How I built an app that makes $2,000 in one month (from scratch)

Arthur sets out to build an app generating $2,000 in monthly profits. He candidly shares his past startup failure and the lessons learned, emphasizing the importance of building and iterating without funding or a perfect idea. The video promises a no-filter look at the process, from finding an idea to achieving profitability.

Seeking a proven concept, Arthur researches top-grossing apps and identifies affirmation apps as a simple yet effective model. He decides to target users in Nordic countries suffering from seasonal depression, creating a 'cozy and warm' UI. He outlines the app's core features, focusing on a minimalist design, and uses Figma for prototyping before coding. A candle mascot named Glow is created to embody the app's theme.

Arthur begins coding using Expo, focusing on the basic structure and local notifications. He highlights the importance of the initial onboarding screen for user retention and designs a welcoming animated mascot. The initial onboarding flow is built, including user data collection for analytics and a special offer screen, similar to existing successful apps.

Despite Expo's limitations, Arthur successfully implements iOS widgets in Swift to display daily affirmations. He then tackles in-app subscriptions, explaining the Apple small business program and using RevenueCat for managing payments. He creates monthly and yearly subscription options with a free trial, configuring different prices for Nordic countries, and designs a paywall template.

Arthur completes all app screens, including the home screen with swipe features, categories, a streak tracker, and a feedback section. He develops a landing page and legal policies using Next.js and ChatGPT for App Store submission. The app is submitted via Expo's EAS tools, and Arthur travels to Bangkok while awaiting Apple's review.

The app is initially rejected due to a mismatch in privacy settings on App Store Connect. Arthur quickly rectifies the issue by unchecking tracking boxes and resubmits the app. He notes that such rejections are common and contribute to maintaining app quality.

After 5 days, the app is finally approved. Arthur focuses on App Store Optimization (ASO) using Astro to find relevant keywords for Nordic countries. He implements a launch strategy offering free lifetime access to early users to gain downloads and reviews. He leverages his YouTube community and creates short-form content to drive initial traffic.

Initial downloads reach nearly 900, with traffic coming from various sources like WeChat and Reddit. YouTube shorts, despite high views, result in low conversion due to a developer-centric audience. RevenueCat shows over 600 free trial claims, but app store reviews are mixed. Arthur notes low retention rates and decides to refine keywords and improve the onboarding experience by adding clarifying screens for notifications.

Arthur shifts to a more aggressive organic content strategy on TikTok, creating POV, product demo, and transformation videos to increase visibility. He also prepares for paid ads by integrating the TikTok events SDK, a challenging process for Expo, and adds a highly requested dark mode feature. He outlines the setup of TikTok ad campaigns, focusing on app installs for iOS users in Nordic countries with a limited budget.

After testing Tik Tok ads, the results show a very low trial-to-paid conversion rate. Arthur attributes this to users not seeing the app's value. He implements a new onboarding with more personalized questions, a 'for you' section, and a tutorial to better explain features. He also creates an analytics dashboard to track user drop-off points during onboarding.

Arthur considers translating ads and the app into native languages for Nordic countries to potentially increase conversions. He also creates a Discord community to share resources, offer support, and foster collaboration among app developers, emphasizing learning by doing and enjoying the process rather than solely focusing on results.

A significant increase in conversions is observed after improving the onboarding process, leading to $160 in revenue from a €30 ad spend. Arthur receives free ad credits from TikTok, boosting his effective ad budget. He identifies a further opportunity to improve onboarding completion and trial start rates. He researches commitment psychology and adds a progress bar to the onboarding flow.

Arthur uses PostHog for A/B testing different onboarding variants, finding that a 'no-pack' version (without a specific screen) yields the best completion and conversion rates. Revenue reaches over $800 in one month. He explains his method of self-financing ad spend due to Apple's payment delays and plans to continue A/B testing paywall designs and prices. Daily downloads and trial conversions show significant improvement, with up to 20.6% of users starting a trial.

Arthur develops a new 'practice' feature using a text-to-speech model for audio affirmations. He shares a crucial tip: enabling the 'billing grace period' in App Store Connect helps recover lost subscriptions. His app rankings in Astro continue to improve organically. He receives recognition from the Expo team, including a blog post feature. Apple Ads, despite free credits, prove ineffective in driving downloads.

With 39 active trials and 61 subscriptions, monthly revenue reaches $2,400. Arthur analyzes his TikTok ad performance, noting that simple rhythmic videos with default iOS wallpapers and user-generated content are most effective, achieving costs as low as 10 cents per download. He discovers new TikTok ad coupons, bringing his total free credits to approximately $475. He plans to create more structured content for the community and intends to build a new app with community involvement.

The app surpasses the $2,000 profit threshold, reaching $2,800 in revenue with solid profit margins due to free TikTok ad credits. Arthur highlights the importance of optimizing onboarding, which accounts for over 80% of conversions, and utilizing Apple's billing grace period. He shares detailed conversion rates: over 14% download to trial and 31% trial to paid, resulting in $1.64 profit per download. He strongly encourages viewers to start building their projects, emphasizing learning through doing and embracing failures as part of the enjoyable process.