1) Executive summary

Goal: to make it effortless and trustworthy to

(a) find a space in <2 minutes and

(b) monetize an idle spot safely.

North-star outcome:

“time-to-park” ≤ 8 minutes from app open to parked, ≥ 70% of first-week users complete one booking or listing.

MVP pillars

Reliable discovery (live availability + clear fit: size, price, distance, safety).

Frictionless checkout (one-tap pay, extensions, refunds).

Lister confidence (easy calendar, pricing guardrails, photo-based verification).

Trust & safety (ID/KYC, license plate, dispute flows, guarantees).

2) Problem framing & market scan

• Core problems for drivers

High cognitive load while moving: comparing distance, price, safety, and fit (SUV vs compact).

Uncertainty: “Will the spot actually be free?” “Is it safe?”

Last-mile stress: navigation ends at the block, not the entrance; unclear rules, fines.

• Core problems for spot owners

Setup friction: photos, rules, calendar, pricing are tedious.

No-shows / misuse risk, payment timing.

• Competitor snapshot (for positioning)

SpotHero/JustPark/ParkMobile: strong inventory & payments, limited peer-to-peer trust signals and granular fit (size/height) in many markets.

AirGarage: business lots, less peer-to-peer.

Opportunity: peer-to-peer liquidity + verifiable reliability (photo time-stamps, plate recognition, guarantees) + fast extensions.

3) Target users & proto-personas

A. Commuter Carla (Driver)

Needs predictable, affordable weekday parking near office/transit.

KPI: <2min to choose a spot, <5% booking failure.

B. Event Eli (Driver)

Evening/weekend demand spikes, willing to pay more for guaranteed proximity and extended hours.

KPI: minimal deviations (wrong entrance/closed gates).

C. Host Hakeem (Resident lister)

Spare driveway/space, wants simple set-and-forget listing and payout.

KPI: occupancy rate, zero disputes.

D. Manager Maya (Business lot)

Off-peak inventory (churches, schools, offices).

KPI: revenue per space, misuse reduction.

Jobs-to-be-done examples:

“When I’m circling a dense area, I want a guaranteed, legal spot I can trust, so I can stop worrying and get there on time.”

“When I have an unused driveway, I want to earn safely without hassles, so I can make passive income.”

4) Research

4.1 Generative

Contextual inquiry / ride-along (n=8–10): I observed search & park behavior (pre-drive vs in-flow), recorded decision cues (price, distance, lighting).

Diary study 2 weeks (n=20): logged parking attempts (success/fail, time-to-park, emotions).

Survey (n=200): willingness to pay, trust drivers (photo, ratings), deterrents (fines, tow risk).

Lister interviews (n=12): setup barriers, rule clarity, photo proof, payout expectations.

Key questions

Which top 3 attributes decide selection? (distance vs price vs safety vs covered/EV)

What trust signals actually reduce anxiety? (recent photo, license-plate check, “last verified 7m ago”)

What listing steps feel heavy?, what can we automate?

4.2 Evaluative (usability)

Unmoderated task tests (n=30 per flow):

1) Find & book under 2 minutes

2) Extend a session

3) List a driveway with calendar rules.

Metrics: SUS, task success, time on task, error rate, heatmaps.

4.3 Quant & analytics setup

Instrument used: search_started, results_viewed, filter_used, spot_selected, checkout_started, paid, navigate_started, arrived, extend, end_and_rate, list_created, availability_updated, dispute_opened.

Defined cohorts (commuter/event/host) + funnels (Search→Pay, List→First booking).

4.4 Safety & field testing

Pilot in 2 zones with mixed density, partner lots + vetted listers.

Incident protocol dry runs (blocked spot, wrong plate, tow risk).

5) Insights & UX hypotheses

Trust beats price by ~20–30% in last-minute decisions. Show “Verified X min ago,” recent photos, and plate/height fit first.

Choice overload hurts on mobile while moving. Default to two/three top picks: Closest good fit and Best value.

Extensions are common. Surface a sticky “+30m” one-tap extension throughout the session.

Listers undervalue time windows. Provide smart default pricing & blackout templates.

Photo evidence resolves disputes fast. Require arrival/exit photo prompts, automate with timestamp & geotag.

6) Product requirements (MVP)

6.1 Driver flow (core)

Search: destination or map pan, auto-detect ETA.

Results: ranked by fit score (distance × availability freshness × price × safety × fit [size/height/EV]).

Detail: entrance photos, rules, “last open 5m ago,” ratings, plate fit check, walking time.

Pay: one-tap (saved card/wallet), clear cancellation window, terms.

Navigate: exact entrance pin + gate code instruction card; “I’m here” check-in.

Session: timer, “Extend +15/+30”, support.

Exit: “End & add photo”, rate, receipt.

6.2 Lister flow (core)

Create listing: address pin, guided photos (entrance, spot, signage), size/height/EV, rules (no overnight, tow policy).

Availability & pricing: calendar templates (weekday commute, event nights), dynamic price suggestions.

Verification: ID + bank, optional plate reader for gated lots.

Live management: block dates, quick close, message buyer, dispute center.

Payouts: weekly/threshold, dashboard.

6.3 Cross-cutting

Trust: badging (“Super Host,” “Recently Verified”), safety blurbs, insurance/guarantee summary.

Support: chat escalation paths, 2-tap SOS for blocked/unsafe situations.

7) Information architecture

Entities: User, Vehicle, Payment Method, Listing, Spot (instance), Booking, Session, Dispute, Payout, Message.

Top-level nav: Find, Bookings, List/Manage, Wallet, Profile.

Empty states: teach/onboarding: “No listings yet” → CTA to add, “No bookings yet” → show nearby popular areas.

8) Key user flows

A) Find & book (happy path)

Homepage (search bar) → results (2 “smart picks” + list) → detail → pay → entrance card → timer.

Microcopy: “Verified 4 minutes ago • 120m walk • 1.9m height limit.”

Edge cases:

Spot becomes unavailable during pay → instant swap suggestions, price hold for 5 minutes.

Gate code failure → quick call/chat, auto-apply credit if unresolved in 3 min.

B) Extend parking

Sticky bar with remaining time, “Extend +30m” (shows new total & cost).

Auto-extend toggle with cap (user-defined).

C) List a spot

Camera coach marks (“Take entrance photo from the street, include gate”).

Rule presets: “No blocking sidewalk”, “Back-in only”, “Quiet hours”.

Calendar wizard: start with a template, smart price prompts based on demand.

D) Dispute & refund

Arrival photo vs lister’s current photo, if mismatch (car occupying), immediate cancel+credit, notify host.

Invisible queue for ops review, 24h payout hold on disputed bookings.

9) Design system snapshot

Color: Charcoal base, off-white text, yellow accents reserved for primary actions/state highlights.

Type: 16/24 body baseline, 28/36 for h2, tight numeric tabular for prices & timers.

Components: Search bar (with voice), Result card (photo, badges, price, walk time), Trust badge, Duration control, Payment sheet, Timer bar, Photo capture coach, Calendar grid.

States: loading (skeletal cards), empty (teach), error (retry, support CTA), offline (cached bookings + SMS backup for gate code).

10) Trust & safety model

Verification: Government ID + selfie (listers), plate capture on arrival (optional) to deter misuse.

Evidence: Mandatory arrival/exit photos; geotag + timestamp.

Guarantee: “Can’t access spot?” auto-credit + nearest alternative.

Ratings: Dual (driver & lister) with specific tags (clear signage, accurate photos, courteous).

Fraud checks: velocity limits, device fingerprinting, stolen-card heuristics.

11) Pricing & policy

Dynamic pricing suggestions from demand heatmaps (events, commute hours).

Cancellation: free until start−5min, after start, pro-rate with transparent fee.

No-show: auto-release after 10–15m, partial refund policy.

Host payout: weekly or instant (fee).

12) Measurement (KPIs & formulas)

Activation (drivers): first booking within 7 days / sign-ups.

Search→Book conversion: paid / search_started.

Time-to-decision: median from results loaded → pay.

Availability freshness: median minutes since last verification.

Fill rate (hosts): booked_hours / available_hours.

Cancellation rate, Dispute rate, On-time arrival %.

CSAT/NPS after session.

Rate on bookings.

13) Experiment roadmap

Ranking: “Trust-weighted” vs “Price-weighted” default.

Card layout: map-first vs list-first, 2-3smart picks vs full list.

Trust badge variants: “Verified X min ago” vs “Live sensor” vs “Super Host”.

Extension nudges: reminder at T−10m vs persistent sticky vs push.

Listing wizard: photo coach on/off, effect on disputes & conversion.

14) Usability test (MVP)

Tasks (10–12 minutes each):

Find and book a spot near “LA Summerville Ave” for 90 minutes.

Extend by 30 minutes mid-session.

Create a listing with weekday-only availability and “no overnight” rule.

Success thresholds:

≥85% task success, median decision time ≤ 90s, SUS ≥ 80.

Critical error rate ≤ 5%.

Observation checklist:

Do users notice “height limit”?

Do they trust the verification stamp more than star ratings?

Where do they hesitate?

15) H.E.A.T (Human Experience Audit Toolkit) - My framework audit on critical screens

Emotion Mapping: pre-booking anxiety ↓ via verification stamps, entrance photos, and “guarantee” copy above CTA.

Cognitive Load Simulation: to keep choices to 2–3 recommended cards, expose filters (EV/covered/security) as quick chips.

Dark Pattern Forensics: transparent fees, show total cost before pay, no default opt-ins.

Loop Fatigue: store vehicle & payment, one-tap repeat of recent destination.

Cultural Contrast: units (ft/m), currency, iconography (no text-only signs), right-to-left readiness.

Interface Soundcheck: subtle haptics and short earcon on “Booked” and “Extended”.

Bored Teen Test: single thumb, 5 taps max from open to paid.

16) Content & microcopy

Result card: “2 min walk • Covered • Verified 7m ago • 1.9m max height”

Guarantee: “If your spot is blocked, we’ll rebook you nearby or refund instantly.”

Photo coach: “Stand on the sidewalk. Capture the entrance including any signs.”

Extension prompt: “Running late? +30m ($6.50). Ends 7:40 PM.”

17) Technical notes

Availability freshness: combines host calendar, optional sensor/IoT, and user “I’ve arrived/left” signals.

Maps: door-to-entrance pin + indoor/lot overlays, precise geofences for “I’m here.”

Offline mode: cache booking + gate code, SMS fallbacks.

Performance: results in <800ms, prefetch detail for top 2 cards.

Privacy: stores location with strict retention, clear opt-ins, PCI-compliant payments.

18) Launch plan (alpha → beta)

Alpha (closed): 200 drivers, 50 hosts in two neighborhoods, concierge support.

Beta (open waitlist): event partnerships (stadiums), business lots off-hours, referral codes.

Success gates: Search→Pay ≥ 25%, Disputes ≤ 2%, Fill rate ≥ 40% weekday, CSAT ≥ 4.5/5.