Research

To begin our project, we wanted to become familiar with the 'Find a Location' feature in its current state, hospital website norms for displaying location information, how other industries display location information, and what the true experience of being a patient both online and offline entails. We started by creating a list of guiding questions that helped shape our research process:

1. Who are UWM’s current site visitors?

2. How do site visitors use the location webpages currently? 

3. Why do visitors use the locations webpages the way they do?

4. What from the in-person hospital visitor experience can we bring online?

5. Which user centered design methods will be helpful in our redesign process?

Assumption Map

Our first step was to create an assumption map. Assumption maps involve ranking a team’s assumptions about a design and its users on two axes: known to unknown and important to unimportant. From there, a team can use research methods to evaluate points that have been labeled as unknown and important. Our assumption map provided focus areas for our next research activities, particularly with the in-person aspects of UW Medicine's locations and the design methods of UWM's competitors.

Competitive Analyses

In conducting our competitive analysis, our primary objective was to gain insights from renowned hospital systems regarding their approaches to "Search for a location to find care." By analyzing their design characteristics, usability highlights, and shortcomings, we aimed to extract lessons that could inform our own project in later stages. The lessons learned from our competitive analysis served as a solid and unique foundation, guiding our decisions and ensuring the development of a robust and user-centered final prototype.

We conducted formal competitive analyses on:

• UC San Diego Health (similar university hospital/clinic system)

• Univ. of Pittsburgh Medical Center (similar university hospital)

• Cedars-Sinai (large hospital/clinic system, non-university)

• Hilton Worldwide (global hotel and hospitality brand with thousands of locations)

Informally, we also conducted quick competitive analyses of Zillow's map and filtering system, AirBnB's amenities section, and Nordstrom's search filtering during our ideation phase. These analyses helped inspire our UI layouts.

Notable Takeaways

UCSD Health's Effective Website Layout

UCSD Health's search locations page showcases an effective layout that facilitates information retrieval. With clear sections, organized content, and intuitive search functionality, the page offers users a straightforward means of accessing relevant information. 

UPMC's Omnipotent Search Bar and Search Filters

One notable feature of UPMC's website was its omnipotent search bar prominently displayed on the homepage. This robust search tool allowed users to search for specific location information in simple English and provided customized results.

UPMC's advanced filter options allowed users to refine search results based on various criteria such as zip code, distance, languages spoken, and patient age group. These improve result precision and save users time by catering to specific needs.

Cedars-Sinai's Interactive Maps and Insurance Information

Cedars-Sinai's website is distinguished by its straightforward accessibility to essential information, such as operating hours, contact details, and location specifics, presented prominently alongside interactive maps. This direct approach minimizes the need for excessive navigation, facilitating a seamless experience for users seeking quick access to healthcare services.

Insurance information was incorporated into the medical provider detail pages, which allows users to find their desired therapist easily and avoids unnecessary costs. This feature allows patients to swiftly find which providers are covered under their insurance plans, facilitating a more streamlined and informed decision-making process.

Hilton Worldwide's "Business Card"

When displaying specific hotel information, Hilton uses a layout similar to a business card. They display essential stay information, such as the name of the hotel, location and contact information, and its amenities. This allows the user to quickly get a good sense of if the hotel fits their needs.

Field Study at UWMC-Montlake

Our next task was for us to "get out of the office" and actually see what the in-person experience of visiting a UWM location is like. Most UWM webpages have little to no information on amenities, but serve as a vital part in one’s stay, especially long-term stays, in the hospital. The lack of online details about these amenities could lead to unnecessary stress and inconvenience for patients and visitors, who may find it difficult to plan their stay or visits effectively. For instance, families of patients, who spend long hours or days in the hospital, would benefit significantly from knowing where they can recharge their devices, find a quiet place to eat, or simply rest. 

We conducted a field study at UW Medical Center-Montlake in order to understand the availability, discoverability, and usability of the amenities inside a UWM hospital. We created scavenger hunt tasks for our team, where we needed to locate and complete various amenity interactions and assess our perceived difficulty. Generally, most tasks were rated as a 3/5 difficulty, with 1 being very difficult and 5 being very easy, signaling room for improvement in communicating amenity information.

Notable Takeaways

During our visit, renovations in progress meant that the way to the parking garage had a detour, which was not communicated online. We would learn on-site that valet is complementary, there is a Parking Cashier window for assistance, and that validation is offered in certain clinics. All of these services are not communicated online.

There is currently no information online about the cafeteria. Inside, the staff were very kind and the food offerings were diverse, though all payment was cashless, which was not communicated before entering. The exit also placed us in an area with staff-only elevators, which we did not know would happen and required us to backtrack to the Cascade Elevators.

We found the vending machines once we had exited the cafeteria and ended up by the internal to hospital elevators. We did not find any more machines during our time there.

There are many areas to sit within the hospital, but designated waiting areas were not clearly labeled. Additionally, most sitting areas on the main floor are located around the entrance. This poses a problem for those with mobility issues who may unintentionally find themselves walking down a long hallway with no place to rest.

Initial User Task Testing

Our goal was to assess the current usability of the UWM website in order to clearly contextualize the pain points we had determined through our previous activities. We employed moderated “talk-aloud” usability tests to achieve this goal through observation of user interactions and the collection of subjective feedback. Test scripts and tasks were written and developed by the Locationologists team with input from our sponsor. Six moderated tests were conducted by the Locationologists via Zoom or in person, facilitating real-time, interview-like usability tasks with instant Q&A sessions following the tests.

Tasks were written based off our team's own experiences searching for functions within the existing system. These included the site's intended uses, such as inputting a zip code or a medical specialty.

Some tasks, like were inspired by the difficulties our team faced when trying to use the Find a Location feature. These tasks do not have obvious answers and were used to see if other users became just as frustrated as us. Additionally, some tasks were features our team discovered completely by accident, such as being able to search a provider's name and have their location appear.

Notable Takeaways

Participant B1 is not a native English speaker, so she encountered several difficulties in using the terminologies (eye exam, immunization, etc.) and understanding the parking directions provided on the website. Unclear search bars and confusing layout led to missed information.

Participant CRM1 frequently voiced a desire to call the locations presented for answers rather than use the system. She also struggled with the search bars, stating “Why does it ask me for the location name first? How would I know where I’m going if I’ve never been here before?”

Participant CRM2, being a tech-averse senior user, convinced herself that all UW Medicine care was located only at the four medical centers presented on the home page, causing her to fail almost every task. Similar wording across locations (e.g., campus, center) as well as unclear “Specialties Available” lists caused her to end up at completely unexpected location pages. 

Participant CZ1 struggled with using the location search system and locating information the most. In the sixth task, the participant spent over five minutes trying to ascertain whether a flu shot was available, navigating away from the relevant page due to unclear listings of services.

Participant CZ2 was frustrated that he couldn’t search for open hours in the search bars, and also mentioned that the inability to search common languages for specialty was quite annoying, especially for people whose first language is not English.

Participant DG1 found it difficult to return to previous pages or return to the “Find a Location” homepage from a detailed location page. Discoverability issues were highlighted as DG1 struggled to identify provided specialties and utilized methods like ctrl+F to search for keywords on pages. 

Affinity Grouping

After organizing and consolidating the results of our usability testing, we categorized similar findings to try to obtain more convincing and representative patterns and conclusions. On Miro, we used sticky notes to first list our respective findings and then create groups of similar sticky notes. Through affinity grouping, we categorized our findings into five main categories: UI, Search System, Parking, Inconsistent/misinformation, and New Amenity Additions. 

Analysis

Users often missed important buttons like "See all clinics" and found the proximity of certain buttons bothersome. Navigation issues were compounded by difficulties in moving back to the previous page. Tabs and menus suffered from vagueness in service descriptions and poor discoverability of specialties. Maps lacked visibility and failed to highlight points of interest like parking lots. Location cards suffered from small corner distances and unclear logic in arrangement or filtering.

Users exhibit varied behaviors with the search system, with some preferring to check all locations while others inputted their zip code directly. There was confusion around certain search features, particularly among older users. Medical terminology presented a challenge for users, who struggled with jargon-heavy descriptions without explanations. This was especially true for users who are non-native English speakers. The reliance on medical terms may pose difficulties for users lacking medical knowledge. However, users appreciated the idea of advanced filters for search refinement.

Parking information needs improvement, with users desiring more prominent and understandable displays. Inconsistent and missing information frustrated users, who wished for clearer descriptions, appointment times, and terminology consistency. Users also expressed desires for transparency in amenities such as parking validation and masking requirements, as well as information on urgent care wait times and visitor policies. Suggestions include implementing features like pharmacy wait time indicators and ratings/comments for medical providers and locations.

Future Plans

At the end of our research, we set the stage for our ideation phase by suggesting the following recommendations. Many of these ideas would become constants in our project, influencing our design decisions at every stage and even ending up in our final high fidelity prototype.

Enhanced Search and Map Navigation Features

Consolidate multiple search bars into a single, powerful search tool that uses plain language recognition, and offers advanced sorting systems. This new system should allow users to search for providers, locations, search based on their symptoms in plain language, and search for other online resources as well. In addition, integrate a map in the search page that not only shows the location of healthcare providers but also includes expandable amenities information.

Improve Consistency in Digital and Real-Life Information

Easily and clearly communicate the knowledge of an information desk to the user, regardless of whether or not a center or clinic has one. Update and simplify information provided on amenities for all locations, especially concerning parking rules and cashless payment systems. Inform hospital visitors of 21st Century amenities such as Wifi and availability of charging outlets. Ensure that a location’s “Medical Specialties” section matches what the searcher expects in order to direct them to proper care. Similar wording can be used across locations, but must not confuse users into believing they are at a different location. 

Information Display Optimization

Ensure that all services provided by a location on the detailed page are clearly listed and categorized. Some important information identified by our users and us that are missing or hard to find in the current system includes: 

• providers’ card info. 

• medical services provided. 

• insurance coverage. 

• existing and new location amenities. 

To improve discoverability and emphasize information hierarchy, consider implementing icons, accent color, filters, and separate sections for different types of services to help users recognize and navigate directly to the information they need.

Accessibility and User Assistance Improvements

Increase font size for essential information and use bolding and color coding to highlight key details such as insurance options and appointment availability. What’s more, medical jargon can be a major barrier to understanding for many users, especially those without a healthcare background. Integrating an AI chatbot to assist users with clear definitions for medical terms used across the site could enhance user engagement and support.

Appendix I: Full Research Report

Includes full competitive analyses, full field study, and full initial user testing results.