PawsRoam
Anywhere, With the Ones You Love
A story about building the product you wish existed in the world — designed for pet owners, built with 7 months of Tokyo research, shipped to Firebase December 2025. From a single conversation at Computex to a live app with 200MB in downloads. This one is personal.
The Origin Story
It started at Computex. I was there doing what I always do at conferences — moving through the floor, listening, finding the people who are thinking about what comes next. That's where I met Kenta Uehira, a VC from Japan. We talked about pet culture, about travel, about the fragmented experience of being a pet owner in a city that wasn't built for your dog or your cat. He listened. Then he said something that stopped me cold:
「福岡で必ず火がつく」
"This will definitely catch fire in Fukuoka."
That sentence became the ignition spark. Not because a VC said it — I wasn't chasing validation. It was because he was right. And I knew it. The problem was real. The market was real. The timing was right.
I started learning Japanese. Not with a class, not with an app — with intention. I wanted to walk into rooms in Tokyo and earn that conversation, not rely on someone else to translate it. I traveled back. I carried business cards. I walked into pet cafes, veterinary clinics, sitter communities. I sat with pet owners and asked them what frustrated them, what scared them, what they wished existed. I was doing the research because I am the user. The problem didn't happen to a persona I invented — it happened to me, and to my friends.
On May 27, 2025, I opened Figma and started building. React, Firebase, Kubernetes. Seven months of heads-down focus. On December 22, 2025, the MVP went live on Firebase. And then — within 7 days — 200MB of downloads.
After 7 months of heads-down building, this was the message I sent to Kenta.
上平さん、
お世話になっております。Ed Chenです。
以前、PawsRoamについて「福岡で必ず火がつく」と
おっしゃっていただいたこと、とても励みになっております。
実は、過去7ヶ月間、製品開発に集中してきました。
現在、アプリはほぼ完成しており、
いよいよ市場への展開を開始する段階に来ました。
上平さんのご意見を伺いたく、
お時間をいただけないでしょうか？
具体的には：
1. 福岡での最適な市場参入戦略
2. 初期に接触すべきペット関連のコミュニティやお店
3. もし可能であれば、ご紹介いただける方々
オンラインでも対面でも、ご都合の良い形で
30分ほどお話しできれば幸いです。
何卒よろしくお願いいたします。
Ed Chen
PawsRoam Founder
Sent after 7 months of building. Not a pitch. An update to someone who believed in this before the product existed.
2. The Problem — I Am the User
"The product I was building wasn't a portfolio piece. It was the thing I needed."
Pet ownership is emotional. Your dog isn't a possession — they're family. And when you travel, or when life gets complicated, the existing tools for managing that relationship are scattered, unreliable, and designed without empathy. I've experienced every one of these pain points personally:
🗺️ Fragmented Discovery
Finding pet-friendly accommodation, cafes, parks, and clinics across a new city means cross-referencing three apps, a Google Maps search, a Reddit thread, and a Facebook group that hasn't been updated since 2022. The information exists — it's just scattered everywhere.
😟 Trust Anxiety in Sitter Marketplaces
Handing your dog to a stranger is terrifying. The current marketplaces have reviews, but reviews can be gamed. Background checks are inconsistent. Verification processes vary wildly by region. When you're leaving your family member with someone you've never met, "mostly good reviews" isn't enough.
🚨 No Emergency Infrastructure
Your cat gets sick at 11pm in a city you don't know. There's no unified system — no SOS button that finds the nearest emergency vet, alerts your sitter, and contacts your emergency contact simultaneously. You're Googling in a panic, in a language you might not speak fluently. That shouldn't be the experience.
These aren't hypothetical user stories. They happened — to me, to friends, to the pet owners I interviewed in Tokyo neighborhoods from Shimokitazawa to Nakameguro. Real problems with real stakes. That's what PawsRoam is built to solve.
3. Product Design — On the Ground in Tokyo
I don't believe in assumptions. Before a single production line of code was written, I was in Tokyo — walking neighborhoods, entering pet cafes, speaking with veterinary clinic staff, sitting with dog owners in Yoyogi Park. This is what that research looked like.
On-the-ground research: pet cafes, veterinary clinics, parks, sitter communities — Tokyo.
4. Core Features
Pet-Friendly Discovery
500+ verified pet-friendly spots. Map-first UX — because location context matters when you're traveling with a dog. Filter by pet type, pet size, establishment type. Every entry is manually verified for accuracy.
Sitter Marketplace
1,000+ verified sitters. Drop-in visits, dog walking, daycare, overnight boarding — all services available. Background checks, review system, sitter profiles with photos. Booking and payment in-app. Available services: Drop-in Visit, Dog Walking, Daycare, Overnight.
Emergency SOS
One-tap emergency contact. Nearest vet locator. Travel safety net — because emergencies don't wait for business hours. When you're in a foreign city and your pet is sick at 11pm, this is the button you wish existed.
App Testing Sessions
5. Technical Architecture — Built to Ship
I made deliberate architectural choices to maximize speed-to-market without sacrificing scalability. Firebase Hosting on the Spark plan meant zero infrastructure cost during MVP validation. Kubernetes handles the backend service orchestration. React 18 powers the client. The stack is pragmatic — built to prove the product, not to impress engineers.
FIREBASE DEPLOYMENT — LIVE METRICS
200MB
Asset downloads in 7 days
(Firebase Hosting bytes, week of Mar 22–28.
Note: Firebase reports data transfer in MB, not user counts. Actual user count tracked separately via Auth metrics.)
Dec 22
Deployed to
Firebase
2025
7 mo
Idea to
MVP
May → Dec 2025
$0
Infrastructure cost
Spark plan
Admin Dashboard
6. Internationalization — Designed for Japan First
Most products add international support as an afterthought. PawsRoam was designed Japan-first from the first commit. The localization wasn't bolted on after the English version was done — it was baked into the architecture from day one. Japanese, English, and Traditional Chinese (繁體中文) are all supported in v1. Not because it was easy. Because it was right.
Market Priority: Fukuoka → Tokyo → Shizuoka
- Kenta's insight about Fukuoka's pet culture validation led to Fukuoka first
- 15.9 million dogs and cats in Japan (2024) — more pets than children under 15
- Japan's declining birth rate is driving pet humanization — pets are family, not accessories
- Ed learned Japanese specifically for this market — the language investment signals the commitment
- High density urban environments (Tokyo, Osaka, Fukuoka) create acute pet-travel friction — the problem is more visible here
Localization Architecture
- i18n framework built into React from initial architecture
- Language switcher in navigation — JP / EN / 繁体
- Date, currency, and address formats localized per region
- Content (spot listings, sitter profiles) tagged with language availability
- Firestore documents structured for multi-locale content delivery
Japan market research — understanding digital distribution channels, local SEO, and how Japanese pet owners discover services online.
7. What's Next — May 2026
8. What PawsRoam Proves — B2C Product Capability
For hiring managers and product teams evaluating B2C experience: PawsRoam isn't a portfolio piece assembled to check boxes. It's a real product with real users, real traction, and a real investor meeting on the calendar. Here's what it demonstrates:
End-to-End Product Ownership
From 0 to MVP: user research in Tokyo → Figma prototypes → React 18 frontend → Firebase backend (Auth + Firestore + Hosting) → Kubernetes for backend services → live deployed product. No handoffs. No dependencies on other engineers. No "I designed it but didn't ship it." This is what full-stack product ownership looks like — one person, end to end, live in production.
Real User Research — Not Assumed Personas
Multiple research trips to Tokyo. Conversations with pet owners in Yoyogi Park, Shimokitazawa, Nakameguro. Interviews with sitters, veterinary clinic staff, pet cafe operators. The app's features weren't derived from competitive analysis — they were derived from watching real people struggle with real problems and asking them directly what they needed. That's B2C product research done right.
International Product Thinking — From V1
JP/EN/繁 from version 1. Japanese VC relationship built from Computex. Japan-first go-to-market. Learning a new language to access a new market. This is what it means to think internationally — not translating a product for a new market, but designing it for that market from the ground up. Built to cross borders, not just ship features.