Nov 13, 2025 · 15m · starter-story

How I Built It: $400K/Month Mobile App (Gravl)

Julian Gargicevich · 10m spoken Pat Walls · 3m spoken
0:00 / 0:00
▶ Watch on YouTube →

gold bands on the timeline = statements, start to end. Hover to read, click to jump. CC turns on captions

Software engineer Julian breaks down how he built and scaled Gravl, an AI-powered workout app generating over $440,000 monthly, through smart competitor reverse-engineering, viral Reddit validation, international paid advertising, and a lean engineering stack.

How this conversation actually went

Every chapter scored 0–10 on four independent dynamics. Hover any point for the reasoning behind the score. Pat holds 26.3% of the talking time here. How this is scored →

Pat as informed peer 1.0 Guest teaching 3.0 Guest disagreement 0.0 Pat pushing back 0.2
05100:0010:001:06–4:01 · Pat as informed peer 2/10 Julian's Background in Engineering and Early Agency Experience Pat acts as a supportive interviewer asking standard exploratory founder questions. Julian explains his background transitioning from an unwanted agency setup to identifying product flaws in market incumbent Fitbod.4:01–6:51 · Pat as informed peer 1/10 Developing the Initial MVP and Custom Workout Logic Pat prompts Julian on his early development timeline and initial traction. Julian details the algorithmic complexity of custom workout logic and using Reddit for initial validation before an ad break.6:52–9:27 · Pat as informed peer 1/10 Scaling Gravl with Paid Ads and International Expansion Pat gently challenges whether Reddit alone drove scale, prompting Julian to break down his paid ad playbook. Julian shares tactical advice on geographic ad arbitrage and creative iteration via Meta ad libraries.9:27–12:09 · Pat as informed peer 0/10 Full Gravl App Walkthrough and Core Feature Demonstration Pat invites Julian to do a full product walkthrough and remains silent throughout the demonstration. Julian explains the onboarding funnel, hard paywall mechanics, and dynamic exercise reordering.12:10–14:55 · Pat as informed peer 1/10 Gravl's Technical Architecture, Backend Stack, and AI Tooling Pat asks about the technical stack and final founder lessons. Julian details his backend architecture alongside an async post-recording breakdown of the company's real P&L and expense margins.1:06–4:01 · Guest teaching 2/10 Julian's Background in Engineering and Early Agency Experience Pat acts as a supportive interviewer asking standard exploratory founder questions. Julian explains his background transitioning from an unwanted agency setup to identifying product flaws in market incumbent Fitbod.4:01–6:51 · Guest teaching 3/10 Developing the Initial MVP and Custom Workout Logic Pat prompts Julian on his early development timeline and initial traction. Julian details the algorithmic complexity of custom workout logic and using Reddit for initial validation before an ad break.6:52–9:27 · Guest teaching 4/10 Scaling Gravl with Paid Ads and International Expansion Pat gently challenges whether Reddit alone drove scale, prompting Julian to break down his paid ad playbook. Julian shares tactical advice on geographic ad arbitrage and creative iteration via Meta ad libraries.9:27–12:09 · Guest teaching 3/10 Full Gravl App Walkthrough and Core Feature Demonstration Pat invites Julian to do a full product walkthrough and remains silent throughout the demonstration. Julian explains the onboarding funnel, hard paywall mechanics, and dynamic exercise reordering.12:10–14:55 · Guest teaching 3/10 Gravl's Technical Architecture, Backend Stack, and AI Tooling Pat asks about the technical stack and final founder lessons. Julian details his backend architecture alongside an async post-recording breakdown of the company's real P&L and expense margins.1:06–4:01 · Guest disagreement 0/10 Julian's Background in Engineering and Early Agency Experience Pat acts as a supportive interviewer asking standard exploratory founder questions. Julian explains his background transitioning from an unwanted agency setup to identifying product flaws in market incumbent Fitbod.4:01–6:51 · Guest disagreement 0/10 Developing the Initial MVP and Custom Workout Logic Pat prompts Julian on his early development timeline and initial traction. Julian details the algorithmic complexity of custom workout logic and using Reddit for initial validation before an ad break.6:52–9:27 · Guest disagreement 0/10 Scaling Gravl with Paid Ads and International Expansion Pat gently challenges whether Reddit alone drove scale, prompting Julian to break down his paid ad playbook. Julian shares tactical advice on geographic ad arbitrage and creative iteration via Meta ad libraries.9:27–12:09 · Guest disagreement 0/10 Full Gravl App Walkthrough and Core Feature Demonstration Pat invites Julian to do a full product walkthrough and remains silent throughout the demonstration. Julian explains the onboarding funnel, hard paywall mechanics, and dynamic exercise reordering.12:10–14:55 · Guest disagreement 0/10 Gravl's Technical Architecture, Backend Stack, and AI Tooling Pat asks about the technical stack and final founder lessons. Julian details his backend architecture alongside an async post-recording breakdown of the company's real P&L and expense margins.1:06–4:01 · Pat pushing back 0/10 Julian's Background in Engineering and Early Agency Experience Pat acts as a supportive interviewer asking standard exploratory founder questions. Julian explains his background transitioning from an unwanted agency setup to identifying product flaws in market incumbent Fitbod.4:01–6:51 · Pat pushing back 0/10 Developing the Initial MVP and Custom Workout Logic Pat prompts Julian on his early development timeline and initial traction. Julian details the algorithmic complexity of custom workout logic and using Reddit for initial validation before an ad break.6:52–9:27 · Pat pushing back 1/10 Scaling Gravl with Paid Ads and International Expansion Pat gently challenges whether Reddit alone drove scale, prompting Julian to break down his paid ad playbook. Julian shares tactical advice on geographic ad arbitrage and creative iteration via Meta ad libraries.9:27–12:09 · Pat pushing back 0/10 Full Gravl App Walkthrough and Core Feature Demonstration Pat invites Julian to do a full product walkthrough and remains silent throughout the demonstration. Julian explains the onboarding funnel, hard paywall mechanics, and dynamic exercise reordering.12:10–14:55 · Pat pushing back 0/10 Gravl's Technical Architecture, Backend Stack, and AI Tooling Pat asks about the technical stack and final founder lessons. Julian details his backend architecture alongside an async post-recording breakdown of the company's real P&L and expense margins.

speaking balance: gold is Pat, purple is the guest (3 minute bins)

0:00 · Pat 44.5% · guest 55.5%0:00 · Pat 44.5% · guest 55.5%3:00 · Pat 13% · guest 87%3:00 · Pat 13% · guest 87%6:00 · Pat 44.6% · guest 55.4%6:00 · Pat 44.6% · guest 55.4%9:00 · Pat 4.8% · guest 95.2%9:00 · Pat 4.8% · guest 95.2%12:00 · Pat 10.9% · guest 89.1%12:00 · Pat 10.9% · guest 89.1%15:00 · Pat 100% · guest 0%15:00 · Pat 100% · guest 0%
Sharpest disagreement ▶ 3:25 Critiquing incumbent Fitbod's workout generation

Julian directly criticizes the market-leading workout app Fitbod, explaining that its generated workouts were weird and physically dangerous.

Hardest push from Pat ▶ 6:52 Challenging the Reddit-only growth narrative

Pat pushes back on Julian's initial launch story, pointing out that organic Reddit posts alone could not scale an app to $400k per month.

Biggest teaching moment ▶ 7:53 Masterclass on geographic ad arbitrage

Julian educates the audience on why bootstrapped founders should translate apps into Spanish and run cheap LATAM ad campaigns rather than burning cash in competitive US markets.

Pat holds their own ▶ 2:43 Reframing agency model vs scalable app economics

Pat succinctly pinpoints the structural scalability limitations of service agencies to redirect Julian toward explaining his product transition.

the scores for every segment, with the reasoning behind each
ChapterTopicPat as informed peerGuest teachingGuest disagreementPat pushing backWhy
Julian's Background in Engineering and Early Agency Experience 2200 Pat acts as a supportive interviewer asking standard exploratory founder questions. Julian explains his background transitioning from an unwanted agency setup to identifying product flaws in market incumbent Fitbod.
Developing the Initial MVP and Custom Workout Logic 1300 Pat prompts Julian on his early development timeline and initial traction. Julian details the algorithmic complexity of custom workout logic and using Reddit for initial validation before an ad break.
Scaling Gravl with Paid Ads and International Expansion 1401 Pat gently challenges whether Reddit alone drove scale, prompting Julian to break down his paid ad playbook. Julian shares tactical advice on geographic ad arbitrage and creative iteration via Meta ad libraries.
Full Gravl App Walkthrough and Core Feature Demonstration 0300 Pat invites Julian to do a full product walkthrough and remains silent throughout the demonstration. Julian explains the onboarding funnel, hard paywall mechanics, and dynamic exercise reordering.
Gravl's Technical Architecture, Backend Stack, and AI Tooling 1300 Pat asks about the technical stack and final founder lessons. Julian details his backend architecture alongside an async post-recording breakdown of the company's real P&L and expense margins.

Statements from this episode (16)

Assertion Not checkable as stated
Gravl reached 70,000 subscribers and generated $440,000 in monthly revenue
“We launched around two years ago, and now we have over 70,000 subscribers. Last month we made over 440 K.”
Julian Gargicevich Nov 13, 2025 ▶ 1:21
Opinion
Gargicevich: Fitbod's generated workouts were weird, bad, and sometimes dangerous
“I noticed that the workouts were weird and even a bit dangerous sometimes. And like the more I used it, I was like, wow, okay, these workouts are actually bad.”
Julian Gargicevich Nov 13, 2025 ▶ 3:36
Disclosure
Gargicevich: Building Gravl's initial MVP took two to three months
“Initial MVP took around two to three months, maybe.”
Julian Gargicevich Nov 13, 2025 ▶ 4:05
Assertion Not checkable as stated
Gravl's launch post on Reddit generated over 300,000 impressions
“I posted this thread on how I build gravel. Back in the day it was called Gains.ai, and I shared kind of the technical specs around it. It got over a couple hundred likes within the first couple hours and over 300,000 impressions. Yeah, we got our first couple…”
Julian Gargicevich Nov 13, 2025 ▶ 4:56
Disclosure
Gravl scaled via paid ads in South America under $50 daily
“For us, the bigger distribution was paid ads. We started running ads as soon as we obviously added the subscription model. I still remember like running our first app, we got a subscription within the 1:10 minutes of like activating the ads. Yeah, that was the…”
Julian Gargicevich Nov 13, 2025 ▶ 7:04
Insight
Gargicevich: Most top-performing mobile app ads are UGC
“You'll see that most of the ads that work out there are UGC content.”
Julian Gargicevich Nov 13, 2025 ▶ 8:52
Insight
Gargicevich: Founders should copy competitors' proven ads via Meta Ad Library
“Good thing about meta ads library is that it's all public. You can go into any of your competitions dashboard and see, you know, which ads are working for them, where they're putting more money. And yeah, there's no secret sauce there. Just copy what works for…”
Julian Gargicevich Nov 13, 2025 ▶ 9:06
Insight
Labeling a product as an AI app directly drives consumer sales
“Make sure that you are showing the users that this is an AI app because that sells now.”
Julian Gargicevich Nov 13, 2025 ▶ 10:12
Insight
Many users will pay at a hard paywall before seeing the product
“Hard paywall before you sign in for the first time. It's like how many users Are actually going to pay before they actually see their product? And the answer is a lot.”
Julian Gargicevich Nov 13, 2025 ▶ 10:46
Insight
Real-human 24/7 support significantly accelerated Gravl's user growth
“And one of the important things that we grew a lot with this, and that was our support. So we have a twenty-four-seven support chat inside the app, and that includes everything from articles to messages, and we'll have someone reply. An actual person, not an A…”
Julian Gargicevich Nov 13, 2025 ▶ 11:49
Disclosure
Gargicevich: Gravl runs on React Native, Expo, and .NET
“We use React Native and Expo. For the backend, I did .NET for most of the core functionality. I've also used Next and React to build some, like, internal admin dashboards.”
Julian Gargicevich Nov 13, 2025 ▶ 12:14
Insight
Gargicevich: Restrict AI coding tools to specific files
“In terms of AI, so I try to stay kind of niche with what I use, cursor a lot, but I'm very specific with files to touch and don't let the AI just go rogue, basically, yeah.”
Julian Gargicevich Nov 13, 2025 ▶ 12:25
Disclosure
Gravl spends approximately one-third of its revenue on paid ads
“Our expense number one is Meta and TikTok, and then maybe a little bit of Google, maybe a little bit of Apple search. I'd say a third is somehow accurate. A third of our revenue that is. This doesn't include making the ads, but just the cost of running them.”
Julian Gargicevich Nov 13, 2025 ▶ 12:52
Disclosure
Gravl spends $50,000 to $80,000 monthly on payroll for 14 people
“Expense number two is salaries. It was mainly just the three of us for our first year. I did the development, and then Matthias and Aaron did the growth and ads. Since then, we've grown the team to probably maybe 13 to 14 people, give or take. There's some, li…”
Julian Gargicevich Nov 13, 2025 ▶ 13:07
Disclosure
Gravl spends just $1,000 monthly on infrastructure despite generating $440K
“And then in terms of infrastructure for running the actual app, like servers and AI builds and other tools, I'd say around a grand a month as well.”
Julian Gargicevich Nov 13, 2025 ▶ 13:37
What-if
Gargicevich: We dragged our previous startup an extra year
“In our case was our previous startup. We probably dragged that for an extra year. We should have probably call it days like way earlier.”
Julian Gargicevich Nov 13, 2025 ▶ 14:41
Made with StarZero

Turn any episode into a week of clips.

This entire site, about 80 episodes transcribed, diarized, checked and made playable, runs on the StarZero media pipeline. Drop in your own episode and the podcast clipper finds the moments worth sharing, cuts them, captions them, and reframes them for every feed.