Why Your App Keeps Only 50% of Its Revenue: Reality vs. Expectation for Founder Earnings

What we talked about.

In this episode of the Sunny Ray Show, host Sunny Ray talks with Arsh and Jeff, co-founders of Addicting Elements, about why app founders often keep only half of their revenue. The pair built House AI, an AI interior design app that reached over 150,000 users across 175 countries in its first few months, and discovered that fees from Apple, attribution tools like AppsFlyer and Branch, and analytics services like Sentry and PostHog can consume 50 to 60 percent of a founder's earnings. That realization led them to build Victura, an in-house platform that consolidates payment flows, attribution, and logging into one system, cutting costs down to roughly 1 percent instead of 15 percent or more. Arsh, who moved from India to the US two years ago, shares how he raised a 100k biotech grant in high school, while Jeff describes his early background in derivatives and quant finance. Together they discuss hackathon culture, the speed AI now allows for shipping products, and their vision for a fairer, more decentralized mobile app economy.

Two young founders explain how third party tools quietly eat half of app revenue, and how Victura claws it back.

The questions, and the answers.

What are you building at Addicting Elements, and why should people care?

We started building mobile apps in December last year. Our first app, House AI, scaled to over 150,000 users in a few months. While building it we realized app founders don't know that Apple, attribution tools, and other third party services take a huge cut, leaving them with only 40 to 50 percent of their revenue. That's why we built Victura, and we're now onboarding other founders.

Which specific tools are you replacing or combining into Victura?

We target payment flows first. Instead of paying Apple's 30 percent cut, we built a web to mobile flow so users pay on the web and we only pay Stripe one or two percent. Second is attribution and MMPs, tools like AppsFlyer or Branch charge per conversion, which piles up to thousands of dollars. We also fold in error logging tools like Sentry and PostHog into one centralized platform.

Tell me about House AI. What's it about and why should people care?

House AI is an interior design app Jeff and I started in December last year. We're live in 175 countries and 40-plus languages, with over 150,000 users in the first couple months and around 20,000 daily users. Building it taught us how the app market works, which let us create our own internal tools that eventually became Victura.

You raised grant money for a biotech idea while still in high school. What did that teach you?

I raised about 100k for a biotech project after my grandma developed diabetic retinopathy. I built a machine learning tool to help diagnose the disease at early stages, did research with a professor at Harvard, and published the paper. I ended up deploying the tool to about nine clinics and saw strong accuracy results from the algorithm.

You moved from India to the US two years ago. What's genuinely different about building here?

The biggest difference is the environment, especially hackathons and the community around them. You meet people who share your vision and can accelerate what you're building. Jeff and I actually met at a hackathon in Florida, then went to others like Cal Hacks and Tree Hacks at Stanford. Finding the right people in your niche makes all the difference.

Third party SDKs take 40 to 50 percent of app revenue. How did that become normal, and why has nobody fixed it until now?

There's very little information out there about app building. When Arsh and I started House AI we dug through YouTube and Reddit and found almost nothing, so maybe it's gatekept. The reality is you can reduce those costs quickly if you know how to use the right tools, and that's what Victura is here to do for every founder.

You built Victura entirely in house instead of wrapping existing tools. What did that cost you and why was it worth it?

Today we have AI to leverage, so work that took a ten person team ten years ago can now be done by one or two people. Everything is on the cloud, so you can deploy and fix bugs with one line of code. Shipping speed is maybe ten to a hundred times faster now, so honestly this is the best time to build anything.

If Victura wins, what does the mobile app economy look like that it doesn't today?

If more people use Victura there will be less spread going to competitors and more growth toward a decentralized network of founders. That creates a new economy that lets more people come in and build new apps and inventions in the digital space, whether they're young builders or people still working a nine to five job.