Jae Hwang - This App Makes $2M/Year With This Viral TikTok Strategy

In 2023 Bloom massively scaled TikTok paid ads. CAC performed worse from mid-2023 and had dropped off by end of 2023, so in 2024 they doubled down on scaling organic growth instead.

Narrative videos: 'we saw a pattern of how billionaires raise their kids' covering what they eat, where they go to school, what education they get, then leading into 'they start learning to invest early' so viewers think starting early could make them a millionaire. Produced at scale with HeyGen AI avatar videos and micro-influencers. The hook borders on get-rich-quick, but the app teaches long-term recurring investing and compounding.

An agency specializing in AI content generation ran 5-10 TikTok accounts that already had decent following and engagement, mass-producing content at about two posts per account per day. When a post went viral, it was propagated to ~20 accounts to capture as many views as possible, then a comment team worked on conversion. An internal team ran the same playbook on new accounts with no following: warm up, then post.

Organic videos avoid 'download Bloom today' because viewers read that as an ad. Videos share screens, narratives and stories instead. Once a video gets views, a comments team posts in the comments that 'this is the app called Bloom', so viewers find the app on their own.

Micro-influencer program run mostly through an agency (Lauren from Pebble, introduced by the host). Creators are chosen by the quality of their videos (camera angle, lighting, narrative, trend awareness), not follower count. Recruited creators start new accounts from zero, warm them up, then scale. Jae also recruits creators himself through SideShift and Handshake, and uses SpyTok to research trends. He spends time on TikTok daily spotting trends and worked with several teams in parallel.

Two contract types. (1) $1,000/month flat retainer plus CPM bonus once a video passes 10K, 50K and 100K views; (2) $500 for 60 videos/month with a higher CPM. CPM bonus is $2-4 per 1,000 views. Bonus starts at 10K views for new creators on new accounts, at 100K for creators with 5-10K followers; capped at $1,000 per video (~1M+ views). Logic in three phases: retainer rewards early volume, 10K-100K bonuses keep creators motivated to make better videos, 100K+ (where conversion kicks in) pays real money. Earlier deals were pure CPM.

Each creator runs two accounts under different names and cross-posts to test the algorithm. When one account picks up better, they post both daily videos there instead of one each. Profiles look personal (a person's face, 'Learn to invest with XYZ') while the content is somewhat branded.

Organic virality spikes then fades after about a month, making downloads swing up and down. Once high-view organic videos taper off, Bloom runs them as paid ads targeted at the right age group and country and doubles down. Organic acts as the testing loop so less paid budget is wasted.

When an organic video becomes an ad, Bloom sometimes adds a final ~15 seconds showing the app (not in the organic version), with a CTA that still sounds organic: 'Download now to flex on your friends' rather than 'to be successful'. Other videos (e.g. the Reddit-story 'underrated iPhone apps') run unchanged with only a small 'Download Bloom' disclaimer at the end.

Brain-rot style videos: a Reddit-style story or 'underrated iPhone apps' narration over maze-runner gameplay, describing Bloom (buy stock with hand-holding, free stock for completing lessons) with a small 'Download Bloom' disclaimer at the end so it reads as organic. Also Reddit-style carousels.

In H1 2025 Bloom targeted StudyTok groups with relatable content for college students on getting good grades and investing money properly, riding the StudyTok and AI-homework trend.

Replaced a single word-heavy onboarding screen with several one-idea slides users tap through like Instagram stories (tap right to advance). Removed the 'Get Started' button from those slides; without it users read everything instead of skipping. The same story format is used for learning content.

Onboarding revised almost every quarter. Order: urgency first (invest X at 18 and have $1M by 65; wait until 30 and end with ~$340K), trust badges (Forbes, Fox, Yahoo!, SIPC, YC), CTA copy changed from 'Continue' to 'Let's go' for college students, animated meme-stock visuals (Netflix, Tesla), frequent color changes so attention doesn't drop, a skip option, and customer reviews last for trust.

Signup CTA promises free stock 'up to' $75. On signup users get roughly $2-4 in free stock; playing games and completing lessons earns in-game currency to buy loot boxes that unlock fractional shares. Stock was chosen over free cash to build a long-term investing mindset. Same subscription funnel used for 2-3 quarters.

Rotated the advertised 'up to' free-stock value between $40, $60 and $75 to see whether it moves subscriptions; $75 is live for this and next quarter.

Planned: spin content from trending Bible/prayer and therapy/self-care movements, e.g. Proverbs 13:11 on not wasting money, or a 'how to become a better human' carousel ending '1. do this, 2. invest wisely, 3. don't waste your money, 4. download Bloom'. Test on one or two accounts in the coming month.

Quitter's organic content and ads target Christians, who care about the pain point, while the app itself is not a Christian app.

Uses Veo (2 is good enough) to generate faceless 'view' videos instead of HeyGen talking heads: camera moving through a house with an ocean-view balcony, or a McLaren/Porsche pulling up, overlaid with text like 'If you invested in Netflix 5 years ago, you might be sitting in this house.' Prompts can be written with GPT. Jae recommends AI video as the cheapest way to grow one account on a ~$10K budget.