Public Library Site

An Information Architecture Opportunity

What design challenge was I asked to solve?

The Upper Sandusky Community Library expressed concerns about patrons experiencing difficulty finding specific library information on their site. This use case applies a fictional client. The purpose of this use case is to demonstrate my problem-solving methods.

“We are in dire need of having our website redesigned from the ground up with emphasis on easy to access information and welcoming thematic elements... while there is an abundance of information, it can be daunting for our users to find that information (i.e. it could be organized much better).”

— Upper Sandusky Community Library

Actions

What was my plan to address this?

Upon reviewing the site, I realized that fundamental organizational best practices should be utilized to make information easier to find. Also through research, I wanted to determine and place emphasis on the tasks and information patrons’ care about most.

The redesign solution not only needed to take into account the patrons’ needs, but the staff’s as well. The library’s resource constraints needed to be considered when coming up with the new solution, making it as cost-effective as possible to implement and maintain.

Start of the Project

I began with a project brief stating the reasons for the project’s initiation. It reiterated the client’s concerns, as well as, additional observations gleaned from a brief analysis of the original site. Among others, the additional observations included items such as:

The site was not responsive, making it difficult to access from various devices.
The use of images for text elements was not compliant with accessibility standards.
The navigation system should have presented the structure of the information hierarchy to indicate the user’s current location.

The brief also included a 6-week time line stating what deliverables, outputs, and artifacts the client could expect to receive.

“A lot of them, it’s to find information. They go to the site to look up books or materials. Look up their account, information about their specific account… what’s checked out, if they have any fines, when the items that are checked out are due back.”

— Participant 1

Research to Understand

I first conducted research to gather information about target user groups and identify their goals, tasks, and what information they’re looking for on the library’s site.

I held interviews with three stakeholders that have first-hand knowledge and experience with public library media, services, patrons, and the website.
I also researched published studies of library site users to gain additional insights, and analyzed competitors’ organizational structures.

Published Studies and Articles:

Zickuhr, Kathryn, Lee Rainie, and Kristen Purcell. “Library Services in the Digital Age.” Pew Internet Libraries, January 22, 2013. http://libraries.pewinternet.org/2013/01/22/library-services/

“Library services used by U.S. online learners on a public or academic library website as of January 2014” Statista, The Statistics Porta https://www.statista.com/statistics/334493/us-online-learners-using-library-services-on-library-wesites/

Chow, Anthony S., Michelle Bridges, and Patricia Commander. “The Website Design and Usability of US Academic and Public Libraries.” Reference & User Services Quarterly, vol.53, no. 3, pp.253-65, Spring 2014.

Personas

During the research efforts, I realized that it’s difficult to keep library users and library website users separate. There are a number of groups of people that use the library for different purposes. However, there are fewer user groups of library websites. Based on the research, I was able to determine the target user groups along with their goals and pain points. The findings facilitated the development of two primary personas and one secondary persona.

Two primary and one secondary persona. (click images to enlarge)

Tasks

From the goals listed in the personas, tasks were created. The tasks have been prioritized in the following table according to the weight of importance and the frequency in which they appear for each persona. These are the tasks I recommended the revised site architecture support.

Content Analysis

Next, I performed content analysis of the current site and compared it to the personas and key tasks. The content analysis helped determine what existing content was reused, reworked, or needed new content to be created from scratch.

“Website wise our demographics, we use Google Analytics, so I look at that every so often, and its about the age group of 25-49 that I’ve seen and mostly women, which we think is basically women bringing their kids in for programs. And that number goes up in the summer. Because they’re looking at events and stuff.”

— Participant 2

Early Design

Once the inventory was complete, I decided on an approach to the classification scheme for labels and taxonomy. I proposed the revised site be a combination of topical with a few task-based labels grouped in the footer. One reason for the combination classification scheme is that it works well for a wide range of site content. Another reason is based on my research of other municipal library sites, I believed the groupings and labels would make sense to the library’s patrons. The labels and taxonomy, however, were tested with target users.

After working out the taxonomy, a sitemap was created to demonstrate access to the key tasks. The plan for the site was to use a simple hierarchy + simple database pattern. The hierarchy + database pattern is good for small sites that only contain a couple levels, but also contain one or more sections that need the power of a database to assemble more detailed information. The project didn’t directly influence the databases in this project, but did tie into third-party database sites.

Proposed Sitemap - Secondary Level Pages.

Tree Test

The labeling and taxonomy from the sitemap designs were tested to ensure online users were able to perform the top ten tasks patrons use the library site to complete. I recruited 5 target users and conducted moderated tests using a UX test facilitation site called, Treejack. Even though these tests can be run un-moderated, the team felt we would get more value in being able to hear participants thinking-aloud, and have the ability to ask follow up questions.

Based on the results and observed hesitations, I collaborated with team members and we refined a few labels in the information architecture (IA). I also decided that descriptive text about databases and download sites would be necessary to fully inform users of the choices they have available.

First-click Task

The adjustments made to the IA were captured in high-level wireframe sketches. A qualitative, first-click study was done to test the revised information architecture. These key workflows were tested using a facilitation product called, Chalkmark. I recruited 5 participants, and conducted moderated testing.

Examples of the Chalkmark Test Questions.

Changes were made based on findings from the testing. The following sitemap comparisons show how 4 of the 5 primary navigation areas are no longer their own landing pages, but rather they lead to sub-navigation choices for greater clarity and focused content. Without the change in content organization including the addition of a few more secondary-level pages, screens would have been overloaded with information.

Team Observations

“...we suspect that library site users are not fully aware of research databases that are available to them as patrons... people do not have a clear mental model of what they are.”

Due to the nature of the test, participants were not able to see the secondary drop-down menu choices from the primary navigation. The database task question did not perform as well as hoped (2 of 5 first-clicked the intended path). Since other team members also observed participants experiencing similar difficulties, we suspected that library site users are not fully aware of research databases that are available to them as patrons. Even though their presence has been elevated to the primary navigation (Research Resources), people don’t have a clear mental model of what they are. As a result, I updated the wireframes with a weekly feature of a different database on the homepage thinking that may help increase patron awareness.

Final Wireframes

Because the wireframes for this project were created for the purposes of testing the proposed information architecture, the desktop pages are the only ones that were used in testing. However, the final set of wireframes for this project include mobile views as possibilities for how the content might work responsively. Further qualitative testing was recommended for the desktop and mobile pages prior to development to ensure the interactions of the pages work for users.

“We’d like to take the content that is on the current site and have it reworked so that it is easy for our patrons to find exactly what it is they are looking for, be it the online catalog, information about preschool story-times or downloading a meeting room application form.”

— Upper Sandusky Community Library

Results

What were the results of my work?

The library’s organizational concerns that initiated the project were comprehensively addressed. In addition to the enlistment of a hybrid classification scheme and the use of a simple hierarchy + simple database pattern for the organizational structure, other IA best practices were incorporated.

Additional issues addressed through the user-centered redesign were:

The site is designed to be responsive, resizing when displayed on mobile devices.
The upper left corner containing the library’s name navigates users back to home.
The navigation system presents the structure of the information hierarchy, indicating the users current location through visual cues and breadcrumbs.
The homepage no longer contains a Flash-based slideshow.
All content is presented in HTML, allowing screen readers and engines the ability to detect it.
All text that is underlined on page load or through hover action is a hyperlink.
Links that open PDFs now forewarn the user that a PDF will open.
Landing pages provide adequate information or explanation about the subject.

Lessons Learned

What if anything did this project teach me about the research, design, or myself?

I understand the basic premise of the first-click test – that if users start down the correct path, their likelihood of completing their task is much higher (87%) than if they click down the wrong path (46%). But when it comes to testing secondary navigation menu choices, this testing approach is frustrating because the participant never sees the secondary menu. Hover might be an option, however, ADA accessibility frowns on the use of hover menus.

Just from this experience of using the first-click testing, I can understand how and why getting the right IA in place for a site or application is a significant effort requiring multiple revisions to find the right solution.