Lo-Fi Prototyping
Build upon the initial UI sketches from our ideation stage, we entered the lo-fi prototyping stage where we transfer our scattered design ideas into a comprehensive, yet basic, screens in online design tools to allow for another round of user testing. To guide the lo-fi prototype stage, we have also come up with a list of questions as our cornerstone:
How do the users perform with our design compared to the original one?
What are the improvements and shortcomings of our design with the original one?
How do we order the sections in the detail page to facilitate an uninterrupted user flow?
Which design of components should we keep when we have more than one of them?
What are some design choices we can bring to our next stage from our user test?
Milestone 3 Activities and Outcomes
Low Fidelity Prototype
Figure. Figma Lo-Fi prototype overview.
The low fidelity prototype marks a crucial stage in our design process, building upon the UI sketches from the previous milestone. In Figma, we recreated these sketches, providing a tangible platform for testing, revision, and refinement towards our final high fidelity prototype.
Our approach involved recreating the individual components from the sketches into Figma, ensuring consistency with our initial design. From there, these components were assembled into screens, reflecting the planned user interface. Additionally, we incorporated interactions into most of the main components, thereby establishing a functional flow that users can engage with during testing.
Throughout this process, we retained the diverse design ideas from the earlier stage, integrating them into our Figma design. This allowed us to explore various design directions and iterate upon them with the help of a small scale user testing.
Included in this report are screenshots comparing the initial UI sketches with their Figma recreations. Furthermore, the Figma design itself will be accessible in the Appendix, offering a more comprehensive view of our prototype.
Omni Search Landing Page
The omni search bar (short for “omnipotent”) is the default page when a user lands on the search page but does not do anything. It serves as the ultimate entry point for all the user flows. It enables search for locations, providers, and other available resources.
Figma Recreation
In the recreation process, we enhanced fidelity by introducing additional tabs with realistic labels. Furthermore, we incorporated an animation for the title that cycles through various functionalities such as "Find a Hospital" and "Find a Provider", before settling on "Find Care". This dynamic animation implies to the users that the system offers comprehensive support in finding care.
Figure. Figma Recreation - Search landing page with two different titles - 1.
Figure. Figma Recreation - Search landing page with two different titles -2.
Figure. Figma Recreation - Search landing page.
Search Result Page
After the user searches with the Omni search bar, they will be brought to a new page showing the search results. This page will be composed of three main parts: Search bar and filters, search results, and interactive map.
The search bar will be placed on the top of the page, however, whether to put the filters as a sidebar or parallel with the search bar is still undecided. The search result will be shown in a list composed with cards.
Beside the search bar, there will be an interactive map in a two column set up. The map will show points of interest on it as pins. Pins will have different icons showing different types of locations, and different colors to show the opening status of the location (open, closing soon, closed). When hovering on the pins, a smaller card of the location will pop up.
Figma Recreation
Most components are kept the same in the recreation of this page, except that all of them have a much higher fidelity. One minor design we have in this page is the dual-color chip for the open status and distance: now before the user provides their location, the distance pill will display “Enter Zip Code”, prompting them to enter their zip code in the top right text entry.
Figure. Figma Recreation - Search Result Page
Cards are recreated with the main design choices and components. The mini and vertical version of the cards had a slightly different design and content to make their sizes reasonable.
Figure. Figma Recreation - Location result cards
Location Detail Page - Introduction
For every location, its detail page will alway consist of a general introduction section. This section will include some image of the location, its name, contact information, location, opening hours, and popular specialties. There will also be a floating sidebar to assist in-page navigation by indicating which part of the page the user is currently reading.
Figma Recreation
The component is recreated without any major changes. Along with introducing a Back button for navigation back to the map, the name of the location has been moved to the top of the page for better clarity. The Make an Appointment button now appears in the providers section.
Figure. Figma Recreation - Introduction section in detail page.
Location Detail Page - Specialty Search
The Service section, as one of the most central features of a location page, is the influencing factor that most users will view and make a decision whether or not to visit the location. In the pre test questions in our user research, users also cited the specialties offered by the location as an important influencing factor.
In our redesign, we have retained the existing specialties listing format that most hospitals have. However, a number of participants in the pre-test demonstrated difficulty in locating this list and were unfamiliar with the specialized medical terminology in the list. Therefore, we propose to add a search field in the Service section as well, which supports everyday language or specialized medical terms as inputs, to help users of all knowledge levels to be able to retrieve the specialties contained in the locations more easily.
If the specialty of the Service search result is provided by this location, a shortcut link to the relevant providers will appear to bridge and facilitate the flow of users confirming the location and confirming the provider. If not provided by this location, a shortcut back to the search page will appear to provide the next call to action after the user flow breaks.
Figma Recreation
This component is created without any major change. Real medical specialties and interactions are added to increase the fidelity.
Figure. Figma Recreation - Specialty search section in detail page, default state.
Figure. Figma Recreation - Specialty search section in detail page, success state.
Location Detail Page - Amenity
One aspect that’s lacking on the current UW Medicine’s website that is vital to a user’s visit is the amenity section. Based on our analysis on hotel systems and our field trip to UW Medicine Montlake, we have come to the conclusion of a tabbed list design for the amenities and include both a detailed parking instruction and a downloadable site map. The instruction and the site map will be two buttons on the same level, one primary and one secondary, right below the section title. Icons will also be added beside the amenities to improve readability. We also made sure to include information we found useful from our on-site visit, such as parking validation and wifi availability.
Figma Recreation
To increase the readability of this section, icons are added in front of the titles within the section. Also, two maps are linked as buttons on the top of the section: one parking instruction map, which will help users see recommended parking lots, and one internal site map, which will help users locate amenities within the center
Figure. Figma Recreation - Amenity section in the detail page
Location Detail Page - Providers
The Providers section shows all healthcare providers within a location and is usually the end point of the user flow where the user contacts a provider or schedules an appointment. This area will also be longer due to the large number of providers at the location, especially the hospital, so it is also the last section of the detail page. However, users can also choose to be directed quickly to the providers section if they found a specific specialty in the previous Services section.
Currently UW Medicine's Providers area arranges them alphabetically by provider name and does not provide any sorting or filtering capabilities. In our initial user testing, we also received feedback that it was difficult to find suitable providers. In order to enable users to locate the desired provider more accurately and quickly, we decided to add filters to the providers area.
Providers will be displayed as cards in the location details page, it will have several filters on top and the available providers below it. Filters will enable users to scope down the providers by the specialties and insurances they cover, and whether they accept new patients or not. We also proposed two types of layout for provider cards display - grid or list, but we will need feedback from usability testing to choose between them.
Information in the providers result will include:
Provider’s photo, name and title
If the provider has board certifications with a universal icon
Their specialties in chips (small rectangular or rounded elements with text or icons used for displaying user-selected options)
Available languages in chips
If they accept new patients in badges on the top corner
The medical specialties filter will display a list of available specialties and the filter will turn into a text entry. When the user was quickly directed from the Service section, the specialty filter will be auto filled with the specialty entered or generated previously.
Figma Recreation
Medical specialty filter was placed in the first place. Accepting new patients filter is now a checkbox rather than a button style. Also, the badge on the provider’s card is now displaying whether they are accepting new patients rather than if they are certified or not. Two layouts of the providers are designed, a grid layout with cards and a list layout, a layout switch is added along with the filters for the users to change if they want.
Figure. Figma Recreation - Provider section grid layout
Figure. Figma Recreation - Provider section list layout.
User Testing
In milestone 3, our focus shifted towards implementing and testing the redesigned features of the UW Medicine website. Our usability testing phase involved a diverse range of participants, including both those familiar with the current UW Medicine system and novices. This approach allowed for a comprehensive evaluation, providing insights into both comparative feedback and fresh, instinctive perspectives. Following the tests, we meticulously analyzed the results, categorizing feedback as "good" for aspects that met satisfaction, "bad" for areas requiring improvement, and "comments" for neutral observations. These findings will inform our next iteration, guiding further refinements to optimize the user experience of the UW Medicine website.
Participants
We aimed to reuse our participants from the initial user testing in milestone 1 for all our user tests in this milestone. These participants were familiar with the current design of the system, allowing them to easily identify changes in our designs. However, due to the constraints in their availability, we were not able to gather enough of our original participants, so we ended up with 3 original participants and 2 new participants. Since we have done a full redesign of the system from scratch, the participants’ prior experience with the original UW Medicine website should not affect their performance and will still provide valuable feedback. The situation of the participants are recorded in the chart below.
Test Questions and Tasks
The pre and post test question and the three tasks in the user test are as below.
Pre-test Questions (Only for new participants)
Multi-select: Is English your first language?
Yes
No
Scale (1 - 5 (1 - very little knowledge, 5 - a lot of knowledge)): What is your knowledge level regarding medical-related words and terminologies in English? 3
Scale (1 - 5 (1 - not comfortable at all, 5 - very comfortable)) How comfortable are you with searching for healthcare providers online? 3
Multi-select: On average, how often do you use UWM’s website in a year?
Never
Less than / about once per year
About once per 3 months
About once per month
More than once per month
Testing Scenario
Task 1
You live in the Seattle area, zip code 98105. One day you missed your step while looking at your phone and you think you broke your ankle on the stairs. You want to find a UW Medicine location that can help you treat your ankle and that’s not too far away.
Task: Search your symptoms and zip code. Read through the results and determine which location you think is best to treat your ankle and that’s within 10 miles of 98105.
What we expect the participant to do:
Enter “broken ankle” in the omni search bar, and enter 98105 in either search bar or the zip code entry
Only one location card has Orthopedics (and the others have vision or immunization)
Find/select the location.
Task 2
Now that you’ve found the location, then see if there are any providers that can help you. You are new to the location so you want to look for providers that are accepting new patients.
Task: See if you can find a provider that treats broken ankles and is accepting new patients.
What we expect the participant to do:
Visit its detail page
Search for broken ankle in the detail page specialty section: “This location provides Orthopedics, which treats broken bones. Click here to see providers at this location.”
Being redirected to available providers
Check the providers cards
Find a provider that have orthopedics as their specialty and accepting new patients
Task 3
Your appointment with Dr. (NAME) is scheduled! One of your family members is driving you to the location and accompanying you during your appointment.
Task: Check if there’s any available parking spot, dining options, and if cash payment is accepted in this location.
What we expect the participant to do:
Use the information available in the amenity section in the location detail page
Confirm the parking, dining, and payment options.
(Link UWM’s available sitemap to the Figma file)
Post-test Questions
Search page
What information do you prefer to see on the search page below the search bar? (Hospital/clinic nearby, care providers, or a map)
Detail page - Specialty
Do you expect the search bar to be auto-filled with the symptoms you searched with the omni search bar (so that you don’t have to type in again)?
If so, would it be confusing since the results are filtered by default?
Detail page - Overview
Do you prefer to have the opening hours collapsed in a drop down or displayed plain?
Detail page - Providers
Do you prefer to have the provider cards in grids or rows? (A/B test provided)
Map page - Popup
What is the ideal size for you for the pop up card?
What information do you think is useful in the pop up?
Do you prefer the popup to have images of the location or not?
Result Analysis
Figure. Screenshot of User Testing Result Analysis FigJam Board.
Search Page
According to the user testing on the Search Page prototype, users appreciate the convenience and versatility of the omni-search bar. The search bar can not only search for locations but also auto-fill common symptoms. It reduces the effort required to type symptoms manually and provides guidance for users unsure of how to describe their symptoms.
Nevertheless, there are still possible improvements that can be made to improve users’ experience. For example, a welcoming title with positive emotion would create a favorable impression, reinforcing brand identity and trustworthiness.
Additionally, multiple users expressed confusion while preserving the suggested contents under the search bar. Wordings like “Online Resources” could be changed to more specific terms like “Health Information” to improve clarity. Consistency in terminology is essential to avoid confusing users.
Last but not least, our internal discussion suggests simplified location cards to improve clarity, making it easy for users to quickly grasp key details. Inside the location card, the usage of plain language for common symptoms ensures accessibility for all literacy levels.
Map Page
During our analysis of user feedback on the Map Page component, we found that the current layout generally meets users' expectations and received positive feedback. Users appreciate the clear layout of the result cards and the intuitiveness of the interactive map, despite its current non-functional state as a mock-up.
However, some issues were identified. First, there are concerns regarding the low contrast in the distance part of the dual-color chip. The dark gray text on a light gray background makes it difficult to read, so we plan to adjust the color scheme to improve readability.
Another issue is the wording "Enter Zip Code" within the dual-color chip. Users find this confusing before they enter their location. To address this, we will revise the wording to clearly indicate that users need to provide their zip code for the system to calculate distances.
Additionally, users expressed confusion about the meaning of different icons on the pins in the interactive map. For instance, some read the pin with a blue triangle as their current location, which actually represents a clinic. We plan to implement a legend on the map to explain the meaning of each icon clearly. Moreover, we will redesign the icons to ensure they are more intuitive and easily understood by users.
Overview
During our analysis of our Overview section of the Location Page component, we found that the current layout was positively received by our testers. The information displayed consistently felt understandable, relevant, and easily accessible.
Regarding the hours displayed, we discussed with our users whether or not they would like the information to be collapsible. Their preferences leaned overwhelmingly toward having a more traditional list of hours, citing unease at not being able to see the information ahead of scheduling an appointment. Furthermore, some users also felt that the bolding of “today’s” hours was not apparent enough, but we believe this issue will be corrected once color is introduced.
Internally, we debated whether or not to stack the address and phone number information as we cannot guarantee a consistent length for addresses, thus making spacing a challenge. We will experiment with this while creating our high fidelity prototype.
Services
In the usability testing of the Service (Specialty search) section, several key insights were gathered.
Participants expressed confusion regarding the instruction text "check below," which led them to expect relevant information within the specialty search section rather than directing them to the providers section. This highlights the need for clearer cues and instructions to guide users effectively.
Additionally, participants found the term "Services" ambiguous for a section displaying specialties, suggesting the necessity for a more precise and targeted name to avoid confusion.
To address users' uncertainty about search capabilities, suggestions were made to provide examples or instructions indicating that plain language and medical terms are both accepted in the search bar.
Amenities
For the Amenities section, we received generally positive feedback from our users, who found the information clearly presented and helpful. However, some were concerned about the section’s placement between the Services section and Providers section, as they felt that the search/filter connection between the two sections was severed by Amenities. In our high fidelity prototype, we will move the Amenities section up to below Overview so as to better show the connection between Services and Providers.
Internally, we discussed how to display the amenities. We discussed a two column format, but felt that it did not allow enough information to be displayed at the same time. So, we will aim for our existing one column format with space to include up to fifteen amenities in our high fidelity prototype.
Providers
Regarding the Providers section, participants appreciated the layout switch feature, with a preference for the grid layout as the default option. This feedback suggests that offering users the ability to customize their viewing experience enhances usability without overwhelming them with unnecessary choices.
Furthermore, participants recommended renaming the badge on provider cards from "new patients" to "accepting new patients" for clearer communication of the providers' availability.
Limitations
During our ideation phase, we planned on redesigning parts of the existing search system that we felt intersected heavily with the Find a Location system, including finding where providers work by name. However, we realized that the existing Find a Provider system, which is not within our scope, was capable of providing such information. In order to make sure we were staying within our sponsor’s scope and making appropriate, well-informed design decisions, we abandoned prototyping searching for provider names in favor of implying the potential for the function’s incorporation into our system. Our low-fidelity prototypes focus more on searching for locations based on location names and user ailments to better serve our sponsor’s needs.
During our user testing, we ran “A/B tests” on dueling component designs our team were curious about. We showed them the same design when testing, then showed them the alternate design, and finally asked for feedback. As our “A/B tests” were not randomized, their results are not quantifiable nor statistically accurate. However, we feel the process of hearing the preferences and rationales of our users was valuable for us to make decisions within the team.
One limitation of our project lies in the low-fidelity prototype, where not all elements are interactive. For instance, features such as the search bar, available results, and back buttons lack interactivity. This constraint may pose challenges for users in discerning what actions are available and where they should click. Although this approach allows us to focus on core functionalities and design principles, it limits the user's ability to fully interact with the system. There might exist other design flaws or considerations that are in the components that we haven’t implemented interactions for. For example, components like the insurance filter are not implemented and thus are not tested by our users in this early stage. Without interactive elements to test these functionalities, we risk overlooking potential usability issues or user preferences that could impact the final product.
Another constraint is the lack of meaningful data in the dummy cards. Since all doctors and most locations are presented with identical information, users may find it difficult to distinguish between them. This limitation could lead to frustration and confusion, as users are unable to make informed decisions based on the information provided. By providing only one meaningful answer, it becomes obvious to users that which one is the right choice, which defeats our purpose of testing the filters and information layout in the results.
Future Plans
In the next phase, we will move to creating a high-fidelity prototype. This means making a detailed and polished version of our design. We'll build on our low-fidelity prototype, adding more visual details and making sure all interactive parts, like checkboxes, dropdown menus, and clickable provider cards, work smoothly and respond quickly. We'll make sure everything looks consistent by using the same colors, fonts, and layout styles. This will help make the prototype engaging and easy to navigate. We'll also add detailed images and branding elements to give it a professional and cohesive look.
Next, we will create a poster to show off our design process and the final product. The poster will highlight key features, how users will navigate the site, and the overall design. It will serve as a visual summary of our project. We will also make a video to demonstrate how our high-fidelity prototype works. The video will show a walkthrough of the user experience, highlighting how to search for locations, use the interactive map, and view detailed provider information. It will show how easily users can move from searching for care to viewing location and provider details.