Surfacing key visitor information on a museum website through improved content hierarchy
Dia Art Foundation Unmoderated Remote User Testing
Overview:
Tool(s) Used:
Full Report:
Scope of Work:
Team:
Year:
Duration:
THE EVALUATION AT A GLANCE
Testing the Visitor Planning Experience on Dia's Website
We conducted an unmoderated remote usability test with nine participants to evaluate how well the website supports visit planning.
Four recommendations were developed to improve content hierarchy, navigation clarity, and information discoverability. All scoped as quick wins that don't require a full site redesign.
SCENARIO
The Starting Point: Accessibility, Events, and Artwork on View
We designed a scenario around a realistic visitor goal: planning a first visit to a Dia location, specifically Dia Beacon, with family and a friend who uses a wheelchair. This scenario reflected a prospective visitor who needed to confirm logistics like accessibility, event schedules, and what's on view before committing to the trip. Three tasks mapped to those questions:
Task 1 - Accessibility: Is Dia Beacon wheelchair accessible?
Task 2 - Events: When is the next free Saturday Studio program?
Task 3 - Artwork: Is Richard Serra's Torqued Ellipse I (1996) on display?
FINDINGS
Nine Prospective Visitors, One Pattern
By the numbers:
9 participants participated
6 / 9 completed the accessibility task, but all 9 struggled
3.8 / 5 average difficulty rating for the accessibility task
1 participant concluded the museum was not wheelchair accessible (it is)
1 participant failed the events task entirely
Severity Heatmap
This heatmap shows how each of the nine participants experienced six issue categories during testing. Each cell is color-coded by self-reported severity. Patterns across rows reveal which issues caused the most widespread friction.
Note: Symbols are included in the color severity sections for accessibility purposes to not rely on color alone to communicate findings.
THE CORE PROBLEM
The Information is There but the Structure Hides It
Every answer participants needed existed somewhere on the site. The problem was a content structure that assumes visitors already understand how Dia organizes its digital presence. Three patterns drove most of the friction:
Dia's homepage represents the foundation as a whole. Dia Beacon has its own sub-site, but 88% of participants didn't realize the location cards on the homepage were entry points.
They scrolled through foundation-level content looking for Beacon-specific details that weren't there.
Location cards had a hover effect on the image only, not the title or description
No button, link text, or visual cue signaled the cards were clickable gateways
Users defaulted to the top navigation, which routes to foundation-wide pages
The original “Locations & Sites” layout featured subtle interaction cues that led to user navigational oversight.
Accessibility information lived inside a collapsed accordion labeled "Accessibility and Service Animals" near the bottom of the Dia Beacon page. Two barriers made it nearly invisible:
Browser Command+F, which many users turned to, can't index text inside collapsed accordion sections
The site's own search returned zero results for "wheelchair"
33% of participants tried Command+F as a last resort, and 55% nearly abandoned the task.
The Control+F feature is failing to locate accessibility information nested within a collapsed section.
The calendar's "Type" filter uses internal categories that don't match the terminology visitors encounter elsewhere. "Saturday Studio" falls under "Learning Program" - but on the Families page, it's listed under "Family Programs."
One participant tried "Special Event," then "Studio Event," then resorted to Ctrl+F
Another couldn't connect "studio event" with "learning event" conceptually
The filter forces trial-and-error instead of recognition
Labeling inconsistency between the "Type" filter in the Calendar page and the Families page.
RECOMMENDATION 1
Give Visitors a Clear Way In
Add a "More >" button to each location card and extend the hover state across the full card - image, title, and description. A standard web convention that immediately communicates clickability.
Comparison showing the addition of a “More >” button and a full-card hover animation to improve the discoverability of the museum sub-site.
RECOMMENDATION 2
Let Visitors Jump to What They Need
Introduce a persistent sidebar navigation listing all sections on the Dia Beacon page. Instead of scrolling through the full page hoping to spot the right accordion, visitors can jump directly to Accessibility, Family Programs, Directions, or any other section.
Eliminates scroll fatigue on a content-heavy page
Makes the full scope of available information visible at a glance
Reduces reliance on Ctrl+F as a workaround
The contextual sidebar navigation provides direct "jump-to" links to improve wayfinding and reduce overscrolling.
RECOMMENDATION 3
Show Visitors What Each Filter Actually Contains
Add hover-state tooltips to each calendar filter option. Hovering over "Learning Program" would show: "Explore artist-led Family Programs for all ages, including Saturday Studio and Play Sets."
Bridges the gap between internal terminology and visitor expectations
Doesn't clutter the interface for returning visitors who already know the system
Aims to prevent the most common trial-and-error behavior
Proposed hover states providing descriptive guidance for each filter category.
RECOMMENDATION 4
Put Accessibility Info Where Visitors are Already Looking
This was the most significant finding. Every participant struggled. They checked Visit, Location Services, Visitor Guidelines, About, Support, and FAQ - most found the answer only through the footer link or by scrolling to the bottom of the Dia Beacon page.
Two Changes:
On the Dia Beacon page: Add an "Accessibility at Dia Beacon" summary block positioned directly after Admission. It includes:
A one-line confirmation that the museum is wheelchair accessible
A scannable list of key features (parking, galleries, bathrooms, elevators)
A link to full accessibility details
Dia Beacon page with accessibility information prioritized.
On the Visit page: Add a short accessibility note under each location's overview, so visitors scanning the page can immediately confirm whether a site meets their needs.
Accessibility information listed under location information on the Visit page.
Introducing accessibility cues on the Dia Beacon (location-specific) page & Visit page can help users quickly confirm whether a Dia location meets their needs with less trial-and-error navigation. This layered approach improves visibility of accessibility information and supports scanning. It will build user confidence through repeated consistent information placement, enabling users to quickly identify accommodations and reducing friction in the visit planning process.
BEYOND THIS REPORT
What These Changes Don't Solve: Site Infrastructure Problems that Need Bigger Investment
Search infrastructure needs to index collapsed content, handle typos, and return relevant results - seven of nine participants hit search failures
The overall IA needs to clarify the relationship between the foundation site and location sub-sites - this was the root cause behind most navigation confusion
Terminology needs standardization across navigation, filters, and content pages - this category had the highest average severity (3.6 / 4)
What I’d Do Next
Run a follow-up usability test to measure whether the recommendations reduce task difficulty
Conduct a card sort to validate the proposed content hierarchy against visitor mental models
Benchmark accessibility information placement against peer institutions
Work with a Dia team to standardize terminology across navigation, filters, and page content
REFLECTION
What This Project Taught Me
Content discoverability is an accessibility Issue.
This project reinforced that, and not just for visitors who need to ensure accessibility of the physical location. When anyone can't locate basic information like event schedules, what's on view, or how to plan their visit, the website is creating barriers to participation and engagement. The wheelchair accessibility task made this most visible, but the same structural problem affected every task: content existed, the architecture just didn't surface it where people looked.
Synthesizing nine think-aloud recordings sharpened my ability to hear the same frustration expressed different ways and recognize it as one structural problem. Translating those patterns into targeted, implementable fixes was the most valuable part of the process. Working within a "quicker wins" constraint shaped every recommendation: respect what's already there, and make it work harder for the people it's meant to serve.