.png)
DIRECTORY DESIGN
This application gives folks experiencing homelessness an opportunity to locate local relevant and time-sensitive resources
CONTEXT AND OVERVIEW
This project piece encapsulates my senior capstone design project, where, as a team of students, we worked on an application called Upswyng. It is a mobile directory application where any person or family at-risk of, or homeless can access accurate up-to-date, and relevant information and resources in the Boulder, CO area​.
Upswyng was created by a group of folks that saw a need in the large homeless population of Boulder, Colorado, and came together. One of the founding members was previously homeless and therefore had insight into resource accessibility level, was able to lend a unique perspective, and had a vast list of resources to be included in the application.
ClienT
Upswyng in Boudler, CO
DURATION
Aug - Dec 2018
Skills
User Testing, Wireframing, Information Architecture, User Research
FOLKS INVOLVED
Teammates:
Located in West Lafayette, IN
-
Abby Perez
-
Ferdinand Macatangay
-
Julian Amado
Upswyng Team:
Located in Boulder, CO
-
Amalie Barras (Product Owner)
-
Mike Homner (Project Champion)
-
Varun & Jacob & Ameya (Technical Teammates)
THE OPPORTUNITY
Problem Statement: Each of the vast number of providers for individuals and families in the homeless community, has their own wide range of resources, making it difficult to navigate and access these resources.
​
Opportunity: Those in the homeless community of Boulder, as well as in many other cities, have access to mobile devices and Internet connection.
​
As the Purdue Team, we aimed to...
-
Understand our users better through multiple rounds of interviews and testing
-
Continuously engage in research including, but not limited to, competitive analysis & analogous research
-
Design for optimization by considering low-spec devices and low-bandwidth connections
​
Intended Outcomes
-
Create all user flow screens for an application that could eventually be a progressive web app
-
Help the homeless population become more aware of their local resources
-
Help providers further their reach/effectiveness
​
BACKGROUND RESEARCH
Technology
It was found that 94% of the homeless population owned a cellphone, but it was noted that the Android operating system was more prevalent among those with devices.
Law
There are many laws that have nuances around what can and cannot be done if an individual is living on the streets. For example, there is a law that says covering oneself with a shirt is allowed - but using a sleeping bag is not..
Point in Time Survey
The "Point in Time Survey" is conducted each year to understand the current homeless population of Boulder, Colorado. It only represents a portion of the homeless population though, as it is not mandatory.
*(592 individuals, ranging in age, ethnicity, and occupation, agreed to participate in the 2017 survey)
​
Coordinated Entry
Coordinated Entry is the process that one has to go through to be eligible for shelters, recourses, etc., by the city of Boulder, Colorado. There were over 1,700 individuals screened in a 6 month period for Coordinated Entry in 2017, indicating that there is a much larger population experiencing homelessness than the "Point in Time Survey" shows.
INSIGHT - There is a wide range of people who become homeless and many reasons why.
Analysis & RECRUITING
Competitive Analysis
There were 5 applications found that served a similar purpose but could be split into two different groups, based on who the user is and what it could provide.
​
​
​
​
​
​
​
​
​
Recruiting in the West Lafayette area
Our Colorado client had a comprehensive list of the resources available in Boulder, Colorado, but our Purdue team needed to connect with the local resources (shelters, soup kitchens, etc.) in the West Lafayette area so that user testing could be conducted in a safe and conducive manner.
​
Resources and potential places for user connection included: Lafayette Transitional Housing Center (LTHC), Lafayette Urban Ministry, Lafayette Crisis Center, Familly Promise of Greater Lafayette, YWCA, Community Foundation of Greater Lafayette
​
Testing with initial existing screens
Our client partners at Upswyng had a set of screens already developed for the mobile web application, so it was crucial for us to test first with these to understand where it was and wasn't working.
Homeless can connect w/specific resources
Individuals can help the homeless
ROUND 1 of TESTING - YWCA
Testing Location: YWCA of Greater Lafayette, IN
Users in Testing Session: 3 individuals, age range from 20-50
Testing Goals: Gain specific feedback on the calendar, map features, and notifications in-app. As well, ask for any additional features or icon ideas that might be helpful if added.
​
Results:
-
Legal help button could be helpful for those going to court and needing help in finding an attorney.
-
Insider knowledge/education button could be helpful for legal or local tips such as if police are able to help get individuals into shelters after hours.
​
Insights:
-
Individual user needs change based on the length of homelessness and location of homelessness (transient-ness).
-
Certain features are helpful specifically to people new to homelessness and the area.
RESEARCH & WIREFRAMING
Analogous Research
Apps like IKEA, Yelp, Travel Advisor, Visit A City, and Google Maps were evaluated to find inspiring ways to design this application in slightly different contexts but that had similar end/experience goals. In these mobile apps, we found many different ways to design an app. Some examples included “cards," location, favorites/bookmarks, easy map access, filter by icon, reviews, and details page.
Speed Sketching
Letting our analogous research findings drive us, we dove into multiple speed sketch sessions. At the end of each, we would chat about what could be carried over from the original app design and what new aspects would be included from our sketches.
2 Sets of Mockups for Testing
There were so many wonderful features that came from our sketching sessions we decided to do a bit of A/B Testing in the next round, to ensure we got to test the features we all thought might lead to the best possible user experience.
So with not one, but two sets of newly created mockups we decided to head into our next round of user testing.
USER TESTING @ LTHC
Testing Location: Lafayette Transitional Housing Center (LTHC) Lafayette, IN
Users in Testing Session:
4 individuals, age range from 35-50
Protocol Script: Users were asked to pick 5 words that they associated with each homepage design (2 design sets done by the Purdue Team and the Orginal design set done by Upswyng Team) from a total of 25 words - about half negative and half positive.
Protocol Goals: Understand our user’s emotional response and the desirability of the homepages.
MOCKUPS
WORDS
RESULTS & INSIGHTS
Results:
Insights:
-
They like having a lot of options on the homepage
-
All users were confused about the Coordinated Entry, Resources, and Crisis Buttons
-
They like simplicity in colors, but had positive reactions to colorful designs
-
Shelter, Health, Food, and Transportation were expressed as the most important
​
Corrections to be made:
-
Move all options to the home/main page
-
Remove the Coordinated Entry button
-
Split Resources button into 3: Hygiene, Health, and Social Services
-
Change wording of Crisis button to "Hotlines"
-
Separate the top 5 most important buttons by color and placement on the homepage
Next Mockup Iteration
USER TESTING 2 @ LTHC
Testing Location: Lafayette Transitional Housing Center (LTHC) Lafayette, IN
Users in Testing Session: 4 individuals, age range from 35-40
Protocol Script Goals: Understand what users think each icon on the homepage would do. See if they understand what actions can be taken on the details page. Gain feedback on additional filter options/features that might be helpful
RESULTS & INSIGHTS
Insights & Results:
-
Required to keep Coordinated Entry due to legal reasons
-
Mentioned staffing opportunities should be added
-
Card expansion to more information was not intuitive
-
Most all users needed & wanted to make the text larger
​
Corrections to be made:
-
Added Coordinated Entry button back on the main page of the application
-
Added Staffing Opportunities button which if clicked broke out into different types of work including: One-Time Jobs, Short Term Positions, Long(er) Term Positions, Train to Work Positions
-
Added arrow buttons on the right-hand side of all cards to indicate more information could be viewed for each card
-
Added text size controls in upper right-hand corner
​
Perspective Point:
most had eyesight so poor they needed glasses but due to their circumstances they were not afforded the chance to have proper prescription glasses for themselves due to lack of money, insurance, or circumstance (i.e. leaving them in the car and car is now impounded)
USER TESTING iN COLORADO
Testing Location: Boulder, Colorado
Users in Testing Session: People experiencing homelessness on the streets of Boulder, Colorado
Protocol Script Goals: Test the current application redesign, from the Purdue Team, on InVision using a script with specific scenarios
Results/Insights: It was very difficult to get people to trust the team that set out to do testing. None of the users wanted any personal details to be recorded, and they did not allow for a voice recording of the interaction. Therefore, testing was difficult and reaped minimal results. The users that the team was able to test the app with all reported that it would be helpful to have direction in regards to the vast number of local resources - validating the core of the application.
USER TESTING 3 @ LTHC
CARD SORTING
Testing Location: Lafayette Transitional Housing Center (LTHC) Lafayette, IN
Users in Testing Session: 3 individuals, age range from 35-50
Testing w/: Card Sorting & Scenario Recall
Protocol Script 1: Card Sorting - Each user received 43 cards with menu items, icon titles, or specific resources in a plastic baggie and was asked to sort them however they best saw fit.
Protocol Goals 1: Understand how they might group certain resources into categories of their own or the ones we had created.
Results/Insights 1: They were able to consistently sort out Transit, Wifi/Tech, and Pets as categories and then put the appropriate recourses under those categories. If we initially separated out all the categories titles, the user was more apt to sort all the corresponding recourses into the correct "titles" than if they had to originally decide if it was a category title at all. Before this testing session, we created a sitemap to understand the flow of the application as it would be. This testing showed us that what we had designed in the sitemap and flow was easily understood and navigable by the users.
Protocol Script 2: Scenario Recall - Each user was asked to recall the last time they needed to use technology to locate a resource or fulfill a need and then walk us as a team through how they might use the application to accomplish that task
​
Protocol Goals 2: Experience natural exploration and navigation of app to understand if intuitive and level of usability.
​
Results/Insights 2:
Text size is mitigated by users through zooming in or rotating - not by the button that we added
Suggested that housing other than shelters should be offered - ex. Affordable Housing Apartments
A better explanation of Coordinated Entry is needed since it will be legally required to be on the application and can be extremely beneficial for those individuals experiencing homelessness
​
SCENARIO RECALL
IMPACTFUL QUOTES
"There’s a lot of people who need help. If you’re on the street, you need help" - user during testing session
"People who don’t know anything about the Greater Lafayette area will need this" - user during testing session
"Just knowing something like this will exist is super reassuring" - user during testing session
FINAL DESIGNS
PRESENTATION & TAKEAWAYS
Stakeholder & Class Presentation
We had the opportunity to present to the Upswyng team in Boulder, Colorado (through skype) while presenting in person to our classmates and professors. We walked through the problem, the research, the user feedback sessions, our iterative design process, and our final deliverables and results.
It was an amazing opportunity to speak about a user group that is so prevalent but so overlooked in the Greater Lafayette area, Boulder, and almost every city in the US, as well as the world.
​​
Takeaways
My favorite part of this project was the human connection at the core of it all. The opportunity to hear the community member's stories, struggles, and successes. In my senior year, I got this chance to see so clearly the people that had been just outside my campus bubble for the past four years.
I realized that you can't live or exist well without a knowledge of those around you. It is in having a deep understanding of the humans behind what you are designing, that one can develop the most usable tool.