Section 1

Introduction

The opening section of the client book — what’s at stake, who’s involved, what we set out to learn, and why it matters globally.

Section 1.1

Executive summary

The thesis: autonomous delivery scales when the last 60 seconds — the handoff — is treated as a human-centered interaction, not a mechanical task.

60sThe friction windowWhere the robot meets the human.

The last minute of the delivery is where the system breaks — and where this study focuses.

Delivery robots are showing up on more streets every month — and breaking in the same place every time: the last minute.

What worksDelivery robots find addresses and avoid cars and people just fine.
What breaksThe last 60 seconds — when the robot stops and the customer has to figure out what to do.
WhyWe're treating a social moment like a vending machine. People don't know which robot is theirs, where to stand, or if anyone will help.
What this study deliversWhere it breaks for whom — and what to change in the app, the robot, and the building.

Key findings

Four insights, four emotional drivers

Each finding rolls up to one of the four core emotional drivers unpacked in §4.2 — I feel…

Coverage is a mystery.

“Can I use one on my street?” There's no clear way to check upfront — areas, hours, buildings are all opaque.

→ I feel informed

Curiosity turns to friction at the curb.

Walking down to meet the robot is where the novelty wears off — especially when the building has no direct access.

→ I feel in control

The system keeps changing on you.

“App said robot, cyclist showed up.” Last-minute swaps and unclear rules break trust before the handoff.

→ I feel trusting

The missed human moment.

No greeting, no thank-you, no recognition. The handoff feels transactional, not warm.

→ I feel connected

Section 1.2

Project context

Market position, the three signals that make the timing right, and the assumptions our own primary research invalidated.

The business problem

Autonomous food delivery is scaling fast — the last 60 seconds isn’t.

Rapidly growing market

Key stat

$8.9BFood-robotics market by 2033A 6× expansion from today’s pilots.

150%Year-over-year growth (2024)The curve is accelerating, not flattening.

~24%Uber Eats’ U.S. shareLargest platform — what Uber ships shapes the category.

~80%Of every delivery dollar is laborRobots are already cheaper per drop.

The economic case for robots is clear — ~80% of delivery cost is driver labor today. But the last 60 seconds — the handoff between robot and human — is where the system breaks. Failed pickups, redelivery requests, and user abandonment within that final minute prevent the technology from moving beyond pilots into a scalable, everyday service.

Why this research now

Three signals make the timing right

Adoption signal

61% / 98%

Demand exists at scale. Build for retention, not awareness.

On campus, 61% of students order weekly with 98% positive sentiment. The behavior is already there.

Discoverability gap

5 of 5

Adoption breaks before the robot moves. Fix discoverability first.

5 of 5 motivated cultural-probe participants couldn't find the robot option in any mainstream app.

Infrastructure inflection

Infrastructure is solidifying. Design API-aware handoffs now or refactor later.

Elevator APIs and building integration top the 2026 industry watch-list — the spec is still soft.

What decisions depend on this research

Five surfaces this work directly informs

App UXHow the robot option is surfaced, opted into, and tracked from order to pickup.
Physical interactionThe unlock gesture, arrival cues (audio + visual), pickup-spot precision, robot disambiguation.
Building integration roadmapConcierge handshake, locker hand-off, elevator API priorities for high-rise customers (Henry persona).
Restaurant onboarding policyConvert satisfied operators (Real Tacos) and skeptical non-adopters (UMAI) into a coherent supply side.
Brand strategy on the sidewalkConvert visibility — currently neutral, generic, and unconverted — into trial.

What primary research invalidated

Three assumptions the data did not support

“The handoff is the problem.”It is, but the upstream problems (discoverability, app-to-fulfillment mismatch, geofence confusion) start failing trust before the robot ever moves. The handoff is the visible symptom, not the root.
“Sustainability sells.”It does — but only when the basic service works. Climate value disappears the moment a customer thinks the food might arrive late or cold.
“Brand visibility on the sidewalk drives adoption.”Sensory-cues intercepts at GT campus showed initial curiosity does not convert to repeat trial. By the third encounter, the robot is invisible. The conversion mechanism doesn’t exist today.

Strategic frame

A behavioral and functional flow study.

This research treats the handoff as a human-centered interaction rather than a mechanical step in a delivery flow. The deliverable is not a redesign of the robot — it’s a clearer picture of the conditions under which a delivery robot becomes a desirable, repeated, and recommended choice.

Section 1.3

Stakeholder map

Everyone whose decisions, behavior, or presence shapes the experience — classified by proximity (Primary / Secondary / Tertiary) and influence mode (Direct / Indirect).

← Indirect

Direct →

Hover or focus a bubble to preview · click to lock. Direct (right of the line) = contact during the delivery event. Indirect (left) = influences from outside.

Stakeholder tiers

From the customer outward

Five tiers, nineteen stakeholders — from the people at the pickup moment to the ecosystem actors shaping the rollout.

Primary tier (core)
People who experience the service directly.
- Uber (research client)Commissions the research, shapes product decisions.
- Customer (end user)Orders and receives the food.
Inner ring
Closely tied to Primary.
- Customer pickup momentThe behavioral event at the heart of the study.
- Robot operatorTeleoperators in command centers — supervise and intervene.
Secondary tier
Don’t use the service themselves but shape how it works.
- Robot makere.g. Serve Robotics
- App / UX design teamShape the digital interface.
- Maintenance & tech supportField + remote operations.
- Restaurant staffDispatch the order, load the robot.
- Customer supportHandle failed deliveries.
- Campus (facilities mgmt)Where pilot deployments often live.
Tertiary tier
Broader ecosystem actors influencing without participating.
- WorkforceHuman couriers whose role is shifting.
- PedestriansNegotiate sidewalk space with robots.
- Human delivery workersMost directly affected labor.
- Accessibility advocatesEquitable access to public space.
- MediaFrames the broader narrative.
- City / local governmentRegulates deployment via policy + pilot zones.
Indirect (competitive + complementary)
Influence the system from outside the delivery event.
- DoorDashRobot, groceries delivery — competitor.
- GrubhubFood order-delivery app, Amazon Prime agreement — competitor.
- Drone makersSimilar / complementary technology shaping public perception.

Know more ↓

Insight

The customer pickup moment sits at the intersection of multiple stakeholders — Uber Eats, the customer, the robot operator, restaurant staff, and indirectly the pedestrians around them. This is precisely why the handoff is where this research focuses.

Section 1.4

Research statement

Intent · Impact · Method — the formal frame for what this research delivers.

Intent

Identify the functional gaps at the robot-to-human handoff moment in autonomous sidewalk food delivery.

Specifically: friction points — failed pickups, redelivery requests, and user abandonment — that occur within the final 60 seconds of the delivery cycle.

Impact

Reduce operational costsLower the need for support calls and redelivery rates for delivery companies.
Increase order completionDrive successful pickup rates across deployment zones.
Improve first-time-use adoptionThe critical metric that determines whether robotic delivery becomes a scalable global business or remains a niche pilot program.
Make the experience desirableNot merely functional — turn early users into advocates.

Method

Online research and desk research.
Contextual observation in active deployment zones.
Intercept interviews with end customers and restaurant staff.
Cultural probes simulating a real robotic-delivery order.
Focus groups with target audiences.
Service failure analysis across active deployment zones (campus + urban).

Failure points are mapped by frequency and revenue impact; interventions are then generated and prioritized using an impact–effort matrix.

Section 1.5

UN SDGs alignment

Three of the UN's 17 Sustainable Development Goals that the robotic-food-delivery problem space materially intersects with — grounded in project data, not theme.

SDG 9

Industry, Innovation and Infrastructure

Distill infrastructure preconditions into discoverable design recommendations — actionable by a platform operator (Uber Eats), a robotics partner, or a municipality.

Read the evidence

Why this goal

Sidewalk delivery robots are a frontier of urban autonomous systems, and their deployment depends on infrastructure most cities haven’t yet built. The project surfaced concrete infrastructure gaps that are not just technical — they’re structural preconditions for whether the technology scales at all.

What the data showed

Major cities lack a standardized “robo-readiness” index. 30–40% of the physical path in current deployment zones is technically unmappable or inconsistent (temporary construction, uneven curbs).
Indoor reach is gated by elevator API integration. Without it, delivery is limited to the ground floor — the #1 industry requirement for scaling indoor delivery in 2026.
Real-world urban testing shows performance drops up to 25% when robots encounter non-standard obstacles.

Official UN SDG 9

SDG 11

Sustainable Cities and Communities

Surface the social and infrastructural conditions under which robotic delivery actually benefits a city — not just the operator. Recommendations are filtered for whether they expand or contract equitable access.

Read the evidence

Why this goal

Robotic delivery is, before anything else, a city question. Whether it succeeds or fails is a question of urban design, equitable access, and shared sidewalk space — not just delivery economics.

What the data showed

Robotic delivery can reduce urban congestion by 29% vs. traditional delivery vehicles.
Eco-friendly delivery robots reduce road travel by 0.7–1.59 miles per delivery, lowering total urban road load.
Field observations and intercepts at Georgia Tech documented 40 dangerous near-misses between pedestrians and sidewalk robots in a five-day study, plus 60 moderate-risk interactions — the robots were to blame in most cases.
Deployment is currently clustered in high-income, high-density areas with the right infrastructure, creating a “sidewalk divide” that excludes lower-income neighborhoods.

Official UN SDG 11

SDG 13

Climate Action

Build the case that climate performance is not just an externality of robotic delivery — it’s a demand-side feature that, when communicated clearly, increases adoption. Solving the handoff is a climate lever, not just a UX lever.

Read the evidence

Why this goal

Sidewalk delivery robots are the lowest-emission delivery option per package across the spectrum. The project elevates this from a marketing line to a measurable design driver.

What the data showed

Per-delivery emissions are reduced by 67–99.9% vs. gas-vehicle baseline.
Gas vehicles emit ~612.7 g CO₂ per mile vs. ~9.3 g for sidewalk robots.
Robotic delivery cuts CO₂ emissions by 16% at the system level vs. traditional delivery vehicles.
Sustainability is a direct predictor of usage intent — though the effect disappears when food arrives late or cold.

Official UN SDG 13

How these three connect

A single logical chain

SDG 9builds the precondition

SDG 11ensures it serves the city

SDG 13is the payoff

Robotic delivery without infrastructure is a pilot.

Infrastructure without equity is a premium service.

Either failure cancels the climate gain.

Where these appear in the rest of the book

Cover page — three SDG icons in the metadata block, signaling the framing up front.
Section 1.1 Executive summary — SDG framing referenced as the project’s strategic lens.
Section 5.2 Opportunities — every recommendation tagged with the SDG(s) it advances.

Up next

Section 2 · Methodology

How the research was done — approach, sample, timeline, and the four primary methods.