Self-Serve Setup For Restaurants
How I built a self-serve flow to pitch a scalable onboarding solution.
Eatsy’s onboarding was slow, manual, and holding back growth. Leadership wanted a self-serve system but they didn’t know where to start.
So I took the lead: identified the core UX problems, explored AI-assisted solutions, and built a working prototype in a few days.
The goal is to turn a vague idea into a clear, scalable vision the team could get behind.
Year
Mar, 2025
Platform
Website / Mobile
Client
Eatsy
Role
Product Designer
The problem was obvious
Eatsy’s operations team was drowning in restaurant onboarding by collecting menus over LINE, WeChat, photos over email, translating random PDFs.
The owners? They never even touched the backend. That worked at first, but it was clear: they couldn’t scale like this.
The CEO told me they wanted to shift this to a self-serve model to let restaurants manage their own info. But they had no clue how that might work. No UX, no flow, not even a rough plan. Just: “it should be easy.”
That’s when I jumped in.
What is Eatsy?
Eatsy is a restaurant booking platform designed for the Asian market. It helps people discover local spots, browse menus, and make reservations — all in one place.
I saw a gap and an opportunity.
When a business wants something but can’t visualize it, that’s a designer’s moment to lead. I wasn’t going to wait for specs or a Jira ticket. I decided to pitch a solution. I started by asking:
What would make a restaurant owner actually want to fill this out?
How do we design this so they don’t need to be trained?
How do we make sure the data they submit is actually usable, not a mess?
The system I pitched
I built a flow that breaks down setup into 6 lightweight chunks. Each one focused, optional, and visually guided. You can do it in 10 minutes, or chip away bit by bit. Here’s the breakdown:
Every step includes previews. No guessing. No surprises.
Owners see exactly what customers will see.
I designed it to feel like setting up a store on Shopify, not filling out a government form.
Business Info: Auto-filled when possible. Preview branding instantly.
Hours: Templates for common restaurant patterns — no tedious typing.
Menu: Upload photos, type names, select tags. Multilingual-ready.
Gallery: Drag-and-drop visuals to tell your story.
Reservations: Set capacity rules, buffer time, cancellation policies.
Specials: Events, seasonal menus, limited-time deals.
User Flow for Restaurant Owners
How I used AI (for real)
I used AI like a creative partner to move fast and think clearly.
The goal wasn’t to automate design. It was to reduce friction between idea and prototype, and it worked. I went from sketch to clickable demo in days. That made the pitch more than just a suggestion. It was a vision they could feel.
Lovable (AI prototyping)
Turned rough ideas into clickable flows fast—way faster than manual wireframing.
I could test directions without the overhead.
Gamma (AI pitch builder)
Turned rough ideas into clickable flows fast—way faster than manual wireframing.
I could test directions without the overhead.
Edge case modeling + flow logic (via AI prompts)
Used AI to test onboarding paths, error states, and user journey variations
Caught blind spots early before they became blockers
My AI-Accelerated Process
The Prototype
🎥 Watch the flow in action
What came out of this?
The CEO and ops team shifted from “We don’t know how to build this” to “This gives us a clear direction.” The pitch helped them:
Align on what matters to restaurant owners
Understand what could be modular, automated, or templatized
See how this system could scale to more cities and languages
Is it live? Not the point.
The impact was this: I helped unstick the strategy.
I used AI tools not for the sake of trend, but to speed up clarity and reduce ambiguity.
