SECOND LIFE TECHNOLOGY
</second life technology>
SECOND LIFE TECHNOLOGY
Conversational chatbot UX for a student e-waste service.
3
service flows automated
text + voice
accessible input modes
solo
entire chatbot integration
SPECIFICATIONS
| ROLE | UX DESIGNER |
|---|---|
| YEAR | 2024 |
| TYPE | CHATBOT UX |
| STATUS | archive |
| STACK | voiceflow · airtable · conversational ux · user research · accessibility · figma |
| LINKS | — |
| DELIVERABLES | 3 artifacts · research → hi-fi |
| TOOLS | voiceflow · airtable · conversational ux · user research · accessibility · figma |
“A student sustainability service needed to handle appointment booking, marketplace device recommendations, and e-waste questions without staffing every enquiry.”
A conversational AI chatbot for a student sustainability service.I built the entire chatbot integration - appointment booking, marketplace recommendations, and e-waste FAQ - using Voiceflow with Airtable APIs and accessibility-first design.
README.TXT — SECOND LIFE TECHNOLOGY (3 KB)[full readme →]
== WHAT IS THIS ==
A conversational AI chatbot for a student sustainability service. I built the entire chatbot integration - appointment booking, marketplace recommendations, and e-waste FAQ - using Voiceflow with Airtable APIs and accessibility-first design.
== </the problem> ==
A student sustainability service needed to handle appointment booking, marketplace device recommendations, and e-waste questions without staffing every enquiry. The starting point — a static, button-heavy decision tree — felt rigid and put barriers in front of exactly the users the service most needed to reach.
role & context
I built the entire chatbot integration solo — conversation design, Voiceflow build, Airtable API wiring, and accessibility — within a broader student sustainability service project.
== </my approach> ==
I built the entire chatbot integration solo: conversation design, the Voiceflow build, Airtable API wiring for real-time scheduling and device filtering, and accessibility from the ground up. The conversation evolved from a decision tree into a dynamic assistant guided by Grice's cooperative principles, refined through user testing.
== </the story> ==
I built the entire chatbot integration for Second Life Technology from the ground up. The system handles appointment booking, marketplace device recommendations, and e-waste education. Evolved from a static button-heavy decision tree into a dynamic assistant using Grice's cooperative principles. Accessibility was foundational - text and voice input, text-to-speech, keyboard navigation, and screen reader support.
== </architecture> ==
Built on Voiceflow with Airtable API integrations for real-time scheduling and device filtering. Intent-based routing with fallback handlers. Complementary wireframes showed chatbot integration within the broader service ecosystem.
== </design decisions> ==
From decision tree to conversation
Rebuilt the static button flow as a dynamic assistant using Grice's cooperative principles — intent-based routing with fallback handlers instead of rigid menus.
Cutting device details from booking
User testing revealed the device-details step was a major barrier. Scrapping it was the single biggest win of the project.
Pre-appointment placement
Positioning the chatbot before the appointment, not after, lowered the barrier to accessing the service at all.
Hybrid input model
Text plus buttons, driven by accessibility research — users choose whichever mode works for them.
Accessibility as foundation
Text and voice input, text-to-speech, keyboard navigation, and screen reader support designed in from day one, not retrofitted.
Live data behind the conversation
Airtable APIs power real-time appointment scheduling and device filtering, so the assistant answers from current data instead of a script.
== </key decisions> ==
DECISION 01
Scrapping device details from the booking flow was the biggest win - user testing revealed it was a major barrier. Placing the chatbot pre-appointment lowered barriers to service access. Hybrid input model (text + buttons) came from accessibility research.
== </what i learned> ==
The biggest UX win was subtraction — removing the device-details step from booking did more than any feature I added.
Where a chatbot sits in the service journey matters as much as what it says. Pre-appointment placement changed who could access the service.
Hybrid input isn't a compromise — accessibility research pointed to text plus buttons as the design that serves everyone.
voiceflow · airtable · conversational ux · user research · accessibility · figma
archive