Every now and then we encounter someone homeless asking for help, yet we often find ourselves averting eye contact, going as far as avoiding them. Why is that? It's become a stereotype to think that interaction may be unsafe or that donations are misused.
As the year comes to a close, we wanted push ourselves as designers and use technology in a more impactful way. Our team designed Street Samaritan in hopes to bring more help and awareness to the impoverished residents of Los Angeles.
Street Samaritan is a community platform app that allows users to directly and transparently donate to those experiencing poverty or homelessness, along with peace of mind.
It's not that people don't want to help those asking on the streets. But we're often dissuaded by stronger concerns, from our own safety to possible misuse of said donations. Let's target to alleviate those reasons.
As the year came to a close, we wanted push ourselves as designers and use technology in a more impactful way. Our team designed "Street Samaritan"— a community platform concept app that allows users to directly and transparently donate to those experiencing poverty or homelessness, with ease and peace of mind.
My Roles:
UX Researcher
Visual Designer
Interaction Designer
Prototyping and testing
The Tools:
Trello, Miro, Figma, Webflow
The Duration:
3 Weeks
The Members:
Martin Carpio, UX/UI Designer
Erin Kim, UX/UI Designer
Joe Lee, UX/UI Designer
Caroline Pham, UX/UI Designer
The Methods:
🎙 User Interviews
📊 Surveys
🕸 Affinity Diagram
🧟♀️ User Personas
🌈 Storyboard
🤩 Competitor and Feature Analysis
🟨 Brainstorming
🤔 Feature Prioritization Matrix
🌊 User Flows and Journey Mapping
🦴 Wireframes
🤓 Usability Testing
🤖 Interactive Prototype
💻 Website M.V.P.
• What are your views of the homeless community?
• What deters you from donating to homeless people and why?
• How often are you on social media? What social media do you use?
• What attracts you when using social media and why?
Our team conducted a total of 8 interviews (2 per team member) through a network of our own peers. For qualitative data, our goal should be to gather the general opinion about donating and how much of an impact social media can have on their donating habits.
Through Google Surveys, with 19 participants and the majority being female, all between 15-24 years old, and making an average income between $30k-$60k a year. This gives insight in which factors can play a role in creating our personas.
Dividing the findings into an affinity diagram, the highlights of data showed that most people were concerned about the transparency of donation, also that donating habits ranged from giving change on the street to more secure donations done through trustworthy campaigns or organizations they found on social media sites.
The homelessness crisis rates steadily grow each year. In Los Angeles alone, more than 60,000 people are living on the streets. At 50 city blocks, the city comprises of the largest unhoused containment in the United States: Skid Row.
Many find conversations about the homelessness issue to be an uncomfortable subject. Our goal is to create a comfortable space for this diverse community to connect.
The idea was to come up with a user-community platform to introduce individuals in need to (whether it be need for clothes, money, a suit, etc.) in hopes to increase visibility and support. Our initial assumption is that users will feel more inclined to help and donate if these individuals were introduced on a more personal basis.
Our key research objectives were the following:
• Understand people’s opinions about homelessness and their inclination to donate to homeless people
• Determine the impact that visual aids and social media have on people’s willingness to donate
• What are your views of the homeless community?
• What deters you from donating to homeless people and why?
• How often are you on social media? What social media do you use?
• What attracts you when using social media and why?
Our research consisted of:
• 8 qualitative, peer-sourced interviews (2 per member)
• A Google Forms survey, completed by 19 responses
Our team conducted a total of 8 interviews (2 per team member) through a network of our own peers. For qualitative data, our goal should be to gather the general opinion about donating and how much of an impact social media can have on their donating habits.
Through Google Surveys, with 19 participants and the majority being female, all between 15-24 years old, and making an average income between $30k-$60k a year. This gives insight in which factors can play a role in creating our personas.
Dividing the findings into an affinity diagram, the highlights of data showed that most people were concerned about the transparency of donation, also that donating habits ranged from giving change on the street to more secure donations done through trustworthy campaigns or organizations they found on social media sites.
Through our Affinity Diagram, we observed a key insight that reinforced our hypothesis,
that users are discouraged by fear and safety factors:
• Most want to donate but are more concerned about the transparency of their donations.
• Donating habits ranged from either giving smaller to no change on the street, to bigger secure donations through campaigns or organizations they find boosted by social media.
"We're unaware of the marginalized. I hope people see it as a public health concern rather than a result of who they are." -L.G.
"You really cannot judge someone by their worst day." -R.G.
"Homelessness is increasingly becoming a huge problem due to a widening gap between wages and cost of living." -B.D.
We realize that this community has two sides open for users, so we made a decision to create 2 user personas:
1. (Sharon) Target user that utilizes the app to make safe and smart donations conveniently
2. (Henry) End user (referred to as "members") receiving benefits
We felt that, to create empathy throughout our work, it is important to be considerate of both users.
Sharon, is our donor, who commutes to DTLA where she works. She regularly passes by the homeless areas and while she does want to help, she is frustrated because she often has no time, and feels unsafe on the chances she does get.
Henry, our donatee who has been experiencing homelessness after losing his job. Henry is in need of resources to support his daily life. He is willing to accept any generosity that reaches out to him but is frustrated because stigma against the homeless community makes it difficult for people to help.
How might we provide kind people like Sharon with a convenient solution to help the impoverished, while maintaining the safety and comfort she needs?
There's a handful amount of organizations out there that are dedicating themselves in fighting the homelessness crisis. We found Handup's mission closest to align with our goals, and Gofundme similar to how we view our product's functionalities.
Using the sticky notes on Miro, we had a productive brainstorming session about the different features we each see our product heading. Shortly after, we got together and sorted our notes out in a matrix.
It was at this point we realized designing different UIs for each user will divide our time and resources, having to split work into two. We decided to focus on designing for Sharon's user UI, but create plans around another UI.
Our aim was to create the app that made donating easy for the user. My goal was to make the product usage convenient and intuitive. These user and task flows are aimed to catch any interaction design issues, and to help create a more seamless interaction by focusing on fewer, more effective, steps.
Complex task flow to show all basic steps in the app.
Flow chart to map out user's journey, and also serve as our usability test's task list.
After receiving the "ok" from the wireframing team, I made sure to highlight and polish consistent actions and interactions between the lo-fi frames, and prepare it for a usability test.
After designing for mobile, we wanted to make a minimum viable web product to trial for our apps responsiveness.
• Empathy is at the very core of this project (and every project to come). It was frustrating to slowly realize we could only design for one side of the UI (user-side), and only draw assumptions for the other (member-side), due to time and feasibility constraints. These were shortcomings we had to realize and learn from.
• Since the team mostly operated on very similar roles, clashes in decisions were expected. Transparency was within our best and common interests, so explaining our choices, driving them from research, and soliciting honest feedback with one another were necessary.
• Because we thought to tackle a complex societal problem with no one solution, we overextended on necessary ideas, only to be met with constraints and limitations. Only when we drove our design decisions with user research did we find what features and functions our product really needed.
• Expand on webpage version for desktop users and create a landing page.
• Create a second member-side interface for the "members/social workers/orgs", creating UI for the other side of users.
• Explore the donation options more, such as defining the "goodiebox" and implement a "giftcard" feature for better usability and more secure transactions.
• More "campaign" creation options to be inclusive for more needs (new laptop, professional clothes, phone charger, etc.).
What started out as a thoughtful idea by wide-eyed designers, to make a simple donation app, entailed more beyond our expectations. We discovered feasibility constraints and limitations, stumbled on vast options, and learned to find compromise between usability and value. It was a challenging, yet humbling, process that I'm grateful to have grown from.
Last but not least, thank you for reading! 🤗
Redesigned the manual data request process for internal auditors and external clients, reducing turnaround time by up to 40% through improved templates, clearer instructions, and a streamlined submission interface.
20 min. read
Disclaimer:
This case study is based on real research, insights, and design work; however, due to confidentiality agreements, certain details have been modified, generalized, or omitted to comply with non-disclosure agreements (NDAs). Any proprietary information, including specific company names, internal data, and strategic details, has been altered or anonymized while preserving the integrity of the design process and key learnings. The purpose of this case study is to illustrate the UX challenges, methodologies, and outcomes without disclosing sensitive or proprietary information.
UX Researcher: Martin Carpio
Client: Customer Services, Training
Target Users: Pricing Analysts (customers), Auditors (internal employees)
Challenge: Customers need to easily and accurately update their data to ensure a timely and error-free distribution to the marketplace. Internal Auditors need customers to input data accurately and consistently to process and resolve customer data requests efficiently.
Activities: User Interviews/Contextual Inquiry, Service Blueprint Mapping, Content Audit
As the company scaled its data distribution services, the Customer Services team noticed a persistent issue: customers often submit error-prone updates to their data stored in our systems. The team found that customers often use their own custom templates to submit file updates. Admittedly, the ideal process to submit data to our system was a bit convoluted, requiring users to switch between multiple (counting 6 or more) applications, rely on unintuitive templates, and navigate a maze of outdated instructions.
Through our CSAT survey and general user feedback, it became evident that the existing process needed considerable improvement in customer effort, efficiency, and accuracy. Here are some of the preliminary findings:
- Processing Delays: Each incorrect or unclear submission required back-and-forth clarifications between customers and our teams, stretching turnaround times by an average of up to 24 hours.
- Low Customer Satisfaction: Friction in the data submission process led to poor CSAT scores, signaling an urgent need for improvement.
“A customer commented that it’s too hard to do business with us.”
Introduction
The Customer Services team approached UX with the intent to redesign their data distribution process. They wanted to know how we can improve efficiency, effort, and accuracy in the process customers use to instruct manual distribution data input changes to our filing systems.
Key insights and recommendation - Service Blueprint Mapping
1. Users go through a fragmented experience.
2. The standardized templates are difficult to discover.
3. Due to the underutilization of templates, manually submitted data can be inaccurate and inconsistent, which causes delays in processing.
Primary recommendation: Enhance template usability and encourage adoption by providing clearer guidance, such as incorporating dropdown options, comprehensive examples, and standardizing input formats.
Key insights and recommendations - Template Content Audit
1. Template users experience ambiguity and confusion when submitting data because the templates have redundant fields and include unclear instructions. So we should streamline templates by removing redundant fields and rewriting instructions in clear, concise language.
2. Users will struggle with accuracy and efficiency when completing forms because manual inputs are required for common fields, instead of using dropdowns or other guided inputs. So we should implement dropdown menus and pre-populated fields to minimize manual entry and reduce errors.
3. Users will have difficulty finding key information because they tend to be buried below the fold, and the visual cues to find them are inconsistent. So we should improve visual hierarchy by placing essential instructions and fields in prominent locations, ensuring consistent design elements guide users through the process.
1. Creating a research plan
Before redesigning the request process, we needed to understand the experience of both customers submitting requests and internal employees processing them. We knew the current process was inefficient, but we needed to pinpoint exactly where and why breakdowns occurred.
To guide our research, we set three key objectives:
a. Identify user challenges when submitting data requests, including how they navigate the current system and what causes delays or rework.
b. Assess how customers find and use templates, training materials, and other support resources.
c. Discover opportunities to streamline the workflow by understanding how users currently interact with the system.
These objectives informed the research questions for our user interviews. Over the next two weeks, I recruited 5 customers and 5 internal employees, and conducted interviews simultaneously over the span of 2-3 weeks.
Stakeholder involvement was critical from the beginning. To ensure alignment, I invited key stakeholders to observe the interviews, allowing them to hear user challenges firsthand and build empathy for the things we were uncovering.
2. Mapping the experience
Following the interviews, I facilitated a Service Blueprinting Workshop with stakeholders, where we mapped out what we learned about each step of the process—from initial request submission to final fulfillment. This exercise helped visualize not just the customer’s journey but also the “front-stage” (customer-facing) and “back-stage” (internal) interactions that made up the entire experience. It truly helped having the stakeholders participate in both the interview sessions and this workshop.
From this mapping exercise, three major points emerged:
a. Users go through a fragmented experience.
Customers tend to experience a fragmented workflow because they must switch between 4-6 applications just to submit a single request. This constant context switching creates frustration and increases the likelihood of mistakes or delays.
Recommendation: Streamline the workflow by integrating key tools and touchpoints into a single, cohesive experience.
b. The standardized templates are difficult to discover.
Templates are a crucial part of the submission process, yet many users either struggle to find them or don’t even know they exist. The issue? Templates are stored in a separate application way outside the main workflow, making them easy to overlook. As a result, customers would often bypass them, use outdated versions, or create their own custom templates, leading to errors and inconsistencies.
Recommendation: Improve the discoverability of templates by embedding them directly within the request flow and proactively communicating updates.
c. Due to the underutilization of templates, manually submitted data can be inaccurate and inconsistent, which causes delays in processing.
Because many customers aren’t using the correct templates, internal employees spend significant time clarifying errors and requesting missing information. This results in extended feedback loops that can delay the processing by up to 24 hours.
Recommendation: Audit content in templates and improve guidance by refining instructions, reducing redundant fields, and incorporating dropdowns for common inputs. Since the Customer Services team wanted to explore a solution with a lighter level of effort, this was the priority to pursue.
Using these insights, we created a current-state service blueprint that highlighted inefficiencies. To align everyone on a future-state vision, I facilitated a Future-State Service Blueprint Workshop the following week. In this session, we designed an ideal customer journey that focused on seamless template access, reduced back-and-forth communication, and a more efficient fulfillment process.
Other recommendations that emerged include:
a. Moderate Lift Recommendation: Enhance the discoverability of templates by linking and surfacing them directly into the request workflow, requiring a change in UI.
b. Heavy Lift Recommendation: Migrate the entire process into a centralized, wizard-style form that consolidates submission steps and reduces tool-switching.
3. Auditing the template
While some of these recommendations required longer-term solutions, the Customer Services team was eager to tackle quick wins first. To kickstart improvements, I conducted a comprehensive content audit of the templates, evaluating them for clarity, usability, and overall functionality.
The content audit revealed several issues that directly contributed to processing delays:
a. Template users experience ambiguity and confusion when submitting data because the templates have redundant fields and include unclear instructions.
Customers frequently made errors because certain fields were unclear, repetitive, or unnecessary. This increased back-and-forth communication and delayed processing.
Recommendation: Streamline templates by removing redundant fields and rewriting instructions in clear, concise language.
b. Users will struggle with accuracy and efficiency when completing forms because manual inputs are required for common fields, instead of using dropdowns or other guided inputs.
Many templates required users to manually enter common data points, rather than selecting from predefined options. This led to frequent formatting mistakes and inconsistencies.
Recommendation: Implement dropdown menus and pre-populated fields to minimize manual entry and reduce errors.
c. Users will have difficulty finding key information because they tend to be buried below the fold, and the visual cues to find them are inconsistent.
Users often overlooked critical instructions and fields because they were buried within the document, below the fold, or lacked proper visual cues.
Recommendation: Improve visual hierarchy by placing essential instructions and fields in prominent locations, ensuring consistent design elements guide users through the process.
4. Redesigning the UI
With the template audit complete and minor improvements already in progress, we turned our focus to solving the bigger usability challenges. To do this, I collaborated with a UI/UX designer to create mockups and prototypes that visualized our proposed solutions.
This phase focused on two key redesigns:
a. Medium Lift Recommendation: Improving Template Discoverability
We designed a revamped support page that prominently featured a direct link to the templates page, ensuring users could easily access the correct forms.
b. Heavy Lift Recommendation: Creating a Centralized, Wizard-Style Form
Instead of requiring users to download, fill out, and re-upload templates, we prototyped a guided, step-by-step submission form that dynamically adapted based on the type of data being submitted. This approach would:
- Eliminate the need for external templates by integrating fields directly into the submission process.
- Reduce errors and inconsistencies by standardizing input formats.
- Accelerate processing times by ensuring that all required fields were filled out correctly from the start.
Although the team has yet to implement these changes, we hypothesize that introducing a guided, intuitive submission experience will reduce processing time by up to 40%. This improvement is expected to streamline the request workflow, minimize input errors, and enable auditors to efficiently process data changes with greater accuracy and less back-and-forth.
