Pre launch / MVP
Competitor research
Identified FitBot as a popular but flawed workout app — workouts were described as 'weird and even dangerous.' Decided to clone FitBot's UI/UX but build a proper workout engine underneath. Built the MVP in roughly 2–3 months.
→ Functional MVP ready to launch
Launch
Reddit
Posted a detailed technical thread on Reddit about how Gravl (then called 'Games AI') was built, including the engineering and technical specs. Targeted developer communities who also had gym interests.
→ 300K+ impressions, hundreds of likes within first few hours, first ~2,000 users
MRR $0 Users 2.0k users
Post launch validation
Organic app feedback
Kept the app free initially and collected bug reports and feature requests from the Reddit-driven user base. Used this signal to confirm product-market fit before monetizing.
→ Sufficient confidence to add subscription paywall and move to paid ads
Monetization start
Paid ads meta
Added a hard paywall (subscription required before user sees the generated workout) and immediately activated Meta ads. Got a paying subscriber within the first 10 minutes of running ads.
→ First paid subscriber within 10 minutes of ad activation
Early scaling
Paid ads spanish market
Localized the app into Spanish and launched paid ads targeting South America, where CPMs and CAC were significantly cheaper than in the US. Julian and Matias are native Spanish speakers, making this a natural fit. Spent less than $50/day on ads.
→ Lower CAC than English-language markets; sustainable early growth
Scaling
Ugc ads
Scaled ad creative through UGC (user-generated content) videos. Used AI video tools and CapCut to produce ads cheaply and at volume. Emphasized that the key is volume of creative testing, not perfection of any single ad.
→ Core driver of paid ad performance at scale
Scaling
Competitor ad research meta library
Used Meta Ads Library to analyze competitor ads — identifying which creatives competitors were spending the most on and reverse-engineering what was working. Explicitly copied successful ad formats as a starting point.
→ Reduced creative experimentation waste; accelerated ad performance
Retention / growth
In app human support
Built a 24/7 in-app support chat staffed by real humans (not AI bots). Julian highlighted this as a key differentiator that users explicitly valued, contributing to retention and positive reviews.
→ Noted as a major contributor to growth and user satisfaction
Current
Paid ads meta tiktok
Continued scaling across Meta and TikTok (with smaller spend on Google and Apple Search Ads). Ad spend runs at roughly 1/3 of total revenue. Team grew from 3 founders to 13–14 people, with Matias and Aaron dedicated to growth and ads.
→ $440K MRR, 70,000 subscribers
MRR $440k Users 70k users