Role

UX Research (User Testing, Data Analysis Strategy, SUS), UX Design

Team

Myself, 3 UX Researchers

Tools

Figma, Tobii Eyetracking Equipment & Software

Timeline & Status

Jan 2025 - May 2025, 5 months

Overview

SVA Library has received feedback that the library's research guides are challenging to navigate, preventing users from finding what they need.

Our team conducted 12 moderated usability tests with Tobii eyeracking software and I lead the data analysis to identify a key mis-match between users' search behaviors and the research guides' content organization, contributing to an overall poor user experience.

From these insights, I re-designed the Design Archives research guide with stronger information architecture and visual heirarchy to serve as a template for the standardization of research guides across the platform.

Context

What are the SVA Library Research Guides and what "resources" do they hold?

The SVA Library Research Guides are curated collections of digital design archives, academic journal databases, and other media repositories to help SVA students jump-start their research projects.

Goal

Improve usability of the SVA Library Research Guides.

Understand how users currently navigate research guides to find relevant resources.

Make feasible recommendations to improve information architecture and usability of the library research guides.

Methodology

Data Triangulation by combining behavioral and attitudinal data

Observe behaviors through moderated testing with eye-tracking

Using Tobii Eye-tracking software, we can non-invasively capture where a user's eye moves on screen while completing tasks, helping researchers identify what page elements draw or hold attention.

Analyze satisfaction with System Usability Scale (SUS)

It's the gold standard.

Uncover insights and inform solutioning with Retrospective Think-Aloud (RTA).

Using RTA helps preserve natural behaviors, while still allowing interviewers to ask further "Why" questions (Neilson Norman).

Reduce bias in analysis with the Rainbow Spreadsheet

A structured, transparent and repeatable workflow.

First Research Insight

Students are "Search-First", and the current search experience does not meet their needs.

After conducting 12 interviews (8 desktop, 4 mobile) with current New York City-based art students and analyzing the data using the rainbow spreadsheet method, we discovered:

67%

of participants mentioned preferring to search over scan for resources

50%

of participants tried to use search to find relevant resources

33%

of participants who used the search function got relevant results

“I wish that the search bar can just immediately take me there. I don't want to spend time looking at everything.”

- SVA Student

Participant

Second Research Insight

When search fails, scanning for resources is not a good experience.

Gaze Plot of SVA Student Participant on the Online Resources Research Guide

Users read headings only - they don't read body descriptions at all.

The gaze plot included to the left shows a high density on the left-hand area of the screen where the resource headings are and a distinct lack of activity on the describing paragraphs.

67%

of participants expressed experiencing information overload and said there was too much text on the pages overall

75%

of participants said they tried to find relevant resources by scanning for keywords

Page organization is not helping students find resources either.

Gif of a participant's gaze jumping across columns

In the gif to the left, the participant scans resource titles across columns and groupings, independent of grouping labels.

83%

of participants either did not notice group labels or found them unhelpful

Bottom Line

Users do find helpful resources most of the time, but the experience is frustrating & inefficient.

Despite a reasonable task success rate, the library's research guides scored in the 6th percentile for usability.

79%

Task Success Rate

41.3

System Usability Scale

Recommendations

Match users' behaviors and expectations with an improved search experience and scannable content.

Before

After

First Priority

Improved search experience

Because users are search first, the search experience needs to be improved to better handle long-form search queries and typos.

Second Priority

Group content with tabs and larger headings to focus user attention.

Tabs help users filter content by search interest. Noticeable headings within each section improve navigability.
.

Redesigned Design Archvies Page

Third Priority

Individual resources should have more visual hierarchy and list keywords relevant to the resource for easier scanning.

In the redesigned resource, the images, icons and title are larger to improve scannability and keywords are included to help users recognize what they're searching for.

Before

After

Impact

“Great recommendations from this group for reducing item and database descriptions on our guides, and for a better way to organize content on the exemplary guide, Design Archives. We can use these suggestions to apply system-wide updates to all of our guides.”

- Digital Services Librarian

Client

Takeways

Bringing multiple data sources together paints the most meaningful picture.

Any of these data sources alone is not enough to reach a conclusion, but together they paint a compelling picture of the user experience. We can make more confident claims and win better buy-in from stakeholders when we can use data triangulation to uncover insights.

Helping students find relevant resources faster on SVA's Library Research Guides