The original is one click away. Open original ↗
How Roger Chen built two number-one apps without paid growth
Executive overview
Most consumer social apps die because they can't generate network density fast enough. Roger Chen solved this twice — first with Lobby (500k DAUs, #1 in five countries) and then with Bro (AI companion, #1 US App Store), both using product-led viral mechanics and near-zero paid marketing.
The core method: prototype in Protopie before writing production code, use cheap paid ads to validate concepts in days not months, then build product features that force organic sharing.
The insight is that online hangouts fail because they lack the spontaneous, low-stakes context of real-world social spaces — and that gap can be engineered.
Validating before building
- Build a Protopie mock-up (not Figma, not code) and use it daily to find what breaks before engineering starts
- Buy small-scale ads ($2–5k) during the idea-to-launch gap to get format feedback fast — ads are for validation early on, not for scaling
- Test 5–6 creative formats; expect results to be unpredictable — the videos you expect to fail often outperform
- For Bro: a 7.9M-view video was filmed using a prototype before the app was finished
- The goal of early ads is two outputs: one format that works, and confirmation people want the product
Product-led growth: how Lobby scaled
- Lobby's core insight: online hangouts need background activity (YouTube, SoundCloud, games) so silence isn't awkward — the model was the college dorm lobby
- Growth stalled for a year until Lobby went viral in Israel; the single measurable difference was 60% of Israeli users brought 5+ friends on day one vs. 20% elsewhere
- This revealed the need for network density — not just relationship density but temporal density (being online at the same moment)
- The group picture feature triggered shared excitement: a countdown + phone vibration + boomerang, then users share to Instagram/Snapchat Stories with watermark — friends who missed it came next time
- The one-minute call feature reduced commitment pressure: a randomised nightly prompt to catch up for exactly 60 seconds, after which the chat auto-ends
- After density features shipped, Lobby hit #1 in the UK, Italy, Germany, Egypt, and Saudi Arabia via organic spread through international school networks
Scaling Bro with creators
- Validated product-market fit with one viral format (floating AI window over texts), then recruited ~50 TikTok creators around relationship/texting hashtags
- The winning TikTok format: real-person hook showing a relatable scenario, then demo — people stop for faces, not app interfaces
- Best-performing videos were unpredictable; the autocorrect-to-crush video (2.3M views) worked because the demo was embedded in the story, not bolted on
- Scaling UGC content is more chaotic than product work — no clear rules, requires volume and patience
AI companions and the social vision
- Bro is designed to augment social relationships, not replace them — AI should live inside apps you already use, not in a separate "black box"
- The floating picture-in-picture model mirrors FaceTime windowing; the framing is a convenient in-between, like DoorDash vs. a restaurant
- Insight from Lobby carried forward: a concrete AI character can act as a "third person" in group chats — a hype man who opens new topics and keeps conversation moving
- The goal is using AI to create context, not content — more authentic human interaction, not more AI-generated output
- Future direction: a social-first app where AI bridges awkward silences the way a yappy friend does in a group
Founder-market fit
- Founder-market fit is real: passion for the problem domain determines whether you can sustain the detail work required
- Roger's team tried an AI interactive story app (Super Movie) in 2023; without personal passion for it, iteration stalled
- The analogy: the founder of AI Dungeon could stare at text-based worlds all day — that's what deep fit looks like
- Consumer social requires emotional resilience; getting "slapped in the face every month" only works if the design process itself is enjoyable
More like this — when you're ready for early access.
Join the waitlist for a personal account and content recommendations based on what you're working on.
No spam. Unsubscribe at any time.
You're on the list. We'll be in touch before launch.