Neighborly
Neighborly Native App
Neighborly Native App: One Framework, Shipped Twice
My role
Led UX strategy, IA, research, and booking flow design
Result
The OPUS jobs framework (Repair / Maintain / Enhance) became the app’s primary navigation and core booking flow
Timeline
5 Months
The problem
The app was live but fell short for both sides of the marketplace. Customers couldn’t easily browse or
understand 20+ service brands, and franchise owners weren’t getting the booking volume or lead quality
the platform promised. The relaunch was scoped to fix that with clearer brand descriptions, offers by
zip code, a real booking flow instead of a lead form, and live job status.
Decision 1: Reuse the strategy instead of redesigning from scratch
The app’s home screen is a direct application of the framework validated on
the web platform. The mental model that organized a 20-brand website migration is what
a customer taps on day one.
Decision 2: Benchmark beyond the category
I audited direct competitors (Urban Company, Pinch, JOBOY) and adjacent apps with high-trust booking
flows (Marriott Bonvoy, Airbnb, Hotels.com, Expedia, TripAdvisor). Home services have more in common with
booking a stay than with e-commerce checkout: infrequent, higher stakes, tied to a time and place. That
shaped the multi-step booking flow and the “upcoming jobs” screen.
Decision 3: Add friction where it prevents bad leads
The flow runs Home → Brand Selection → Brand Details → Service Type → Booking. I added a
lightweight service type bumper between brand details and booking, because a single brand (Five Star Painting,
for example) offers service types with different booking requirements, and skipping that step was sending
franchise owners mismatched leads.
Decision 4: Design one flow for two mindsets
The booking flow itself, service frequency, schedule selection, contact details, review, confirmation, was
designed around the same two personas from OPUS:
- Diane (Adapter) - Diane gets visibility and control: a step
indicator, a review step, and exact preference inputs.
- Elizabeth (Optimizer) - Elizabeth gets speed: guest checkout,
sensible defaults, and a “call us directly” option on every screen.
Decision 5: Bring people back without making them remember
Offers surface by zip code and brand. A persistent “upcoming jobs” view consolidates
every scheduled service across every brand the customer has used, with a fallback to add
a phone or job number for call-center bookings.
Results
Framework carried from web platform to app navigation and booking
Yes
Booking completion rate (flow started → confirmed)
70%
Mismatched leads sent to franchise owners
−35% after the service type bumper
Bookings completed as guest
~80%
Guest to registered user conversion
~40%
Repeat-booking rate across brands
~30%