Onz
Overview
Onz is a food decision-intelligence app that helps people pick a restaurant in under three minutes by cutting through option overload and decision fatigue. Instead of endless listings and generic ratings, Onz centres on curated public and private reviews — turning scattered opinions into actionable, context-aware recommendations. The responsive web app launched with early active users and is designed to evolve into an AI-powered recommendation engine as the dataset grows
Categories
Food Discovery App
Webapp Design
Date
Feb 1, 2026
Client
Sean Khoo
The Problem
People spend over 10 minutes deciding where to eat — longer when choosing for a group. The options are endless, the information is scattered, and by the time you've picked somewhere, the experience already feels exhausting.
This isn't a food problem. It's a clarity problem. Too many options, no personalisation, and no guided path to a decision.
Validating the Problem
Before designing anything, I needed to know whether this was a real problem or just a personal frustration. I ran a survey with 47 participants across Malaysia, Singapore, and Melbourne, aged 20–34.
PROBLEM VALIDATION
91.5% face decision difficulty (sometimes to always)
51.1% spend 10+ minutes deciding
57.4% don't know what they're craving (#1 FRUSTRATION)
55.3% overwhelmed by too many options
36.2% fear picking somewhere disappointing
SOLUTION VALIDATION
59.6% HIGHLY INTERESTED (4-5 rating) in single recommendation
80.9% would use few times per week or MORE
68.1% want MOOD-BASED curation (TOP FEATURE) 🎯
48.9% want AI single recommendation
44.7% want personal food diary
TARGET MARKET
55.3% aged 23-27 (primary target)
66.0% dine out 3+ times per week
91.5% discover via social media (Instagram/TikTok)
59.6% prefer app over web
COMPETITIVE GAPS
55.3% frustrated: same restaurants always appear
46.8% frustrated: takes too long to find
42.6% frustrated: too many choices confuse
36.2% frustrated: reviews don't match taste
Only 6.4% say current apps work fine (HUGE opportunity!)
MONETISATION REALITY
55.3% probably/definitely NOT willing to pay upfront
44.7% willing or maybe willing (depends on features)
46.8% prefer FREEMIUM model ⭐
34.0% prefer free with ads
PROBLEM VALIDATION
91.5% face decision difficulty (sometimes to always)
51.1% spend 10+ minutes deciding
57.4% don't know what they're craving (#1 FRUSTRATION)
55.3% overwhelmed by too many options
36.2% fear picking somewhere disappointing
SOLUTION VALIDATION
59.6% HIGHLY INTERESTED (4-5 rating) in single recommendation
80.9% would use few times per week or MORE
68.1% want MOOD-BASED curation (TOP FEATURE) 🎯
48.9% want AI single recommendation
44.7% want personal food diary
TARGET MARKET
55.3% aged 23-27 (primary target)
66.0% dine out 3+ times per week
91.5% discover via social media (Instagram/TikTok)
59.6% prefer app over web
COMPETITIVE GAPS
55.3% frustrated: same restaurants always appear
46.8% frustrated: takes too long to find
42.6% frustrated: too many choices confuse
36.2% frustrated: reviews don't match taste
Only 6.4% say current apps work fine (HUGE opportunity!)
MONETISATION REALITY
55.3% probably/definitely NOT willing to pay upfront
44.7% willing or maybe willing (depends on features)
46.8% prefer FREEMIUM model ⭐
34.0% prefer free with ads
PROBLEM VALIDATION
91.5% face decision difficulty (sometimes to always)
51.1% spend 10+ minutes deciding
57.4% don't know what they're craving (#1 FRUSTRATION)
55.3% overwhelmed by too many options
36.2% fear picking somewhere disappointing
SOLUTION VALIDATION
59.6% HIGHLY INTERESTED (4-5 rating) in single recommendation
80.9% would use few times per week or MORE
68.1% want MOOD-BASED curation (TOP FEATURE) 🎯
48.9% want AI single recommendation
44.7% want personal food diary
TARGET MARKET
55.3% aged 23-27 (primary target)
66.0% dine out 3+ times per week
91.5% discover via social media (Instagram/TikTok)
59.6% prefer app over web
COMPETITIVE GAPS
55.3% frustrated: same restaurants always appear
46.8% frustrated: takes too long to find
42.6% frustrated: too many choices confuse
36.2% frustrated: reviews don't match taste
Only 6.4% say current apps work fine (HUGE opportunity!)
MONETISATION REALITY
55.3% probably/definitely NOT willing to pay upfront
44.7% willing or maybe willing (depends on features)
46.8% prefer FREEMIUM model ⭐
34.0% prefer free with ads
PROBLEM VALIDATION
91.5% face decision difficulty (sometimes to always)
51.1% spend 10+ minutes deciding
57.4% don't know what they're craving (#1 FRUSTRATION)
55.3% overwhelmed by too many options
36.2% fear picking somewhere disappointing
SOLUTION VALIDATION
59.6% HIGHLY INTERESTED (4-5 rating) in single recommendation
80.9% would use few times per week or MORE
68.1% want MOOD-BASED curation (TOP FEATURE) 🎯
48.9% want AI single recommendation
44.7% want personal food diary
TARGET MARKET
55.3% aged 23-27 (primary target)
66.0% dine out 3+ times per week
91.5% discover via social media (Instagram/TikTok)
59.6% prefer app over web
COMPETITIVE GAPS
55.3% frustrated: same restaurants always appear
46.8% frustrated: takes too long to find
42.6% frustrated: too many choices confuse
36.2% frustrated: reviews don't match taste
Only 6.4% say current apps work fine (HUGE opportunity!)
MONETISATION REALITY
55.3% probably/definitely NOT willing to pay upfront
44.7% willing or maybe willing (depends on features)
46.8% prefer FREEMIUM model ⭐
34.0% prefer free with ads
The Opportunity
One line guided every design decision:
Reduce the time to decide on a meal to under 3 minutes, by creating a restaurant discovery platform that's curated to each user.
Who I'm Designing For
Onz can be used by anyone, but I focused the design around three personas that represent the core frustrations:
1. The Mood-Based Foodie — overwhelmed by options, wants 3 good choices not 500
2. The Social Coordinator — plans where the group eats, juggling everyone's preferences
3. The Efficient Explorer — wants somewhere new but doesn't have time to research
Design Decisions
Decision 1:
Personal taste profiles over generic ratings
A 4.2-star restaurant tells you what the crowd thinks — not whether you'll like it. Onz builds a personal taste profile from your reviews, likes, and saves, then surfaces relevant options instead of generic "Top 10" lists.
Decision 2: Public and private reviews
Public reviews help others discover places. Private reviews are honest — no social pressure to be diplomatic. That honesty makes the taste profile more accurate, which makes future recommendations better.
Decision 3: Social recommendations over algorithmic ones (for now)
The long-term vision includes AI-powered recommendations. But for the MVP, "your friend Sarah loved this place" requires zero explanation — while an AI recommendation requires users to trust a system they've never used. Social first builds engagement while collecting the data the AI will eventually need.
The MVP
Launched February 1st, 2026 at getonz.com. Starting with Klang Valley, Malaysia. ~8 active users and growing.
What's included: Location-based discovery, personal and public reviews, food journal, taste profile building.
What's deliberately not included (yet): AI recommendations (Phase 3 — needs sufficient data), restaurant portal (Phase 2), group decision tools (future).
This is a sequencing decision, not a limitation. The review platform generates personal data. The data enables the AI. The AI delivers the core promise.
Discover Restaurant Reviews near you.
Location tracking technology, for on-the-go review searching.
Write personal reviews or recommend restaurants to your friends.
Using a qualitative approach that suits your preferences.
Responsive Platform
Laptop/Desktop friendly, if you are pre-planning your meals.
Recommend reviews to your friends across all platforms.
Quick & accessible way to spread the word.
Track your personal restaurant whereabouts
All public or private reviews will be stored here.
What I Learned
Clarity beats features. The most common feedback wasn't "I wish it had more." It was "this makes sense." For a product tackling decision overload, simplicity is the feature.
Design the data, not just the screens. The review flow asks specific questions — what did you order, how was the wait, would you go back — that seem like UX choices but are actually data architecture decisions powering future AI recommendations.
Start with trust, scale with technology. Users need a reason to engage before you can personalise their experience. Social proof provides that reason while the system learns.
More works
Not sure what's confusing your users?
Or email me at hello@kaizerstudios.com —
We'll respond within 24 hours
©Sean Khoo 2026. All rights reserved.
Not sure what's confusing your users?
Or email me at hello@kaizerstudios.com —
We'll respond within 24 hours
©Sean Khoo 2026. All rights reserved.
Not sure what's confusing your users?
Or email me at hello@kaizerstudios.com —
We'll respond within 24 hours
©Sean Khoo 2026. All rights reserved.