Analyzing usability of NYCParks.com with eye-tracking behavioral metrics
Using eye-tracking, heatmap analysis and SUS scale analysis to evaluate user behavior and navigation patterns on NYCParks.com, uncovering usability challenges and informing data-driven improvements.
September 2024 to December 2024
Public Sector
3 UX Researchers, 1 Senior UX Researcher
Tobii eye-tracker system, Figma, Google suite and recruitment platforms
My Role
Prepared a research plan and proposal for the study. Recognized the research goal.
Planned and performed recruitment efforts for participants in New York City.
Prepared the script and task as per the research goal. Co-conducted 20+ eye-tracking and interview sessions
Analyzed gaze plots videos, gaze tracker analysis, heat maps and interview responses to generate usability insights and provide data-driven recommendations to improve website usability.
Performed SUS scale analysis for further insights.
Impact created
Through eye-tracking and interviews, uncovered specific user struggles such as navigation inefficiencies and content discoverability, with usability metrics like a 65% task success rate and frequent gaze plot deviations indicating areas for redesign.
Behavioral patterns analyzed: Mapped user gaze behavior, identifying high-focus and low-engagement zones, helping to reframe content strategy for optimal user attention and resource discoverability.
25
Participants tested
110
Heatmap data screens analyzed
Project Overview
This project aimed to evaluate the usability of NYCParks.com by leveraging advanced eye-tracking technology and heatmap analysis. Through 20+ user sessions, the research focused on understanding user navigation patterns, identifying pain points, and uncovering opportunities to enhance the website’s accessibility and efficiency. By combining gaze plot analysis, heatmaps, and qualitative interviews, actionable insights were generated to address challenges such as navigation inefficiencies, unclear content hierarchy, and user task completion barriers.
The findings provided NYC Parks with a data-driven foundation for improving user satisfaction and engagement, setting the stage for iterative design enhancements and a more user-friendly digital experience.
What we observed and why
We were focused on analyzing the usability based on navigation, information architecture and usability issues.
For this purpose, we designed tasks that would first require the user to locate the provision on the website. Here we observed the paths the users took in order to find. For example, for a task- find a group around your neighborhood that performs park cleanups, we observed which navigation path they thought they would find the same. Conincidentally, only 25% of users found the navigation path as users did not associate the task with the navigation bar card 'Get Involved'
Secondly, each task had a specificity that might require them to utilize filters. We chose to study this as filters were present in multiple manners across the website. We later realized that some of these were not easily discoverable leading to user frustrations.
To start with, upon evaluating task-to-time metrics, we identified that tasks with feature visibility/ discoverability issues resulted in longer completion times and increased user frustration.
This project aimed to evaluate the usability of NYCParks.com by leveraging advanced eye-tracking technology and heatmap analysis. Through 20+ user sessions, the research focused on understanding user navigation patterns, identifying pain points, and uncovering opportunities to enhance the website’s accessibility and efficiency. By combining gaze plot analysis, heatmaps, and qualitative interviews, actionable insights were generated to address challenges such as navigation inefficiencies, unclear content hierarchy, and user task completion barriers.
The findings provided NYC Parks with a data-driven foundation for improving user satisfaction and engagement, setting the stage for iterative design enhancements and a more user-friendly digital experience.
The tasks were designed focused on navigation usability and features discoverability usability. We observed
SUS Score Analysis
Below-Average Usability: The SUS score of 57.4, placing the website in the 23.5th percentile, indicates that its usability is significantly below average compared to other websites. P-value analysis shows no statistically significant difference between the population mean and the test mean, highlighting the absence of measurable improvement in usability metrics.
DELVING IN DEEPER WITH EYE-TRACKING…
The analysis revealed patterns of recurring usability issues with the website.
Usability Issues with the Date Filter Visibility
60% of users miss noticing the right Filter panel.
Many users overlooked the date filter at the top right of the page, forgetting it as they scrolled. Without using it, they resorted to manually scanning events, making the search process less efficient. The filter’s low visibility and lack of contextual cues reduced its discoverability and ease of use.
Users forget the Filter affordance due to the long page scroll.
Users often miss the date filter located at the top right and instead scroll through the entire page searching for relevant events. When they don’t find what they need, some scroll back up and only then notice the filter. This results in longer task completion times, increased frustration, and a higher likelihood of drop-offs.
The Filter panel on pages need to be clearly visible at all times.
The Filter panel would need to scroll with the page/ stay fixed beneath the top nav bar to encourage users to engage with the full set of filter options and reduce user frustration experienced in trying to locate what they are trying to find.
In addition, we believe reducing the text density on the page and make all text fields more visually distinct will firstly reduce the cognitive overload.
Search input Inconsistencies and User Frustration with Location Lookup
Users face challenges in address search and park location discovery because zipcode and 'neighborhood' inputs varying drastically across the website.
Most participants experienced difficulty in looking up addresses. The frustration showed when users at the end chose to resort to opening a Google tab and finding the park. We realized this happens because the location input field differs across the website as 'zip-code', 'neighborhood', 'enter address', 'borough' or a physical map. Many users either knew the zip-code of the area they were searching for and not the neighborhood name, and vice-versa creating major user frustration at specific points in the website.
This was a completely unanticipated issue for us, but it became a repeated user issue pattern observed where users expressed verbal and visible frustration.
We suggested streamlining the location input for user intuition and ease.
Based on the behavioral analysis, we suggested that users should always have the option to search an address using both the neighborhood and zipcode, or only one of it. This would accommodate scenarios where users are familiar with the area but may not know the exact ZIP code and vice versa.
Navigation and Naming Inconsistencies
Users are confused by the change in page naming
Users experienced confusion due to inconsistencies between the page names they clicked on and the names displayed on the landing page or breadcrumbs. The lack of standardized naming across text links and breadcrumbs caused uncertainty about their current location on the site. Seen below is an example indicating 3 different names for the same page.
Users are confused by the visibility of system status.
According to Heuristics Violation - Visibility of System Status, the design must keep users informed about ongoing processes through timely and appropriate feedback. However, on the website users experienced confusion due to sudden changes in page nesting across different categories. These abrupt shifts often left users unaware of their current location on the website, resulting in disorientation and frustration.
We suggest
Standardize Page Naming: Ensure consistent and clear naming conventions across text links, landing pages, and breadcrumbs to align user expectations and reduce confusion.
Improve Navigation Feedback: Provide clear visual indicators and contextual cues when navigating between nested categories to help users understand their current location and path.
Please reach out to me for further information on detailed eye-tracking plots and maps, research plan, and client pitch deck.
My key take-aways
Value of Different Research Methods in Insight Generation:
By analyzing plots, maps, and comparing them across the user base, I gained a deeper understanding of how observational and statistical data can reveal different insights. Cross-referencing these findings with interview notes and SUS scores helped me identify distinct typologies and more accurately interpret user behavior.
Experience Leading Eye-Tracking Sessions:
Moderating eye-tracking sessions was a valuable learning experience. From setting up the equipment to managing session durations, I became proficient in both the technical and observational aspects of eye-tracking, which enhanced my ability to analyze user interactions in real-time.
Gaze Plot and Heatmap Analysis:
Analyzing numerous gaze plots and heatmaps sharpened my ability to identify subtle patterns in user behavior. This experience greatly improved my skills in detecting hidden trends and understanding how users engage with interfaces, making me a more effective researcher in visual data interpretation.
