobst Library atrium to do guerrilla testing. Offering granola bars as an incentive to participate, we conducted five-minute interviews with students. (We nicknamed this approach “speed dating” because of its formal testing qualities and rapid nature.)  Each session was divided into three sections: a pre-test questionnaire that asked students to provide demographic information, a test in which we asked students to explore a guide of their choice and think out loud so that we could better understand their process, and a post-test in which students filled out a survey about their experience of using a guide. With this testing method we were able to gather 27 interviews in two hours.

After synthesizing the data for both the focus group and guerrilla testing, our major findings included the following:

  • Students often rely on websites like Google Scholar and Wikipedia as a starting point because they find them straightforward and direct.
  • Students were also more likely to use guides if they were a part of a course curriculum.
  • Most students were not fans of the tags and tag clouds in guides
  • External links in the guides opened in the same browser tab, which disoriented users.

This information allowed us to make recommendations that would resolve problems and address user needs. We created recommendations in the form of a style guide that addressed four main areas: navigation, visibility, usability, and consistency/branding. To further illustrate the recommendations, we inserted a style guide in Version 2 of the LibGuides CMS and created a redesign mockup of the NYU LibGuides homepage (Figure 2) that incorporated user feedback and a clearer starting point. Following the UCD Process, we did user research through our focus group, evaluated the NYU LibGuides interface with guerrilla testing, and redesigned the interface by creating a mockup. The full reports of all four NYU Libraries UX Lab projects can be accessed and downloaded so that others can read about our process and methods. We hope that our work will help others to implement their own testing.


 Figure 2: A redesign of the NYU Libraries LibGuides landing page

After the end of the Spring semester, the team presented all four NYU Libraries projects at a Metropolitan New York Library Council (METRO) event called, “User Experience Design for Academic Libraries: Project Reflections, Methods, And Models with the Pratt / NYU Libraries UX Practicum.”  We received positive feedback and were told by audience members that their libraries were facing similar usability issues. Some libraries have dedicated UX librarians or departments, while others at least know the value of UX and are beginning to incorporate these concepts with existing staff. While UX methods do take time to plan and implement, some methods, such as guerrilla testing, are relatively quick and inexpensive. More importantly, providing this kind of platform for users to be heard and to participate in the process of designing library services and products gives patrons a more meaningful experience.


Figure 3: LibGuides Version 2 style guide