Video Summary - 0 to $300k/mo in 45 days with my ai app (just copy me)

Business outcome (Halo AI)

Founder: Dillian (bootstrapped)
Product: Halo AI — AI photo editor (upload photo + write prompt → generate edited output)
Speed to scale: $0 → ~$300,000 MRR in 45 days

Key scaling milestones

Social reach:
- 0 → 600M views in 60 days
- 0 → 1.2B views in 120 days
Creator scale: 85 creators posting
Output scale: ~300 posts/day
- Early: ~10/day → later ~25 shown in timeline

Profit / margins

Reported: ~50% margin at current scale
- Later interview mention: ~$150k profit at $300k MRR
Strategy: reinvest 100% into growth (marketing + scaling distribution)
Early constraint: initial creator payments funded via ~$100k personal loan
- Later goal: reach higher scale without sacrificing unit economics

Core growth strategy: UGC “content machine” (simple, math-driven)

Dillian models revenue as:

Revenue ≈ Total Views × Conversion Rate (from those views)

This leads to a “maximize two levers” operating playbook:

Increase views
Increase conversion

UGC views engine equation (inputs)

Total Views = (# Creators) × (# Videos per Creator) × (Posts per Video / platforms) × (Views per Post / format quality)

Practical implications:

Hire more creators
Increase creator output volume
Cross-post to more platforms
Optimize the single highest-performing format (avoid format diversification)

The “single format” playbook (no content angle variance)

Only one TikTok-style format used end-to-end
Claim: ~1.2B views driven by one format
- Thesis: repeatability > experimentation chaos

“Golden goose” format evaluation criteria

Viral potential
Repeatable by creators
Converts (drives app installs / paid users)

Example winning format structure

Absurd hook + generated image
- e.g., “Pranking my dad I cooked his koi fish”
App demo loading screen
- big app name/logo as a subtle CTA
Story build-up + payoff using a text-message conversation
- audience waits for the twist
No post-ending beyond the conversation payoff
Mostly pranks (designed for high shareability)

Viral distribution / power law expectations (creator planning)

Virality is modeled as “lumpy”:

~5% of videos reach 100K+ views
~1% reach 1M+ views

Therefore scaling volume matters:

Many creators posting daily → daily output compounds into large daily view totals.

Creator acquisition, onboarding, and QA process (ops system)

Where creators come from

Reddit groups (UGC Creators / UGC Brands)
Hundreds of Facebook groups
Mentioned options: paid creator marketplaces
Process:
1. Post job across groups
2. Receive ~500–600+ applicants initially (early scale)
3. Screen for fit
4. Run paid trials

Trial design (5-day paid test)

Trial length: ~5 days
Trial payment: $20 per video
Trial output: 5 videos per creator × 4 platforms (TikTok, Instagram, Facebook, YouTube)
Cost per trial: ~$100 per creator

Evaluation approach (not a hard “views threshold”)

Because algorithms follow power-law behavior, 5 videos isn’t enough to “expect” a breakout
Uses heuristics + qualitative checks, including:
- Consistency: posts daily + proactive communication
- Viral editing intuition: hook / emotional build / payoff structure
- English-first-language heuristic
- Gen Z relevance heuristic (when targeting younger audiences)
- US targeting fit: US residence or US VPN, plus app/tech familiarity preference
- Format compliance (since the format is fixed)

Ongoing operations: Discord as the execution layer

Creators join a Discord
Each creator gets a private channel
Daily communication + feedback loop
Channels for:
- Wins
- Ideas
- Feedback

Community/camaraderie is treated as a critical retention + motivation mechanism (not “post-and-forget”).

Creator pay structure: performance-based + optimized for simplicity

Pay structures tested (then standardized)

He explored 4 common models:

Fixed rate (no views incentive)
CPM with cap
Mixed (base per video + CPM)
Milestone bonuses (base + bonuses at view thresholds)

What they used operationally (standard for everyone)

Base: $20 per video
Bonuses: milestone-based with a base
Initially creators could choose CPM vs milestone, but:
- 29/30 chose milestone even though CPM paid slightly more on paper

Why standardize to milestones:

Better aligns incentives to maximize views
Operationally easier: pay triggers at thresholds (avoids recomputing CPM differences over time)

Milestones + payout cliffs (examples)

Early milestones mentioned: 20K, 100K, 500K, 1M
Later refined milestones mentioned: tiered cliffs (details partially unclear in subtitles), including 200, 500, 800 and additional point steps

Payout logic (as described):

Bonus roughly aligned to ~$1 per 1,000 views on average, but “cheaper than $1 CPM” due to structure
Mentions of residual views as additional benefit under certain interpretations (not fully formalized)

Key risk/reward mechanic: payout only once per multi-platform video

A creator video is posted across multiple platforms
One bonus trigger even if multiple platforms surpass thresholds:
- No stacking multiple bonuses per creator for the same video milestone
Purpose: increase motivation while controlling company costs

The “viral amplification” loop: replicate winners instantly

When one creator video goes viral, automation sends it to all creators to recreate the same viral script/format the same day
Claim:
- a viral hit becomes a “spike multiplier” across the creator network
- due to power-law dynamics + replication, view acceleration compounds even if not everyone matches viral scale

AI tooling for creator coaching + trend ingestion

“OpenClaw” agent (internal creative coach)

Trained on million+ view videos
Workflow:
- download videos
- use Gemini for transcripts
- store data as local text files
Functions:
- generate new scripts / video formats consistent with what’s worked
- pattern-match successful structures
Usage claim:
- creators ask OpenClaw daily for scripts and improvements

Trend integration (TikTok signals)

OpenClaw connected to TikTok trend data via a provider:
- trending keywords and hashtags
- sound mentions noted as important; hashtags used to identify trends rather than directly drive virality

Funnel + distribution strategy (execution detail)

Distribution channels (top-of-funnel)

Posting on 4 platforms:
- Instagram, TikTok, Facebook, YouTube

Funnel steps described:

Views
App store / web search / store discovery → download
Trial or paywall
Subscription conversion

Audience sizing reference (US focus)

US targeting heuristic cited:
- ~100M US people with Instagram + iPhone as an addressable pool reference

Monetization: paywall experiments and onboarding design

Pricing examples shown

Weekly plan: $8.99/week
Testing:
- weekly: “has been really good”
- yearly tested, removed, then testing yearly again

Additional paywall variants mentioned:

Trial cancel paywall
Abandon cart offer
“Exit offer”

Paywall optimization: two-step “soft + hard” paywall

Onboarding paywall (skippable) + gated paywall at first key action
Reported conversions:
- each paywall individually: ~16.5% conversion
- combined: higher total conversion than a single hard paywall

Hypothesis:

Users need to see value before committing (trust issue)
Paywall triggers right when users hit high intent (“submit prompt” / attempt generation)

Onboarding flow design (reduce drop-off)

Users can:
- browse templates
- upload photo + type prompt
Paywall appears on submit, right before they see the output:
- “At the border” behavior: once they take action, they convert.

Capacity planning and scaling targets

Forecasting and plateau

Forecasted plateau at ~$300k MRR, claimed to hit “exactly”
Ops remained “stable” while scaling creators/videos/views

Next scaling goal

Goal: scale from $300k MRR → $1M MRR by June
Plan:
- continue scaling UGC
- expand distribution
- add paid ads
- add other channels

Break-even timeline (cashflow framing)

Break-even in first ~4–5 weeks of early scaling from 0 → $300k
After loan investment:
- claimed break-even for more than a month, then profit

Unit economics / profitability claims (high level)

Early UGC insight: conversion stability when a video reaches ~1M views
Claimed early ~6:1 ROAS / return multiple:
- “Okay, just take out all my money…”
Current:
- claimed ~50% margin
- growth approach: “grow at all costs”, reinvesting everything into acquisition/distribution

Key operational recommendations implied by the playbook

Pick one winner format and enforce it (avoid creator “angle drift”)
Scale views mechanically using the UGC views equation (creators × output × platforms × format performance)
Use paid trials + qualitative viral heuristics (don’t require immediate viral outcomes in trial)
Standardize pay with milestone cliffs to reduce ops complexity and align incentives
Replicate winners fast via automation (turn one viral outlier into multi-account distribution)
Coach creators continuously in a community hub (Discord) to sustain consistency
Optimize paywalls around intent timing (soft/skip early, hard gate at first “submit” action)

Presenters / sources

Presenter / guest: Dillian Burma (Halo AI; creator/operator)
Host: Joseph Choi (Founder, Consumer Club; Superwall podcast)

0 to $300k/mo in 45 days with my ai app (just copy me)

Key takeaways