Appendix

Insights

• Low staff training levels / familiarity with the system.
• Zero personalization, ticket-based assembling.
• Minimal/short interaction between the staff and the robot (pickup → assigned robot → close lid).
• Overall satisfied.

Quotes

• “We don’t receive many robot orders.”
• “We just put the food inside when it arrives.”

Insights

• Pricing preference → consistent ordering.
• Exclusive spot pickup driven by the system.
• Pickup flow: push notification → spot → opening lid through app → pick up → close lid.
• Zero social interaction.
• Preferred by those in a rush. Overall satisfied.

Quotes

• “I just wait where the app tells me.”
• “You unlock it with your phone.”

Insights

• Clear platform preference.
• Low emotional engagement.
• Pure functionality focus.
• Overall satisfied.

Quotes

• “I didn’t use Uber Eats, I used Starship.”
• “It’s simple, but nothing special.”

Insights

• Accurate pickup location.
• Delivery format / user is not updated whether it is human or robot prior to start.
• Food condition: no damage.
• Overall satisfied.

Insights

• Large pool of high-frequency delivery users who don’t know robot delivery is available — discoverability inside the apps is the bottleneck.
• For mid-rise without doormen, robots are not a downgrade — the resident already comes down for human couriers.
• Social proof and a clearly visible price advantage are the simplest unlocks for trial.
• Real-time tracking (already loved with human couriers) is table stakes.

Quotes

• “I just don’t know it’s an option. I’d want to know how I get the food.”
• “If the app literally said ‘try a robot delivery, $2 off’, I’d click. Or if my roommates tried it first and posted about it.”

Insights

• Memorable and fun once, but novelty doesn’t translate into a default for routine orders.
• For a low-rise house with setback from the street, hand-off location matters more than the technology — robots stopping at the curb create friction.
• Older customers may want a clearer audio/visual signal when the robot arrives.
• Loss of social niceties (saying thank you) is real — emotional warmth of human couriers is a perceived loss.

Quotes

• “I missed having someone to say thank you to. That part was weird.”

Insights

• High-rise residents lose the front-desk drop-off advantage with robots — a real workflow regression.
• Reliability and consistency are the strongest robot benefits for busy professionals.
• Hand-off in public, on a sidewalk, in front of strangers is socially awkward; voice or visual cues could reduce that.
• Building integrations (concierge handshake, locker hand-off) are a meaningful unlock for dense urban deployment.

Quotes

• “By the third or fourth time, it stopped feeling cool and started feeling like more work than just a person handing me a bag at the lobby.”
• “No tipping.”

Insights

• Fixed pickup points push the hand-off problem from the robot to the customer, especially at night or in bad weather.
• Disambiguating between multiple robots in the same area is a real failure mode — visual or audio identification needed.
• Cultural and social norms (greeting, politeness) make silent hand-offs feel strange for some users.
• The ‘identity / story’ value of robot delivery is real — early adopters share photos — but doesn’t sustain repeat usage on its own.

Quotes

• “Feels walking up to the robot, don’t know if you are supposed to talk or something else.”

Insights

• Robots have to clearly undercut the human option in total price, not just delivery fee.
• Limited cultural exposure to delivery apps means he is less app-fluent and less willing to explore — onboarding clarity matters.
• Robot visibility around campus is high but does NOT translate into trial; users don’t know how to actually order one.
• About the robots: ‘Honestly I think it’s funny, but also kind of weird?’

Quotes

• “And I don’t trust the apps fully, like I see prices in the app are higher than in the actual restaurant.”

Insights

• Lack of explicit customer demand is a real signal — restaurant managers won’t push internal change without that pull.
• Successful adoption needs corporate alignment + on-site training + commercial transparency (commission savings vs. labor cost).
• Restaurant concerns about urban environment (crosswalks, vandalism, vulnerable populations near robots) are real and need a platform answer.
• Office-tower delivery (above-ground hand-off) remains a structural mismatch with sidewalk robots.

Quotes

• “Our customers haven’t asked for it. We hear about other things — gluten-free shells, late-night hours — but no one’s emailed me asking for a robot.”

Insights

• Closed environments (campus, gated, dense urban beltline) are the sweet spot — open city environments have too many failure modes.
• Robots unlock incremental orders during low-staff windows (late night, between rushes) without adding labor.
• ‘Class ran late’ is a campus-specific failure mode — robots’ fixed wait window doesn’t match student schedules and produces refund churn.
• Robot disambiguation in clustered hand-off zones is a real pain point; the right student finding the right robot needs better visual/audio cues.

Insights

• Could not find robot option in apps — abandoned the test.
• Notification suggested robotic delivery, then a cyclist appeared. Expectation gap between app communication and actual fulfillment breaks trust before the hand-off.
• Outside the GT geofence, Starship is unavailable. Mainstream apps (Uber Eats, DoorDash) failed to deliver even within plausible range. Geography is destiny.
• 5 of 5 motivated participants — actively trying — failed. Adoption isn’t a UX problem at the hand-off; it’s a discoverability and availability problem upstream.
• No participant could surface the option through normal app flows. Even keyword search returned ambiguous results.

Quotes

• “Notification suggested a robot; cyclist actually showed up.”
• “Got a cyclist; Starship out of range; Kroger never assigned a robot.”

Insights

• Dominant emotional register is neutrality. Robots are perceived as a tool, not an experience.
• Students see a generic ‘delivery robot’ — not ‘Uber Eats’ or ‘Starship’. Brand identity does not survive the trip from app to sidewalk.
• On the sidewalk, no signal says ‘I am working,’ ‘I am waiting for someone,’ or ‘I belong to this brand’. This silence is the design gap.
• Initial curiosity (‘I took a photo,’ ‘we filmed it for Instagram’) does not convert into repeat trial. By the third encounter the robot is invisible.