Hey {{ subscriber.first_name | strip | default: "there" }},,

Welcome to month 3 of my app portfolio journey. I'm sharing everything - the wins, the experiments, and the lessons learned as we build toward 10 profitable apps.

TL;DR

This month shows some early signs of growth:

• $123 revenue (+712% 🚀)
• 506 downloads (+52%)
• 6/10 apps launched (goal: 10 apps by year-end)

1. Launched Baby Kick Counter 👶

We added our 6th app to the portfolio! Lukas built this in just 2 weeks following our internal playbook and starter project.

The Game-Changer: AI-generated UI. We're not designers, so discovering UXPilot transformed our development process. Lukas experimented with it, and the results were fantastic - professional-looking interfaces without the design struggle. This is now part of our standard playbook.

Results:

• 129 downloads (100% organic)
• $49 in sales
• Zero marketing spend

2. Improve app screenshots 🎨

The screenshots for our AI Girlfriend app were outdated and, frankly, terrible. Since screenshots are crucial for conversion (they're what users see before deciding to download), we completely redesigned them.

Key Learnings:

• Show benefits, not features
• Use keywords in the screenshots - they are being indexed as well

Before:

After:

Impact:

• Conversion rate doubled
• Downloads doubled

3. Monetization Experiments 💰

We ran several tests to optimize revenue:

Paywall Timing: Now showing the paywall every time users open the app (previously was after X uses)

Price Testing: Started A/B testing different price points for AI Girlfriend. The surprising finding? Cheaper packages significantly outperformed our original pricing. While we need more data for statistical significance, the early difference is substantial. We'll continue running this test for more conclusive results.

• Control (red line): $4.99 Weekly, $29.99 Annual
• Test (blue line): $2.95 Weekly, $19.99 Annual

So far, the cheaper version has increased the LTV (lifetime value) of the customer by 371%.

4. Embracing AI Development Tools 🤖

I challenged myself live on YouTube to build an app in 1 hour using Claude Code. The result? A working fitness tracker prototype with core features is functional.

Reality Check: While impressive for prototyping, I haven't mastered production-ready apps through prompting alone. Recently tried building a new app entirely with AI - ended up scrapping it and coding from scratch. The issue might've been my unclear UX vision, leading to messy outputs.

What Actually Works: Using AI for specific components:

• Individual screens
• Reusable components
• Specific features

I'm curious - how are AI tools working in your development workflow?

October Plans 🎯

1. Launch 2 more apps (getting closer to our 10-app goal!)
2. Maybe document one app's journey live? Would you be interested in watching the entire process from idea to launch?
3. Kickstart our notJust.Incubator with 10 amazing founders that will build and launch their own apps

The Bottom Line: Month 3 showed us that small optimizations (screenshots, pricing, paywalls) can have massive impacts. We're not just throwing apps at the wall - we're learning from experimenting and sharing this with you as well. Because why not?

What would you like to see in next month's report? Hit reply and let me know!

Building in public, one app at a time.

🔁 In case you missed it

The BEST ASO strategy for your mobile apps In this episode, we sit down with indie developer Wilmer Terrero, who has grown a portfolio of 14+ mobile apps using ASO (App Store Optimization) while working full-time as a software engineer. Watch now

5 steps to Launch Your First Mobile App Fast In this video, we'll share the exact 5-step system we use at notJust.dev to go from app idea to App Store in as little as 3 days. We’ve already launched 6 apps this year, and we’re on track for 10. Watch now

Did you learn something new today?

If you found this email valuable, forward it to one friend or coworker who can also benefit from it. That would be much appreciated 🙏

Vadim Savin Helping you become a better developer together with the notJust.dev team