SmartDrop: AIoT Assistive Eye Drop Helper
Exploring collaboration experience strategies inside a Fortune 500 manufacturing enterprise
Services
Product Design
Year
2022
Website
6 Months
Website
During these 6 months, I worked on designing an assistive solution that helps older adults and people with limited dexterity use eye drops more easily and confidently. My goal was to design a seamless experience that combines hardware and software, rooted in real user needs and constraints from the healthcare space.
“I can’t tell if the drop went in or not. I just hope for the best.”
This project started with a small, everyday action that turned out to be a big challenge: using eye drops.
For many older adults, applying eye drops isn’t easy. Some struggle with shaky hands or limited vision. Others simply forget when or how many drops to use. Even when they try, they’re often unsure if they’ve done it correctly.
One user told us, “I can’t tell if the drop went in or not. I just hope for the best.”
We realized this wasn’t just a physical task issue—it was emotional too. There was stress, uncertainty, and frustration wrapped into something that should’ve been routine.
Caregivers shared their side too. They often had to remind loved ones, check in constantly, or even physically help with the process. It was tiring, and it added pressure on both sides.
So, here comes to our design question:
How might we design an assistive system that makes applying eye drops easier, more accurate, and less stressful—especially for users with vision issues or limited dexterity?
Design Process
We began by observing. We watched real users try to apply their eye drops—at home, in clinics, and in interviews. We also tested existing market tools, which were often bulky, confusing, or simply ineffective.
We also interviewed older adults, caregivers, and SME experts to understand pain points from all angles.
From there, we sketched out the full journey—from prescription to daily use—and began prototyping both hardware and software. We brainstormed features, voted on priorities, and quickly built a working flow.
Across 5 full design-test-iterate cycles, we refined the experience. We simplified the user interface, improved instructions, added voice support, and reshaped the device for comfort and feedback.
What we learned: Insights from the collaboration experience
Our final system combines a Handheld Device and Mobile App, design to enhance independence and accuracy in user’s medication delivery.
1. Handheld Device
The device gently holds the eyedrop bottle, helps users align it correctly, and provides haptic or visual feedback during use. It even detects whether the drop was dispensed successfully.
2 “This feature seems useful — I just didn’t know it was there.”
Search was among the most frequently used features — and one of the most frustrating. People often entered keywords, waited, then left without clicking anything. In interviews, users described search as a guessing game. Behavioral data showed a high bounce rate within 10–15 seconds. Information was scattered across group chats, individual messages, files, and business systems — but there was no clear map of what lived where, or what could be retrieved. The problem wasn’t just the interface. It was a lack of system-level intelligence.
Design for discoverability, not just access.
I recommended improving the visual prominence of filters and notifications using clearer icons, consistent placement, and supporting text labels. Contextual prompts like “Too many updates? Filter unread messages” can be introduced during high-volume chat activity. Progressive onboarding or lightweight tooltips can reinforce these tools without disrupting workflow.
3 “I opened the chat — and then got lost.”
The message window — the core space for collaboration — frequently left users confused. Interviews and click maps showed low interaction rates with built-in productivity tools like file requests, polls, or quick approvals. Tools were inconsistently labeled and scattered across devices, requiring repeated relearning. Heuristic audits flagged issues with clarity, consistency, and grouping. “I didn’t even know that button was for meetings,” one user said. The space was full — but cognitively empty.
Restructure the chat space around user tasks.
I proposed a task-first layout with three clear zones: messaging, contextual tools, and action shortcuts. High-frequency features like file sharing and meeting invites should be prominent; low-frequency tools grouped under expandable menus. I also emphasized parity between desktop and mobile to reduce learning effort. Better layout isn’t about adding more — it’s about making the right thing show up at the right time.
4 “I want to know what they’re working on — but there’s nothing here.”
The personal profile page was heavily visited — and just as frequently exited. Users expected basic work context: What’s this person working on? Are they online? How can I reach them? But most profiles only offered names and titles. As one team lead put it, “We’re in the same project group, but I honestly don’t know what she does.” In a distributed organization, relationships need more than just contact. They need cues for connection.
Bring work context into contact pages.
I proposed redesigning personal profiles to surface relevant signals like current role, project alignment, availability status, and team relationships. Lightweight integrations with OKR tools or scheduling systems allow quick visibility into shared priorities. Adding fields like “works with X team” or “handles contract approvals” can make collaboration smoother and more intentional — especially across functions.
Zooming out — where do we go from here?
Effective collaboration in a manufacturing enterprise isn’t just about communication. It’s about visibility, alignment, and trust — across time zones, functions, and tools.
Midea Connect is meant to support this system. But when it doesn't reflect how people actually work, it becomes invisible in the wrong way: present, but no longer helpful.
As UX researchers, I led the discovery process, helped define product north stars, and advocated for experience to be treated as a measurable organizational asset. We mapped UX strategy to business outcomes and built a long-term roadmap across functions — shifting collaboration design from a support task to a strategic capability.
Bringing user thinking into digital transformation revealed something deeper: Poor experience isn’t just frustrating — it’s operationally expensive.
As collaboration becomes more dynamic, our challenge isn’t adding features — it’s designing for context. We need tools that adapt to roles, predict intent, and stay out of the way when needed.
In modern work, being seen is just as important as being connected. Sometimes, it’s the missing link between coordination and chaos.