LetterKey
Social Impact
Accessibility Design
2025
Lead Product Designer
Saves Fei Yue 25% Staff Time Thorough AI Simplified Letters
Project Info
When I first joined the project
LetterKey began as part of the 2025 Build for Good (BFG) Community Hackathon, co-organised by Open Government Products and the People’s Association, where my team set out to address the struggles seniors face when interpreting government letters. Following the hackathon, the project was selected for the Accelerator Programme and awarded S$20,000 in funding to further develop and scale the solution.
As the lead product designer, I drove the end-to-end design process from uncovering user pain points to delivering high-fidelity prototypes. I designed and conducted UX research, structured the product’s information architecture, designed intuitive user flows, and collaborated closely with developers.
Challenge
Reality on the ground
We first began by immersing ourselves at Senja's Fei Yue Community Centre. Shadowing staff and conducting 4 user interviews with them. Through our experience volunteering there, we observed 2 recurring problems with the existing workflow that eventually led us to our product vision.
⚠️ Problem
For seniors, government letters were long, technical, and filled with unfamiliar terms
Implication
Many had to rely on staff for help, with support requests lasting between 45 minutes to 2 hours each session
⚠️ Problem
Staff were stretched thin as there are only five full timers balancing these sessions with their other daily responsibilities
Implication
The result was a bottleneck that leaving some seniors falling through the cracks and staff overwhelmed
Process
Designing the Solution
To address this, our team envisioned a physical kiosk solution that helps seniors scan, translate, and understand complex letters with ease.
I came up with a proof of concept, designing the interface with senior's mobility and limited tech literacy in mind and approached the design with two guiding principles: simplify the experience and offer ample touch points for accessibility.
This translated into a three-step flow that seniors could follow independently:
Select your preferred language
Take a photo of your letter
Read or listen to your simplified letter
Testing with Seniors
To stress test our hypothesis and assess its' usability, I designed a UX research proposal with the objective of challenging and validating our assumptions and hypothesis, guiding 4 seniors participants through the end to end flow of our proof of concept.
The Usability Testing results were expected, yet still surprising. I had not anticipated senior's tech literacy to be lower than expected. Below are the results.
✅ What Worked Well
All participants played the audio feature, showing its effectiveness
Validates Hypothesis 2All participants expressed positive reception of the kiosk's core idea
Validates Hypothesis 43 out of 4 seniors rated the product 5/5 for ease of use
Validates Hypothesis 4❌ What needs to be Improved
3 out of 4 seniors struggled with terms like “upload”
How might we simplify technical terms so seniors understand what action is required?2 out of 4 seniors found some icons unintuitive
How might we clarify icons or add supporting cues to reduce guesswork when navigating the interface?1 participant asked to store summaries on their phone
How might we enable seniors to save or share summaries so they can revisit important information
💡 Insights
3 out of 4 seniors prefer In-person assistance for help
Validates Hypothesis 32 out of 4 seniors expressed difficulty interpreting official and financial documents
Validates Hypothesis 1
Finale
Solution
Following the Usability Testing, we took the insights and continued to iterate on the UI, directly addressing the challenges raised during the session, concluding the first phase of the project.
✏️ Improvements (Earlier Findings)
Minimised tech jargons simplifies UX copy
How might we simplify technical terms so seniors understand what action is required?UX Copy added to clarify button functions
How might we clarify icons or add supporting cues to reduce guesswork when navigating the interface?Send to WhatsApp feature for summaries
How might we enable seniors to save or share summaries so they can revisit important information
✏️ Other Improvements
Physical markers for document placement
Markers guide seniors with limited hand mobility to align documents easily, reducing effort and scanning errors.Reduced user flow complexity
How might we clarify icons or add supporting cues to reduce guesswork when navigating the interface?Lyrics style highlight feature during playback
How might we enable seniors to save or share summaries so they can revisit important information
Impact
To conclude, by simplifying letters into clear summaries and storable audio playback, we created an alternative to the 45–120 minutes of staff assistance usually required for each senior. With just one scan, seniors could independently understand their documents within minutes.
Reducing dependency on staff and freeing up human resources
Empowering seniors with the confidence to manage personal documents
Reflect
What's Next?
Being selected to further develop our solution gave us access to funding and new partnerships. Our team is now in talks and have partnered with organisations to scale LetterKey and planned to conduct further user interviews and usability testings in the coming weeks to validate at scale and fine tune the product.
Design for Good
Nothing felt more “end-to-end” than this experience. From heading down on the ground to observe operations, to validating our hypotheses with user interviews, to iterating prototypes based on usability testing.
Designing for seniors was a challenge tougher than i expected. Unlike other age groups, user research and interviews were especially challenging due to minor language barriers as majority spoke in proper mandarin and dialects. Also, older seniors
Being on the ground and seeing how the tool empowered seniors to take control of their own information reaffirmed my passion for designing meaningful solutions. It demonstrated that design is most impactful when it addresses real human needs.
