Role
UX Designer & Product Designer
Year
2026
Tools
FIGMA · INFORMATION ARCHITECTURE · DASHBOARDS
DURATION
The starting point
I live in Verbier.
I noticed something.
Guest side
Restaurant side
Mountain restaurants run on compressed seasons and tight margins.
A no-show during peak February is a table that could have been turned twice.
Most still manage reservations by phone, with no deposit and no real recourse — paying someone to handle bookings between service runs.
Problem statement
"How might we create a reservation experience that's fast and frictionless enough for a premium guest — while building in enough commitment to protect the restaurant from no-shows?"
Before touching Figma
Three decisions
that shaped everything.
Who we're designing for
Meet Marcus Sinclair
"If I have to call to book a table,
I'll find somewhere else."
Marcus Sinclair · 38 y/o · London · Management Consultant
Goals
Frustrations
How it came together
From napkin sketch
to native app.

Lo-fi
The biggest call here: removing the reservation placeholder from the home screen. It was redundant. The carousel below did the same job better. Less noise, cleaner hierarchy.
Discovery

Home

Restaurant
Prioritized direct search and top-rated venues above the fold to cut down browsing time for high-intent users.
Booking

Reservation

Confirm & Book
Replaced the standard calendar wall with an immediate time-slot selector.
Users see exactly what’s available the moment they land.
Resolution

Success
Kept the confirmation screen strictly functional.
The map and reservation details are front-and-center so the user can take a quick screenshot and move on.
Visual direction
The UI needs to feel as premium as the dining experience it facilitates.
Warm Parchment
Sparse Bronze
Graphite
Bone
Six screens.
One clear journey.




WATCHING IT WORK
The Flow, Stitched Together
A few decisions worth explaining
Design Decisions

Why the bottom sheet
Why a card hold, not a deposit
A deposit implies distrust.
A hold — 20 CHF per guest, only triggered on no-show — solves the restaurant's problem without penalising the guest upfront. "Nothing is charged today" took a few iterations to land.
Four words, significant difference.


Why the hero rotates on return
Showing Le Rouge when Marcus comes back signals the app knows his context and is already thinking ahead.
The difference between a platform that logs transactions and one that feels like a local recommendation.
Handling the Unhappy Path
Graceful Cancellations
The Intent
The Reassurance
The Resolution

One state changes when it's too late
ONE MORE THING I FOUND IN THE FIELD
Late Isn't Rude. It's Bad Math.

I wasn't looking for this one. It came up during a chairlift chat with a ski instructor.
If this were real
How I'd know it worked.
This is a concept, not a shipped product — so there's no real data yet. But if I were validating it, here's exactly what I'd be watching.
Booking completion
Where do people drop off? My bet is the card hold screen — that's the moment trust is tested the most.
Time to confirm
If it takes Marcus longer than 90 seconds, something's wrong. That's the number I'd design against.
No-show rate
The metric that actually matters to the restaurant. This is the real test of whether the card hold works.
Repeat bookings
Completion is one thing. Coming back is another. That's the difference between solving a transaction and building a habit.









