
How We Ran UX Research with Malaysia’s Underserved Communities
When designing digital solutions for underserved communities who are often overlooked, research can’t be business as usual. It has to be human-first, humble, and grounded in reality.
We partnered with Friendsuretech—a fintech innovator—on their mission to design myKawan, a financial literacy tool for Malaysia’s B40 communities. Together, we aimed to uncover the real-world challenges these communities face around money, trust, and technology so that the final product could truly serve their needs.
And while empathy drove the approach, the business case is just as compelling. Inclusive financial tools don’t just build social equity; they build market opportunity. A study of Chinese banks, for instance, found that when fintech solutions focused on underserved groups, it led to stronger lending growth and financial resilience, especially in rural areas. Reaching underserved communities isn’t a detour from profitability; it can be a driver of it.
This wasn’t just about testing a prototype. It was about building understanding from the ground up. And that meant doing research a little bit differently.
In this Article
Who Are the Underserved? Giving Voice to Malaysia’s Low-Income Communities
Before we could design a meaningful research approach, we first had to be clear about who we were designing for. Conversations about inclusion often start with good intentions, but without a shared understanding of who is being left out, it’s easy to miss the mark.
In this project, the focus was on Malaysia’s B40 group: the bottom 40% of households by income. This group often faces layered, systemic barriers that go far beyond financial strain. Limited digital access, lower financial literacy, geographic isolation, and a lack of tailored financial services all contribute to their exclusion from the wider economy. As such, this research effort is specifically aimed at reaching these underserved communities often left out of digital product development.
In short, many in the B40 community are navigating life without the tools, support, or opportunities that others might take for granted. They are also frequently overlooked in traditional product development, not due to a lack of relevance, but because they’re harder to reach through conventional UX research methods.
What made this project stand out was Friendsuretech’s early commitment to going beyond assumptions in building a financial literacy tool that actually works for the underserved. It meant listening to real stories, real behaviors, and real constraints.
Together, we tailored a research approach that met communities where they are; physically, emotionally, and digitally.
This commitment aligns with a bigger picture too. Financial inclusion isn’t just a feature but a foundation for social equity. The United Nations links it to eight of its Sustainable Development Goals, and Bank Negara Malaysia has highlighted inclusive finance as key to national progress.
Together with Friendsuretech, we aimed to translate these ideals into action by starting with something simple but powerful: listening.
Breaking Barriers: Adapting Our UX Research Operations
From the beginning, we knew reaching Malaysia’s B40 community wouldn’t be as simple as sending out a few screening surveys and booking slots in our UX lab. Many of the tools and assumptions we typically rely on just didn’t apply here.
In our journey to understand the B40 group, we had to rethink every step—from outreach to participation—to truly hear their voices. Here’s how we turned the main obstacles in this journey into opportunities for meaningful research.
💡 Challenge 1: Limited Access to the Low-Income Communities
Our first hurdle became clear almost immediately: we couldn’t reach participants through any of the conventional recruitment strategies. Recruitment channels like social media and user panels just weren’t reliable enough to get the demographic we’re looking for.
These communities don’t typically show up in databases or online records, and financial vulnerability isn’t something most people openly share. Add to that a healthy skepticism toward unknown outreach, and our efforts often felt like spam—or worse, a scam. Malaysia saw nearly 3 million scam calls in 2024 alone, an increase of 82.8% from the year before. Without a strong connection, there was no reason for them to engage.
🚀 What We Did: Partnering with People They Already Trusted
So, we turned to trusted community leaders, NGOs, and local organizations that already had strong relationships with these communities. These partners became our bridge to the people we hoped to reach.
As they had already built relationships through previous support or outreach, their involvement gave our research credibility. Instead of showing up as strangers asking personal questions, we were introduced through familiar, trusted faces. That shift, from cold outreach to community-based connection, made all the difference.
💡 Challenge 2: Earning Trust and Overcoming Hesitation
With the community leaders on board, they helped spread the word about our research to those who might be a good fit. But even when people agreed to participate, financial topics brought hesitation.
For many, this was the first time they were asked to share their personal money habits and struggles with outsiders. A few even wondered if they had anything valuable to contribute at all. This reflected something deeper: a learned invisibility, a sense that their voices had rarely been heard, let alone valued, in product development.
🚀 What We Did: Slowing Down to Create Comfort, Bit by Bit
So instead of overwhelming participants with too much upfront, we took things slow. We first met with community leaders to explain the research clearly and openly. They then passed that understanding on to their communities, helping build awareness and trust in small, steady steps.
By the time participants showed up for the actual sessions, they already had a sense of what we were doing and why. This gradual introduction gave them space to warm up to the idea, and by the time we explored more personal topics, they were ready to share more openly. This approach showed them that their everyday experiences did matter and were exactly what we needed to hear.
Our team with one of the community members helping drive recruitment
💡 Challenge 3: Logistical Constraints
Just as things were starting to click, we ran into another issue: logistics. Even when participants were open to joining, the logistics of showing up often became the dealbreaker. Traveling to our research facility meant navigating unfamiliar routes, figuring out transportation, dealing with parking, and then still needing to make time in a day already packed with responsibilities.
For those just getting to know us through community introductions, trust was still in its early stages. Without a strong relationship yet in place, the effort required to attend—on top of everything else—felt like too much, even for those who were genuinely interested.
We were especially eager to include two key groups who both had their own obstacles in joining our research:
- University students, juggling lectures, assignments, and part-time jobs with little breathing room.
- Housewives, whose days were fully booked with childcare, cooking, and managing the household.
🚀 What We Did: Bringing Research to Them
So instead of asking them to come to us, we went to them. We held sessions at community centers in their neighborhoods and offered flexible time slots, including weekends and evenings.
By making the process as easy and respectful as possible, we removed the pressure and made space for their voices to be heard, on their terms.
Here’s a glimpse into these sessions, where conversations unfolded in trusted spaces.
Focus group discussion with students in a local temple
One on one interview held in a neighborhood association room
Focus group discussion with housewives held in neighborhood association rooms
Key Takeaways: How to Do It Right
Working with underserved communities isn’t just about changing research methods, it’s about shifting mindsets and rethinking the research operations side from the ground up. Unlike typical UX studies, where participants sign up through online platforms, visit a UX lab and be compensated then and there, our work with Malaysia’s B40 communities demanded a more flexible, relationship-first approach.
Here are some of the lessons that stuck with us, and what others might find helpful if they’re hoping to do more inclusive, community-grounded research:
- Community-Driven Recruitment: Rather than sourcing participants through digital channels or databases, we turned to community leaders and local organizations. These partnerships were essential in reaching people who are often left out of the research conversation.
🔹 Takeaway: If the community doesn’t know you, start with someone they already trust. Local partnerships aren’t just helpful, they’re foundational.
- Relationship First, Research Second: We didn’t jump straight into focus groups and surveys. Instead, we spent time getting to know people through informal chats, meet-and-greets, and simply being present.
🔹 Takeaway: Big words and technical terms can make people feel like they’re not “qualified” to participate. Use plain language. Be patient. And remember: most people don’t think their everyday experience is valuable, until you show them otherwise.
- On-Site Flexibility: A formal research lab wasn’t the right fit for this project. We went where participants felt most at ease: community halls, temples, neighborhood association rooms. We adapted to their schedules too, conducting sessions on evenings and weekends to avoid disrupting their daily lives.
🔹 Takeaway(s):
-
-
- Whether it’s a neighborhood temple or a quiet community center, conducting research in a space that feels safe and known makes all the difference. It helps participants open up and reminds us that the lab doesn’t always have to look like one.
- Housewives, students, and working individuals all juggle full plates. Offering evening or weekend sessions—or just showing you’re willing to work around their schedules—signals that you value their time and insight.
-
- Planning for the Unexpected: From spotty Wi-Fi to noisy environments, we learned to plan for (and roll with) the unexpected. Our setup had to be portable, reliable, and ready to adjust on the fly without losing focus on the people in front of us.
🔹 Takeaway: Having portable Wi-Fi, alternative recording methods and the ability to adapt on-site means you can stay focused on the person in front of you, not the setup. It also signals professionalism without rigidity.
Conclusion: Human-Centered Research in Action
This research journey went far beyond gathering data. It was about listening, adapting, and building meaningful connections with underserved communities whose voices often go unheard.
This wasn’t just another product design sprint but a shared mission to build a financial literacy platform that could genuinely support Malaysia’s low-income communities.
By embedding empathy into every step of the process, Friendsuretech took meaningful strides toward financial inclusion, supported by our research expertise.
This experience reminded us that conducting UX research with low-income communities requires more than just the right methods: it demands empathy, flexibility, and a commitment to inclusion to design a financial literacy platform that truly resonates with its audience.
In the end, the impact of this research can’t be measured by findings alone. It’s found in the conversations that were had, the trust that was built, and the steps taken toward designing tools that reflect and respect the people they’re meant to serve. We hope this case study contributes to more inclusive research in Malaysia and beyond.
About the Author:

Wai Yee is a UX Researcher with four years of experience driving research initiatives in the fintech and insurance sectors, shaping digital strategies and product experiences. Outside of work, you’ll likely find her chasing hiking trails or perfecting her cup of coffee.