Synopsis:

Sno-Isle Library (Sno-Isle.Org) is a library system serving Island and Snohomish Counties of Washington serving roughly 836,000 people across the two counties. During the Summer of 2016 I was asked to do some initial user research for the website to figure out discoverability and engagement (Phase I). I was then asked back in Summer of 2017 to expand upon my research findings and recommendations (Phase II).

My Role: User Researcher

As a User Researcher I:

  • Created a project timeline and research plan in collaboration with Sno-Isle Library.
  • Developed a deep understanding of the business goals for Sno-Isle Library in regards to its website.
  • Cooridnated and scheduled user interviews and testing.
  • Remotely conducted interviews with library patrons around Snohomish and Island counties of Northern Washington.
  • Consolidated insights from users to find key tasks to test on the library's website.
  • Tested users' ability to complete tasks on the library's website within a timely manner.
  • Delivered usability recommendations to internal library stakeholders for website improvement.


In Phase I I acted as lone researcher. Phase II: I worked with Kristin Kinnamon of Kinnamon Communications. Here we acted as a research team sharing duties.

Business Case

Improve the Discoverability and Engagement across the Sno-Isle Website's non-catalog pages. Sno-Isle wanted to provide customers with relevant resources to improving their library experiences, and highlight resources they may not have know were available online.

Project Scope:

Phase I:

  • Determine how users discover information in non-catalog pages;
  • Understand inherent value of library's website to non-catalog pages;
  • Discover any points of friction (if any) within users' experience on the Sno-Isle;
  • Library non-catalog pages; 6 Months in 2016;

    
Phase II:

  • Define users of Kids' pages on Sno-Isle Library;
  • Understand mental model of users of Kids' pages of Sno-Isle Library;
  • Determine how to improve users' experiences navigating the non-catalog kids' pages;
  • Six months in 2017

Users:

Phase I:
Sno-Isle Library serves roughly 836,100 people across the two counties in Northern Washington. The population that come to the library's website and use its services varies across every kind of demographic variable.

Google Analytics revealed that the majority of the users (73%) who engaged with the library's pages tended to be women across all age groups, speak English, and use desktops (80%).

The Library charged me with making sure that the website worked for everyone that would come to the library's website to find information within the library or about the library.

Phase II:
Findings in Phase I required Phase II to focus specifically on parents and guardians of children. The team here primarily focused on parents/guardians with children aged 0-12 that used the library.

Pain Points

The Library's Non-Catalog pages were once housed in different departments across the organization. As a result, non-catalog pages didn't prioritized different functionality and information as important. As the website grew, this led to confusion about where to find certain information on the website. This Phase of the research revealed that users had particularly difficulty:

  •     Finding books for your children to read
  •     Finding information about the town history
  •     Finding information about starting a small business at home

    
These tasks were flagged as particularly important services that the library wanted to offer to its patrons.

Phase II delved deeper into the navigational issues of the children's section of the website. It showed that:

  • Confusing terminology for finding children's books;
  • Multiple listing of children books pages that led to confusing lists;
  • Subpages didn't meet expectations of parent's wants or goals

Research Questions:

Phase I:

  • What are user goals when they come to the non-catalog pages of the library?
  • Can users complete their tasks when they come to the non-catalog pages of the library?

    
Phase II:

  • How does the navigation of the kid's pages help or hinder user goals?
  • How has the new navigation structure helped users ability to discover information on the Sno-Isle website?
  • What information is important to parents on the kids pages of the library?
  • Where are users disengaging from the library's kids pages?    

Methods and Process

    Discovery Phase
        Behavioral Data Analysis
                Traffic and Search Analytics
                    Google Analytics, User Flow through websites, bounce rates
                Content Analysis
                    IA Structure
                    Page Level Structure
                Stakeholder Interviews
                    Executive Team Interviews
                    Lead Librarian Interviews

User Research

  • Conduct Video and phone interviews with Subject Matter Experts (SMEs) about pain points, website goals, known user data
  • User Survey sent out to users throughout the Sno-Isle Library district through mailing lists. Analyzed the data about user goals and behaviors.
  • 30 - 60 Minute long user interviews to determine goals of users when using the website, engaging with the library.
  • Time on Task - Test usability of overall navigation structure by using a time on task test. Tasks were ranked by success, fail, and partial and then averaged.
  • Test old and new navigation structures using card sorting and tree testing to improve the information architecture.
  • OptimalSort – Conduct closed card-sorts with staff to ensure that the classification scheme could account for all the existing articles in the tag system for support staff to use.
  • Ethnographic Research - Follow Users of Children's Pages to understand their customer journey using the site
  • Treejack – Librarians tested the proposed new navigation structure to see if they could find information clearly.
  • Test both existing website navigation and the proposed new structure in two separate research studies with users and support staff, in order to gauge findability improvement.
  • Conduct two rounds of testing on the new IA to refine the navigation labels.
Screen Shot 2016-12-15 at 1.45.52 AMScreen Shot 2016-12-15 at 1.45.52 AM

Challenges

Definition of Scope: In Phase I, Sno-Isle knew they had a problem but were not sure what to look for or how to go about determining the problem. Many of my initial challenges were just doing discovery of determining if their knowledge of the problem was correct and what to do about it. In Phase II, with several problems clearly highlighted, it became imperative that my team focused on solving one problem (navigation through the kids pages) at a time.

Resolution: I worked with the Web Team and the Communications Team to understand the scope of the problem they considered. After doing internal stakeholder interviews, I reported back to the Communications Director and the Web Team of my general understanding of the problem and suggested courses of actions to diagnose the problems.

Participation and buy-in for internal users: Support staff needed to participate in testing to make sure the navigation and naming built for users worked for them too.

Resolution: During my interviews with the known internal stakeholders, I identified secondary stakeholders and took time to work with them. After listening to their concerns, I got buy in from several stakeholders. As we were testing new navigational systems, I came back to them and gathered their insights as well.

Outcomes

Clearer understanding of user goals: before coming on to the project, the web team and the communications team had no idea how users engaged with their website, or what their goals were. After this investigation, we understood that users come to the non-catalog pages to find books, particularly books Librarians recommend.

Restructuring of Information Architecture: Menu Navigation and layout of the website changed to accomodate user goals of finding recommendations.

This project convinced internal stakeholders to green light Phase II: Phase I's investigation found significant difficulties with finding kid's books recommendations. This would become the focus of Phase II's research.

Phase II:
Understanding of parental goals: Through contextual analysis, ethnographies, user interviews, and usability testing, we were able to determine parental goals and have the library support them.

Changes in "reading level" filter options: One point of confusion that consistently came up was the options provided for filtering results for parents. This led to changing the filter options for parents when searching for books for their children.

This project convinced the Library to invest in Phase III and lead to a complete overhaul of the navigational structure.

Insights

Things move slowly in governmental organizations. Between Phase I and Phase II of research, a year had gone by. There were many internal politics that I avoided as an outside consultant.

The website is a reflection of the internal culture, especially at larger organizations. Since historically different pages were "owned" by different internal subject matter experts, making significant changes to the IA or navigation required me consulting across the organization with stakeholders spread throughout the org, often in physically different locations.

Collaboration with stakeholders is essential when research is difficult and findings may be controversial. It's important to have buy in from multiple stakeholders as you begin your process.