I have been making games for around 6 years now.

Across mobile, web and other platforms, I have probably built 120+ prototypes.

The funny part?

The first 100 barely made me any money.

The next 20 or so got much closer to 6 figures in revenue.

Some examples:

700,000+ ratings!

1. Watermelon/Suika variation — 65M+ players

GOTY Winner!

1. Draw To Smash

2. Element Merge variation

3. Cozy Rope Puzzle

4. Drop Animals

5. …and many more

My ability to make games did not suddenly become 10× better.

What changed was that I finally understood distribution.

Instead of publishing a game somewhere and hoping people magically found it, I started working with web-game platforms that already had millions of players.

Platforms like CrazyGames get millions of players every month

Think CrazyGames, Coolmath Games, GameDistribution, Poki, YouTube Playables and similar portals.

They already have the audience. Millions of players visit these websites every month.

Your job is to build something their audience wants to play.

Building the game is the easy part now

Coming up with game ideas does not always come naturally to me.

Usually, I start by scouting mobile stores and web-game portals to see what people are already playing.

I look at games with lots of ratings, play a few of them and try to understand:

• What is the core mechanic?

• Why does it feel satisfying?

• How quickly can you understand the game?

• What keeps you playing?

• Where could the idea be changed or improved?

The goal is not to clone the game.

I usually take the core idea, change the theme, add or remove mechanics and put my own spin on it.

Most of my games are currently 2D puzzle, casual, arcade or high-score games because they are small enough to build and test quickly.

Although lately, I have also been moving more into 3D games.

Racing and shooting games can work well on web too.

For development, I use AI heavily.

I vibe-code games in Unity or HTML/JavaScript using tools like AgentsGuild, Claude Code and Codex.

AgentsGuild allows me to export builds for various web-platforms with simple commands like: /youtube, /poki, /crazygames etc.

My workflow is basically:

idea → prototype → test → feedback → iterate → ship

AI now handles a huge amount of the coding.

I give it references, explain mechanics, test what it produces and keep correcting it until the game feels right.

My own games website!

You can see some of the games I build on anul.space.

For web specifically, engines like Phaser, Three.js, PlayCanvas or Cocos can be easier to work with.

I still use Unity because I know it well, although Unity web builds usually require more optimization.

Distribution is the actual business

This is the part I misunderstood for years.

Making a good game does not automatically mean people will play it.

You need distribution.

Web platforms solve that problem because they already have millions of users actively looking for games.

Once I have a playable prototype, I start approaching platforms.

Most portals have a developer or game-submission page where you can send them your game.

Usually, I host a playable version on my games website and submit a link.

The platform then evaluates things like:

• Is the game fun?

• Does it fit their audience?

• Does it feel different enough?

• Will people play it for long enough?

• Does it work properly on desktop and mobile?

• Can advertising fit naturally into the experience?

Platforms have also become much more selective.

Process overview

Sometimes you submit a game and hear nothing.

Sometimes they reject it.

Sometimes they send you a long list of changes.

That is normal.

If I do not hear back, I may submit again later or contact someone from the company through LinkedIn or X.

Once a platform is interested, there is usually an iteration process.

They might ask you to change the onboarding, improve the controls, fix mobile responsiveness, adjust difficulty or rethink where ads appear.

After that comes QA.

You fix bugs, meet their technical requirements and keep iterating until the game is ready.

Once approved, the game goes live in front of an audience you would otherwise have to spend months or thousands of dollars building yourself.

That is the leverage.

Engagement = more distribution = more money

Web games usually make money through advertising.

There are ads around the game, and there can also be ads inside the game.

One format I particularly like is rewarded ads.

Instead of forcing someone to watch an ad, you give them something useful in exchange.

Rewarded Ads Examples

For example:

• Extra life

• Coins

• Power-ups

• Continue after losing

• Faster progression

• Bonus rewards

The important part is designing the game around engagement, not just ads.

Another game revenue from CrazyGames

If somebody plays your game for 30 seconds, there are very few opportunities to monetize them.

Engagement is key

If they play for 20 minutes, return tomorrow and keep progressing, that player becomes much more valuable.

And platforms care about the same thing.

If players click your game and immediately leave, the platform has little reason to keep promoting it.

If players stay, replay and engage, the platform has more reason to give the game visibility.

So there is a useful loop:

better game → longer sessions → more monetization → more platform promotion → more players

This is why I think about monetization while designing the game rather than adding ads at the very end.

Launching the game is only the beginning

Once the game goes live, I start looking at data.

Most platforms give developers dashboards where you can track things like players, sessions and revenue.

I also like adding my own analytics.

For example, if a game has 20 levels, I want to know:

How many players reach level 2?

How many reach level 5?

Where do most people quit?

Are people using rewarded ads?

Are they replaying levels?

This makes improving the game much less subjective.

Instead of saying:

“I think level 7 feels difficult.”

You can see that 35% of players are quitting on level 7.

Now you have something to fix.

You can also keep updating games after launch.

New levels, seasonal content, Halloween themes, Christmas events or small gameplay improvements can give an older game another push.

And every game teaches you something for the next one.

That is probably the biggest advantage I have after 120+ prototypes.

I have a much better idea of what to build, what not to build and what platforms are actually looking for.

The real advantage is the relationship

One thing that became increasingly valuable over time was building relationships with publishers and platforms.

Once you have shipped multiple games with them, you can start asking:

What genres are working right now?

What mechanics are players responding to?

What kind of games are you looking for?

They have access to far more player data than an individual developer does.

So instead of spending 2 months building something and then trying to convince a publisher to take it, you can sometimes get feedback on the idea before you even start.

That dramatically reduces the risk.

And this is ultimately how I think about web games today.

AI has made building much cheaper and faster.

But AI is not the biggest advantage.

Distribution is.

You can build the best game in the world, but if only 100 people ever see it, there is a limit to what it can become.

Build something fun, get it in front of platforms that already have millions of players, study the data, improve it and repeat.

That simple loop took me from my first 100 prototypes making almost nothing to games that have now reached tens of millions of players.

Reach out to me

My LinkedIn: https://www.linkedin.com/in/anulagarwal/

My X: https://x.com/anulagarwal

My YouTube: https://www.youtube.com/@Anoool1/videos

Back in June 2025, my apps were making around $50/month only.

Since then, I’ve experimented a lot, launched more apps, and slowly grown my portfolio.

I’m a solo developer — I build apps and games, and create content around the process.

RevenueCat Analytics Dashboard

Last month, I made ~$1,600/mo from my mobile apps. Some stats:

• 100% organic revenue

• 100% built with AI

• ~200 downloads/day across IOS & Android

• Almost equal split between both for revenue

• 80% of the revenue comes from 20% of the apps

• Built 40+ apps, only kept ~20 live

• Almost 100 paid subscribers

Reaching 100 paid subscribers soon

Some of my apps

The best part is you just build once and publish it. You don’t need to spend a lot of time managing it or constantly adding new features.

Occasional customer support is needed but again I use AI to solve those, for example this customer faced an issue in my app:

Customer Support Queries

And I was able to solve in less than 30 mins with AI. He was shocked and surprised at the speed and my care for the support.

I will cover in this article how all of this came to be: the development & distribution of mobile apps.

What is a mobile app?

Example of mobile apps

Anything that solves a problem and can be used on your phone can be an app.

I’ve always believed people will pay when an app solves a real problem and provides enough value.

It can be something simple like a timer, calculator, or journal, or something much bigger like a social media platform.

You publish these apps on the App Store or Google Play, where users discover and download them. They can make money through subscriptions, premium features, or ads.

I mainly focus on smaller, simpler apps rather than trying to compete with the apps at the top of the charts.

One real app example

One of my old apps

For example, I built this app to help pregnant mothers count the baby kicks and track over time how fast/slow it becomes. This app makes me ~$100/mo.

How to develop a mobile app?

Traditionally, building an app meant learning programming languages, writing code, and testing everything on a real device.

For iOS, that usually means Xcode + Swift. For iOS and Android together, you can use cross-platform frameworks like Flutter or React Native.

A simple step by step approach

But with AI, do you really need to know these languages deeply?

Not necessarily.

I built 40+ mobile apps without knowing Swift or Flutter beforehand. I do come from a coding background though, so understanding logic, product and design helps me guide AI tools and judge whether the output is actually good.

You may not need to write all the code yourself anymore, but knowing how software works still makes a big difference.

What “100% built with AI” means

AI writes almost all the code, but I still decide what to build, guide the product, test every flow, fix the design and make the monetization decisions. AI handles implementation, but judgment still comes from you.

Using Opus 5.5 for my apps

I use Codex & Claude (in terminal) to develop these apps. I have setup my own App Template in Xcode/Swift that allows me to reuse features easily, such as:

• Attractive onboarding screens

• Paywall with links

• AI APIs (for image scanners, image generators, chatbots) just plugin your AI API key

• In-App purchases/subscription management with RevenueCat

• Firebase to track analytics

• Notifications

• Settings/Contact

It can save you upto 1–2 hours of the initial development time in every new project.

So it allows me to reuse the features without spending time to develop them again and again.

You can get it here: https://anulagarwal.gumroad.com/l/cukmwo

Using AI to develop an app

Agentic tools like Claude Code and Codex are now very capable, but they still need good direction.

I usually explain the feature, share references or screenshots, let the AI build it, test the result, and keep giving feedback until it feels right.

Think of it like sculpting clay: prompt, test, refine, repeat.

AI can still break existing screens, overcomplicate simple features, or miss edge cases. I reduce this by working in small steps, testing on real devices, and using Git before major changes.

How to distribute/market a mobile app?

There are a few main ways to market an app:

• ASO

• Paid ads

• Influencers

• Reddit, X and LinkedIn

• Instagram, TikTok and YouTube

I mainly focused on ASO, which is basically SEO for the App Store and Google Play. You find what users are searching for, target those keywords, and improve your title, icon and screenshots to increase downloads.

For ASO, I recommend watching Adam Lyttle, who talks a lot about growing iOS apps through search.

I’m also starting to explore paid ads more. I previously ran Google Ads profitably for my mobile game, Squishy Cats.

The biggest lesson here is simple: distribution matters more than almost anything else.

Without traffic, you have no users and no data. Once people start using your app, you can improve the product, pricing and monetization.

For ASO specifically, your title and screenshots matter a lot. Your store page is what sells the download before users ever open the app.

How to monetize your apps?

Ask users to pay $$ in the form of subscriptions or one-time payments.

I have a soft paywall in my apps — which means, users can explore the app for free but to use the core feature they need to subscribe to it. And they do!

Paywall from one of my other apps

You need to experiment with pricing and different offers, but I start off with this:

• Weekly: $4.99/week

• Lifetime: $24.99

You can adjust numbers and see how it works out for you. In some apps I have a higher price, but close to this range.

What kind of apps to choose?

I did a lot of ASO research to find app ideas with decent search demand and relatively low competition.

I used Astro for this. They also have an MCP server you can connect to your AI agent to help with keyword and market research.

If your main marketing channel is short-form video, the approach is slightly different. Your app needs a strong hook and should be understandable in 1 sentence.

If viewers immediately understand what the app does and why it is useful, converting them becomes much easier.

I am now generating motion graphics videos using Opus 5.5 and automating publishing using my web-tool.

Final thoughts

After building 40+ apps, I’ve realized development is no longer the hardest part. AI makes building much faster, but it does not guarantee success.

You still need to solve a real problem, give users a reason to pay, and find a reliable way to reach them. Many of my apps didn’t work, but every launch taught me more about keywords, screenshots, pricing, and paywalls.

Start small, validate demand, build the simplest version, and publish. A few useful apps can slowly grow into meaningful income.

I’m currently at around $1,600/month.

Next goal: $5,000/month.

I’ll post again when I get there.

Reach out to me

My LinkedIn: https://www.linkedin.com/in/anulagarwal/

My X: https://x.com/anulagarwal

My YouTube: https://www.youtube.com/@Anoool1/videos

How I Build iPhone Games Using AI — and Make Money

A game I built using AI reached 65 million+ players and made $25,000+ in revenue.

It was not a massive production. I did not have a large engineering team, a studio full of artists, or millions of dollars to spend on user acquisition.

The game generated over $25,000+ in revenue for me as a solo-developer.

You can view my detailed article on the whole process here:

https://medium.com/@anulagarwal12/my-ai-generated-game-earned-25-000-here-is-how-it-happened-ea5612d91464

I had a simple game idea, an AI-assisted development workflow, and a distribution opportunity that helped the game reach far more players than I expected.

That game was originally built for the web, not the iPhone. But the biggest lessons had very little to do with the platform.

The things that mattered were:

1. Choosing an idea I could actually finish

2. Building and testing it quickly

3. Adding a sensible way to make money

4. Finding places where people could discover it

These are the same 4 things I now think about when building mobile games.

AI has made the first 2 steps dramatically easier. A solo developer can now use AI to write code, debug errors, build menus, integrate SDKs, generate levels, create store assets, and even plan marketing content.

But AI has not solved everything.

It will not automatically choose a good idea. It will not tell you whether your game feels satisfying. And it definitely will not bring players just because you successfully published something on the App Store.

So, this is not a guide to typing:

“Build me a successful iPhone game.”

It is the more practical process I would follow to go from an idea to a published game that has a real chance of making money.

1. Start With an Idea You Can Actually Finish

After working on more than 100 game prototypes, the biggest mistake I see is starting too large.

People want to build:

• An open-world game

• A multiplayer survival game

• An RPG with 50 characters

• A simulation with hundreds of systems

AI can make development faster, but it does not make a huge game small.

For your first iPhone game, I would aim for:

• 1 core mechanic

• 1 main input

• 30–90 second sessions

• 10–20 initial levels

• A playable prototype within 7 days

Think of mechanics such as sorting, merging, stacking, drawing paths, dodging obstacles, solving word puzzles, or upgrading a character.

My 50-point idea test

I would score each idea from 1 to 10 across these 5 questions:

1. Can someone understand it in under 3 seconds?

2. Can I build the first version in 7 days?

3. Can the mechanic support at least 20 levels?

4. Will it look interesting in a short video?

5. Can I naturally add ads, content unlocks, or skins?

The maximum score is 50.

I would be careful about starting anything below 35.

The game that reached 65 million+ players was extremely simple. Players could understand it quickly, start instantly, and play in short sessions.

That taught me an important lesson:

A simple game with strong distribution can outperform a more polished game that nobody discovers.

AI is useful here because it can quickly generate and compare ideas.

For example:

Generate 20 simple iPhone game ideas.

Constraints:
- One-thumb controls
- 2D graphics
- Understandable within 3 seconds
- 30 to 90-second sessions
- Prototype possible within 7 days
- Supports rewarded ads and cosmetic items
- Gameplay should look good in vertical videosScore every idea out of 50.

Use AI to expand your options, but make the final decision yourself. It can generate ideas, but it cannot fully judge whether something feels fresh or satisfying.

2. Choose the Right Development Path

You do not need to spend 2 weeks researching every game engine.

The best option is usually the one that gets a playable build onto your phone fastest.

Xcode, Swift and SpriteKit

This is the native route for Apple devices.

I would choose it when:

• The game is only for iPhone and iPad

• It is a simple 2D game

• I need native Apple features

• I want direct control over performance and device behaviour

You can use SwiftUI for menus and SpriteKit for gameplay.

AI is fairly good at writing Swift, especially when you give it the existing project and one small task at a time.

Flutter and Flame

Flutter is useful when you want to release the same game on both iPhone and Android.

It works well for:

• Puzzle games

• Word games

• Card games

• Board games

• Quiz games

• Idle games

Flutter can handle menus, shops, level selection, and statistics, while Flame manages the game loop and gameplay objects.

I would not use it for a visually demanding 3D game, but it is practical for many small 2D games.

JavaScript and Capacitor

This can be the fastest option if you already know web development.

You can build the game using:

• HTML

• CSS

• JavaScript or TypeScript

• Phaser, PixiJS, or another web library

You then package the web project as an iPhone app using Capacitor.

I like this route because you can test the mechanic immediately in a browser. It is especially useful for grid games, merge games, card games, idle games, and simple arcade games.

It also keeps the web open as another distribution channel.

That matters to me because my 65 million-player game was a web game. Making it easy to access played a major role in its growth.

My simple rule

• iPhone-only 2D game: Swift and SpriteKit

• iPhone and Android casual game: Flutter and Flame

• Existing JavaScript game: Capacitor

• Advanced 3D game: Unity or another full game engine

Players do not care which framework you used. They care whether the game loads quickly and feels good.

My AI Development Workflow

I never ask AI to build the complete game in one giant prompt.

That usually creates a lot of code that becomes difficult to debug.

Instead, I work in small stages.

I also use AgentsGuild — A collection of AI Agents + Skills that allows you to build games for any platforms efficiently.

It helps you create high-quality games with AI that do not look like slop at all. You can get it here:

https://agentsguild.co/

Step 1: Write a one-page specification

Define:

• Core mechanic

• Controls

• Win and lose conditions

• Level structure

• Progression

• Required screens

• Monetization

For a small game, you do not need a 50-page design document.

Step 2: Build one playable level

The first version should only contain:

• One character or object

• One level

• One win condition

• One lose condition

• A restart button

No shop, skins, login, or daily rewards.

First answer one question:

Is the core interaction enjoyable for 30 seconds?

If the answer is no, adding more features will not save it.

Step 3: Complete the basic loop

Add:

1. Main menu

2. Gameplay

3. Results screen

4. Retry button

5. Next-level button

6. Save system

At this stage, someone should be able to play without you explaining everything.

Step 4: Add content through data

Once the mechanic works, AI can help generate:

• Levels

• Obstacle combinations

• Difficulty progression

• Enemy variations

• Upgrade values

• Daily challenges

I try to keep levels inside configuration or data files instead of hardcoding everything.

Then I can ask:

Create 10 new levels using the existing format. Increase difficulty gradually and introduce only one new obstacle.

This is much safer than letting AI repeatedly rewrite the main gameplay code.

Step 5: Add monetization and analytics

Before launch, I would track at least:

1. game_started

2. level_started

3. level_completed

4. rewarded_ad_watched

5. purchase_completed

You do not need 200 events.

You need enough information to understand whether players are starting, progressing, returning, and paying.

Step 6: Test on a real iPhone

I test:

• Touch controls

• Different screen sizes

• Audio interruptions

• Weak internet

• Airplane mode

• Ad failures

• Purchases

• App backgrounding

• Older devices

One lesson I have learned is that a feature is not finished when the AI says it is finished.

It is finished when I have tested it.

Also, commit or back up the project after every working feature. AI can fix one problem while accidentally breaking 3 unrelated systems.

3. Choose a Simple Monetization Model

My AI-built game made more than $25,000 from 65 million+ players.

That works out to roughly $0.0004 per player.

The revenue was still meaningful because the game was inexpensive to build and reached a huge audience. But it shows why both distribution and monetization matter.

For a small iPhone game, I would start with one primary monetization method and one secondary method.

Model 1: Ads and Remove Ads

This is the simplest option for many free casual games.

You can use:

• Rewarded ads

• Interstitial ads

• A one-time Remove Ads purchase

Rewarded ads work well when the player receives something useful:

• One extra life

• One extra move

• Double coins

• A hint

• A retry

• A bonus chest

Interstitial ads should appear during natural breaks, not while someone is actively playing.

For example:

• After completing a level

• After losing a run

• After several rounds

• Before returning to the menu

You might price Remove Ads at $1.99, $2.99, or $4.99, depending on the game.

Model 2: Content Lock

This works well for level-based games.

For example:

• First 10 levels are free

• Remaining 40 levels cost $4.99

• The purchase also removes forced ads

The free section should give players enough time to understand and enjoy the game.

Do not show the payment screen before they have experienced something satisfying.

Model 3: Skins and Cosmetic Items

Cosmetics work when players repeatedly see the same character, vehicle, board, or object.

You could sell:

• Character skins

• Board themes

• Trails

• Particle effects

• Victory animations

• Environment packs

For the first version, I would keep it small:

• 1 default skin

• 3 unlockable skins

• 2 premium skins

• 1 bundle

Cosmetics also create content opportunities.

A post such as:

Which of these 3 skins should I add?

can generate engagement while helping you decide what to build.

Avoid subscriptions unless you have recurring value

A small offline game with 20 levels probably does not need a weekly subscription.

Subscriptions make more sense when you provide:

• New weekly levels

• Tournaments

• Regular cosmetic drops

• A large content library

• Live-service features

For most first games, Remove Ads, Unlock All Levels, or a Skin Pack is easier to explain and test.

A Simple Revenue Example

Imagine your game reaches:

• 10,000 monthly players

• 6 sessions per player

• 1 interstitial per session

• 60,000 monthly impressions

At an illustrative $8 eCPM:

60,000 ÷ 1,000 × $8 = $480

Now imagine 1% of players purchase Remove Ads for $2.99:

10,000 × 1% = 100 buyers

100 × $2.99 = $299

That gives approximately:

$480 + $299 = $779 gross monthly revenue

Real numbers can vary heavily by country, retention, ad frequency, store fees, and purchase conversion.

Your first target does not need to be $10,000 per month.

A better progression is:

1. First 100 players

2. First returning player

3. First purchase

4. First $100

5. First profitable month

6. First $1,000 month

The first time someone pays for something you built, the project stops feeling like an experiment and starts feeling like a small business.

4. Distribution Is More Important Than Most Developers Expect

I used to think development was the difficult part.

After releasing many projects, I now think distribution is often harder.

The game that reached 65 million+ players did not grow only because it was fun. It grew because it was easy to access and appeared in places where people were already looking for games.

For an iPhone game, I would focus on 4 channels:

1. App Store optimization

2. Reels and short-form videos

3. Reddit and relevant communities

4. Building in public

App Store Optimization

Your product page should answer 3 questions immediately:

1. What is the game?

2. What does the player do?

3. Why does it look fun?

Your first 3 screenshots are the most important.

I would structure them like this:

Screenshot 1: Main fantasy

• Build the Biggest Army

• Create Your Dream Island

• Escape the Impossible Maze

Screenshot 2: Core mechanic

• Merge Matching Fighters

• Draw a Path to Escape

• Sort Every Color

Screenshot 3: Progression

• Unlock 30 Upgrades

• Discover New Worlds

• Beat 100 Levels

Avoid vague copy such as:

The Ultimate Gaming Experience

It tells the player nothing.

Create 3 icon variations

Your icon is an advertisement.

It should have:

• One main subject

• A clear silhouette

• Strong contrast

• Minimal small details

• No unnecessary text

I would create at least 3 versions before launch.

The design you personally like most may not be the one that gets the most clicks.

Turn 5 Gameplay Clips Into 30 Videos

You do not need 30 completely different recordings.

Use this formula:

5 gameplay moments
× 3 hooks
× 2 editing styles
= 30 videos

Your 5 gameplay moments could be:

1. Completing a difficult level

2. Failing in a funny way

3. Unlocking a rare skin

4. Creating a huge chain reaction

5. Comparing the first prototype with the final game

Your hooks could be:

• “Can anyone beat this level?”

• “I asked AI to build this game.”

• “This bug accidentally made the game better.”

Then create 2 versions:

• Text and music

• Voiceover and commentary

You now have 30 pieces of content from 5 recordings.

Show the game in the first second. Do not spend 5 seconds introducing yourself before anything happens.

AI Is Also the Marketing Story

Building the game with AI gives you another angle for your content.

You can post about:

• Building an iPhone game in 7 days

• Asking AI to create a new feature

• Showing the terrible first prototype

• Fixing an AI-generated bug

• Revealing the first day’s downloads

• Sharing the first $10 in revenue

• Comparing version 1 with version 20

People may discover your game because they are interested in AI, programming, entrepreneurship, or solo development.

They do not necessarily need to be searching for a new mobile game.

Use Reddit for Feedback, Not Only Promotion

A direct “download my game” post will usually be ignored.

A more useful post might be:

I built this one-thumb puzzle game using AI. Can you understand the mechanic within 5 seconds?

Or:

My AI-built game received 500 downloads in its first week. Here is where the players came from.

Numbers, lessons, and honest development stories make the post more interesting than a basic advertisement.

Always read the community rules before sharing links.

Focus on the First 100 Players

You do not need millions of players immediately.

Your first 100 real players can already answer important questions:

• Do they understand the first level?

• Where do they stop playing?

• Do they return the next day?

• Do they watch rewarded ads?

• Does anyone open the shop?

• Which video generates installs?

• Does anyone purchase?

Your first release is an experiment.

You are trying to learn:

1. Do people understand the game?

2. Do they enjoy it enough to continue?

3. Can it make money without damaging the experience?

A Practical 30-Day Plan

AI Makes Building Cheaper, Not Success Automatic

AI has completely changed how I build games.

It can write code, fix bugs, build menus, generate levels, integrate services, and remove a lot of repetitive work.

But it cannot fully decide:

• Which idea is worth building

• Whether the game feels satisfying

• Whether an ad is annoying

• Whether a level is too difficult

• Why someone stopped playing

• Whether your screenshots earn the click

Those are still product decisions.

The process I would follow is:

1. Choose one small idea.

2. Build the mechanic within 7 days.

3. Test it with real players.

4. Add simple monetization.

5. Create 30 pieces of content.

6. Find the first 100 players.

7. Improve the game using real behaviour.

8. Repeat.

My AI-built game reaching 65 million+ players and making $25,000+ showed me what is possible.

It also taught me that code is only one part of the outcome.

The idea matters. Distribution matters. And finishing matters more than endlessly planning.

Build the smallest game you can finish, publish, and learn from.

Then get one stranger to play it, one player to return, and one customer to pay.

That is when it starts becoming a business.

I started building apps in June 2025, almost a year ago.

Back then I was making only $50/mo. Over time, I have experimented a lot and grew my apps portfolio.

I am a solo-developer: I build games, apps and create content around developing them.

Last month, I made ~$1,100/mo from my mobile apps. Some stats:

• 90% organic revenue

• 100% built with AI

• ~100 downloads/day across IOS & Android

• Almost equal split between both for revenue

• 70% of the revenue comes from 20% of the apps

• Built 40+ apps, only kept ~20 live

The best part is you just build once and publish it. You don’t need to spend a lot of time managing it or constantly adding new features.

Occasional customer support is needed but again I use AI to solve those, for example this customer faced an issue in my app:

And I was able to solve in less than 30 mins with AI. He was shocked and surprised at the speed and my care for the support.

I will cover in this article how all of this came to be: the development & distribution of mobile apps.

What is a mobile app?

Anything that solves a problem and can be packaged in your mobile is an app. I had a belief that people will always pay $$ to get their problems solved when they see value in it.

It can be anything simple like a timer, calculator, journal diary, or complex like a social media platform that targets the problem of not being social enough.

You upload these apps on App Store or Google Play Store where millions of people search for new apps to download every month.

In exchange, these apps ask users to subscribe & pay for premium features or show ads.

I focus on small-scale apps, unlike the top charts as seen in the image above.

One real app example

For example, I built this app to help pregnant mothers count the baby kicks and track over time how fast/slow it becomes. This app makes me ~$100/mo.

How to develop a mobile app?

Developing a mobile app is pretty much like developing any kind of software — you need to learn coding lanugages, write code and deploy the app to your device to test it out.

If you are targeting IOS only apps, then Xcode is the software where you write code, and Swift is the language.

If you are targeting both IOS & Android then there are cross-platform libraries like React Native or Flutter.

But today with AI, do you really need to understand these languages?

No. I built 40+ mobile apps with having 0 knowledge of these lanugages, although I come from a coding background — I have design sense and logic knowledge that allows me to talk to AI tools in a certain way that helps me achieve the desired result easily.
General coding and product background helps you supervise AI.

What “100% built with AI” means

AI writes almost all the code, but I still decide what to build, guide the product, test every flow, fix the design and make the monetization decisions. AI handles implementation, but judgment still comes from you.

I use Codex & Claude (in terminal) to develop these apps. I have setup my own App Template in Xcode/Swift that allows me to reuse features easily, such as:

• Attractive onboarding screens

• Paywall with links

• AI APIs (for image scanners, image generators, chatbots) just plugin your AI API key

• In-App purchases/subscription management with RevenueCat

• Firebase to track analytics

• Notifications

• Settings/Contact

It can save you upto 1–2 hours of the initial development time in every new project.

So it allows me to reuse the features without spending time to develop them again and again.

You can get it here: https://anulagarwal.gumroad.com/l/cukmwo

Using AI to develop an app

Agentic AI tools like Claude Code or Codex are very good with their latest models. You need to guide them properly, with references, screenshots, and detailed idea explaination.

It is like sculpting a piece of clay into something valuable — you need to give it prompts, let it work, test the output and give back feedback again.

What works — what does not work? Why it works? What do you actually want?

My AI development loop
I explain the feature, share references or screenshots, let Claude Code or Codex build it, and then test the result. I keep giving specific feedback until the feature works and feels right.

What AI still struggles with
AI can break existing screens, overcomplicate simple features, or miss edge cases in subscriptions and APIs. I avoid this by working in smaller steps, testing on real devices and using Git before major changes.

How to distribute/market a mobile app?

There are multiple ways to do this:

• ASO: App Store Optimization

• Paid Ads: Google/Meta/TikTok Ads

• Influencers: Pay influencers to promote your content

• Text based Social Platforms: Reddit/X/LinkedIn

• Video based Social Platforms: Instagram/TikTok/YouTube

I chose the way of ASO — optimizing my store listings in a way that appears in search results of what users search the most.

ASO is basically SEO for the App Store and Google Play. You research what users are searching for, target those keywords in your title and listing, and improve your icon and screenshots so more people click and download your app.

For ASO, just watch all the videos of Adam Lyttle — who makes $100k/mo from his ASO based IOS apps.

I am now diving deeper into paid ads too, as I previously had success of running Google Ads profitably with my mobile game — Squishy Cats.

In future articles I will dive deeper into other forms of marketing channels.

Now remember — distribution is THE MOST IMPORTANT factor for your app. You need to get traffic to your app otherwise nothing else matters, because once you have enough people using your app — you have data, where you can tweak features, monetization strategies etc.

For ASO, the most important bit is having a great title + screenshots that convert. Your store page listing is the key selling point here — not your app.

How to monetize your apps?

Ask users to pay $$ in the form of subscriptions or one-time payments.

I have a soft paywall in my apps — which means, users can explore the app for free but to use the core feature they need to subscribe to it. And they do!

You need to experiment with pricing and different offers, but I start off with this:

• Weekly: $4.99/week

• Lifetime: $24.99

You can adjust numbers and see how it works out for you. In some apps I have a higher price, but close to this range.

What kind of apps to choose?

I personally did a lot of ASO research to come up with ideas that helped me decide what keywords users search for the most and have low competition.

I used the Astro App for this — they have an MCP server too that you can connect to your AI agent and help with research.

But if your marketing channels are video content via social platforms — your app needs to have a great hook and easily understandable within a sentence. So viewers can convert easily.

Final thoughts

After building 40+ apps, I realized development is no longer the hardest part. AI and reusable systems make it much faster, but they do not guarantee success.

You still need to solve a real problem, give users a reason to pay, and find a reliable way to reach them. Many of my apps did not work, but every launch taught me more about keywords, screenshots, pricing and paywalls.

Start small, validate demand, build the simplest version and publish it. A few useful, low-maintenance apps can slowly grow into meaningful income.

I am currently at around $1,100/month, and my next goal is $5,000/month.

Reach out to me

My LinkedIn: https://www.linkedin.com/in/anulagarwal/

My X: https://x.com/anulagarwal

My YouTube: https://www.youtube.com/@Anoool1/videos

my mobile game, squishy cats, was shipped 1 year ago.

i had the idea of building a game around cute/liquid cats and softbody physics.

i am a solo-dev, i prefer speed and quick validation over everything.

my small guide for solo-devs self-publishing mobile games that helped me reach 100k+ downloads:

1) don’t 1:1 clone trending puzzle games
pixel flow was a recent hit - copy it, run paid ads, and you’ll most likely burn money

why?
those games have big publishers running endless a/b tests and spending hard on UA behind them.
you’re not racing a game, you’re racing money entirely

if you think hey i can be the ‘first’ to clone - so are 100 other Turkish and Vietnamese studios with better infrastructure and systems that allow them to make clone of mobile games in <1 week.

go after fresh ideas inspired from proven ideas/mechanics.
they are less competitive and easier to ‘stand out’.


2) for paid UA - google ads is the easiest place to setup and start with

android to understand if your game has decent metrics or not

>ROAs campaign > install volume

>they also get on a call to help you setup everything in 30 mins

>firebase is used to track

>leverage your best performing organic videos

3) be a content marketer

pre-2020 to get organic installs just Google Store or App Store was enough but today you need to be on socials
>market your game/app on socials - it forces you to see it from different angles & makes you a better storyteller

>hook is the most important bit, it will make or break your traction

>keep analysing video KPIs - where the drop-offs happen, how many people stayed to watch etc.

>and the organic videos that get traction - push them into paid ads.
it drops your CPIs

✅ i stopped organic TikToks because 100% focused on paid-ads now!

✅ i first created organic content on TikTok/Instagram and used best performing videos for paid-ads.

and then I created 100s of more videos in <30 mins using my video-tool:

try it out here: https://script2shorts.app/

it helps you use various hooks and different game clips, stitches them together, adds scripts/voiceover narrations and captions to produce engaging videos in bulk.

3a) also leverage reddit/x/linkedin

if you can make appealing videos for your game - reddit/x/linkedin is a great place to post and see if the video is actually appealing enough and the idea sticks with viewers.

in #3 & #3a

-> your goal is not to validate your actual game but just the concept of you making appealing videos and being a great marketer.

because paid UA is expensive and in early days you need a lot of players to check engagement data, solving for in-expensive marketability in early days is a great success.

4) leverage AI into everything

>all system/tech related tasks like:
remote configs, faster builds, level editors - automate every repeatable task
it saves you a ton of time
AI is the only moat left in dev. it’s what gives you speed
develop fast, iterate with patience

>i am integrating Claude Code with Google Ads now to automate more stuff and bring me daily reports!

5) keep learning
once your game is live and tests are running, you’re mostly waiting.
use that time to upskill and bring it back into your next game (or this one) - play more games, explore ideas, etc.

because of my last post on LinkedIn, plenty of indie-devs DMed me asking on how to self-publish and run paid-ads

^i hope this served as a guide!

I will cover monetization in a 2nd part - follow/stay tuned :D

Follow me on X where I share more about building games/apps:

https://x.com/anulagarwal

in this post, i cover:

-the game itself
-how AI was used (for complete code & 2D art)
-tech stack
-distribution strategy

play the game:

• web:

  https://www.coolmathgames.com/0-suika-watermelon-game

brief intro

i was laid off at the end of 2022. i was working as a game producer in Germany.

racism + internal management issues bought me to this.

i was happy to come back to my country (India) and to work on my own ideas/games.

in 2023, i started building my own games - for mobile & web-browser platforms.

AI tools like ChatGPT (3/3.5) had just been slowly taking over and as a curious dev, I started to experiment with those.

>my theory was web-browser games can be profitable as they provide free distribution in front of millions of players if you have a decent enough game.

more on this theory in my video on how to make passive income with games:

riding a trend

in the middle of 2023, i saw a trending game rising: Watermelon Suika Game.

it is a highscore based game, with MILLIONS of views on YouTube and other socials.

the gameplay was pretty straightforward - you move left-right and drop fruits in a box. same fruits if touched, evolve into a bigger fruit.

art was a constraint for me as I suck at creating art - and this game needed specialized fruits.

i thought okay let’s try ChatGPT (DALL-E) to create images.

and it gave me what i want! i polished a bit in photoshop - removed the sparkles and cleared the background.

now I had 11 fruits all generated with AI.
i used ChatGPT again to write ALL the code for me - in C# as Unity codes in C#.

i had to manually copy-paste code, bring back any bugs or changes I did - unlike today where Claude Code/Codex can do this 100x faster.

mechanics i coded:

• moving the controller to drop fruits

• dropping fruits

• evolution

• win/lose condition

• new game modes

  

i took inspiration from some other variations of the trending game and added some VFX in Unity to make it polished.

the game was completed and i think i did a pretty decent job with development. now time to distribute!

when it hit $10,000 in revenue, i created a video showcasing how I used AI to make this game:

FREE distribution strategy

up until now my belief was only mobile games made money. but I was so wrong.

web game-platforms are websites where you go and play games online: CrazyGames, Coolmath Games, Poki, etc.

and these platforms always require more games from developers on their website.

I distributed my game to various platforms such as:

• GameDistribution

• Coolmath Games

• GamePix

• ....and many more

now there are 2 ways to earn from web-games:

• ad-revenue: you integrate their SDK in your game, call ads, earn revenue

• license-fee: you ‘sell’ your game’s license to these platforms for a one-time upfront fee. and no earnings on future revenue.

some platforms only do license-fee, and others do ad-revenue only.

i did a mix of both with various platforms.

on average for license fee, i was charging $1500-$2500.
and for more changes in the game, there was an additional fees!

now there were various variation of the trending game - but mine was of the highest quality for some reason.

my game:

top #1 on coolmath games mobile (20 million MAU)

• top #3 on coolmath games desktop

• GOTY on coolmath games

• 700,000+ reviews

• 50 million+ players across all platforms

• $20,000+ revenue from web-platforms (license fee + ad-revenue)

mobile version

i had also ported the game to mobile and the engagement data was INSANE.

D1 retention: 42%
D7 retention: 13%
Daily playtime: 50 minutes

proving that the game was indeed a high quality game.

I made some mistakes for mobile - I did not move fast enough. I was testing with a publisher and they did not seem very keen on publishing it - I should have moved fast with another publisher.

I would have added another $40-50k easily if published with a mobile publisher.

but organics got me ~$4,000 revenue (ad-revenue).

vibe-coded games can earn $$$

if you are vibe-coding games, consider reaching out to these platforms to see if they can distribute your games - it is a very lucrative approach as they provide free distribution!

connect with me

do subscribe to my YouTube where I am sharing latest update on my games/apps:

https://www.youtube.com/@Anoool1

my X:

https://x.com/anulagarwal

Click once and post all your game-videos across various social platforms like TikTok, Instagram, YouTube, Pinterest, Threads & more.

This post is sponsored by cross-post. A one-click solution for posting content!

So far, X has paid me ~$500 for 4 weeks of posting. It pays you based on your impressions and engagement - I mostly talk about my games/apps that I am building!

Here is how it all began:

I got scammed (kind of) in Sri Lanka.
My bill was supposed to be 6500 LKR but turnt out to be 9220 LKR. +35% on taxes. I did not like it.

I took a photo of the bill and shared it on X. It went instantly viral:

Gaining me ~500k views. When the tweet started taking off, I was in an Airplane about to leave Sri Lanka - and when I landed, it had exploded.

30k, 40k, 50k - I knew would go viral before posting even because of how controversial it was. And that was the time to:

Milk the tweet.

So I started to comment under it, reply to other tweets, quote tweet it: everything that would help me get the most impresssions out of it.

Quote-Tweet + Replies gave me an additional 30-40% impressions from he original tweet.

With all that being said, I loved Sri Lanka - one of the most relaxing trips ever of mine.

Trend-Riding

I rode a trend:

There was a hype around AI + Video tools and I quote tweeted the announcement of a product launch. Because it was trendy, it got great visibility!

Original thoughts >> Copy posts

Another one of my tweets blew up and I kind of expected it to:

My experience with Apple Search in the latest OS update was terrible so I shared a quick video showing how broken it is, and it went viral.
Turns out plenty of people face the same issue and are able to relate to it.

But did you notice, I also mentioned that Apple is a $4 trillion company - hence enhancing the value of the tweet?

And I had noticed that tweets related to Apple ALWAYS do well.
Again I milked it well.
Original Tweet: 1.8mn impressions
Replies: ~300k impressions

And I don’t even count the replies under other user comments.

Again all of this is just to reach the initial 5mn impressions that is required to be monetized on this platform.

If you think virality is a sum of quantity and luck, then you are wrong.

Checkout this article that explains it pretty well:

Those 1-2 banger tweets will outperform the 100-200 average tweets by a huge delta.

TLDR:

• Share original experiences & thoughts, but share it in a catchy way

• Choose mass-topics instead of niched out ones

• ^Travel does well, Brand related stuff

• Things that people can relate to throughout the globe

• Once you have a potential banger, milk it

• Quote Tweet it, reply under it, reply to others - enhancing your impressions.

• Ride the trends - see what is popping and be creative around it!

Join my game/app-dev Discord (2,300+ members):
https://discord.gg/4PFDRsShtX

Until next time,

Anoool

Start with an honest question: what are you actually good at?

Not what you think you’re good at. Not what your LinkedIn says. What is the thing you do that most people around you genuinely can’t?

If the answer doesn’t come fast, that might be your answer. You’re probably a generalist - spreading yourself across everything, mastering nothing. And the world is already full of those.

The real winners are specialists. People who go long and deep on one skill, or a tight cluster of skills that complement each other so well they almost become one thing.

Hard Skills Are Becoming Table Stakes

Here’s the uncomfortable part: a lot of the skills people are grinding right now are quietly becoming worthless.

Coding. UI design. Graphic work. These used to be the barrier. Learn them, and you were ahead. But with AI making all of it accessible to anyone with a laptop and a decent prompt - the barrier is gone. The skill still exists, but the value of owning it doesn’t.

Most people’s response to this? Sharpen the hard skills harder. Use AI to go faster, do more, produce more. But if you zoom out one level - if you think about what that actually produces - you start to see the problem. You’re getting better at something that’s getting cheaper every month.

The skills that are actually appreciating? Storytelling. Original thinking. Design thinking. The ability to see a problem differently than everyone else in the room. These are still rare. AI can assist them, but it can’t replicate them, because they come from you - from your specific experiences and perspective.

The Photographer Analogy

Think about photographers in the 90s. That was a real, valuable, hard-won skill. Film, lighting, composition - not everyone could do it, so the ones who could charged well for it.

Then smartphones happened. Good cameras in everyone’s pocket. The skill didn’t disappear, but its scarcity did. And when something stops being scarce, it stops being valuable.

Ask yourself the same question about whatever you do: Is this skill available to the mass market now? Can someone pick it up in a weekend with the right tools? If yes - that’s not a moat, it’s a treadmill.

Originality Is the Only Real Moat

So what actually holds value?

Perspective you earned. Thinking patterns that were shaped by something specific you lived through - an industry you worked in, a problem you had to solve that nobody else was solving, a life path that just happened to be unusual.

Maybe you spent years in a field that most tech people have never touched, and that crossover is rare. Maybe you faced a specific kind of failure or constraint that changed how you approach problems. Maybe you’ve just done a set of things, in a specific sequence, that almost no one else has - and that combination is quietly a skill in itself.

These things aren’t on your resume. But they’re what make you different from the 10,000 other people who also know how to code or design or ship products.

What If You Don’t Have That Yet?

Then go get it. Deliberately.

Not “do things differently” -that’s just optimizing inside the same loop. Actually do different things. Travel. Talk to people outside your bubble. Work in an industry you know nothing about. Put yourself in situations where your existing playbook doesn’t work.

Originality isn’t a personality trait. It’s a byproduct of accumulated experience. The more varied your inputs, the more distinct your outputs.

For builders specifically - different experiences mean you’ve faced different problems. And the ability to see a problem that others don’t see yet, and build something around it, is worth more than any single hard skill you could grind.

You don’t need to be completely original. Even a bit of genuine differentiation, in the right direction, is enough to stand out in a crowd of people doing the same things faster.

The question is just whether you’re actively building that, or waiting for it to show up on its own.

It won’t.

There’s a developer named Oliver Henry who has an old gaming PC sitting under his desk.

Oliver Henry - creator of Larry!

He calls it Larry.

Larry is not a gaming PC anymore. Larry is an AI agent. And in the last 5 days, Larry has generated over 500,000 TikTok views, pushed Oliver’s MRR to $714, and created content that’s converting into actual paying subscribers.

Snugly’s MRR

Oliver didn’t design a single image. Didn’t write a single caption. Barely even opened TikTok.

I had to break down exactly how he did this - because honestly, this is the most practical AI automation case study I’ve seen.

Wait, Who Is Larry?

Larry is Oliver’s old gaming PC - an NVIDIA 2070 Super that was collecting dust after he stopped gaming. When he heard about OpenClaw, he wiped the drive, installed Ubuntu, and turned it into an autonomous AI agent.

OpenClaw is open-source software that gives an AI (in this case, Claude) a persistent identity, memory, and access to real tools. It’s not a chatbot. It’s an agent that lives on your machine and actually does stuff.

Here’s what Larry has access to:

• His own personality and memory that persists between sessions

• Ability to read and write files on Oliver’s machine

• Image generation through OpenAI’s API

• Custom code he writes himself for text overlays

• Posting to TikTok via Postiz API

• Skill files that teach him specific workflows

• Memory files where he logs every lesson learned

Oliver talks to Larry on WhatsApp. He tells him to research winning formulas. Larry does everything else.

The Results (In Under a Week)

• 500K+ total TikTok views

• 234K views on the top post

• 4 posts over 100K views

• 108 paying subscribers across his two apps

• ~$588/month MRR and growing

• Cost per post: roughly $0.50 in API calls

• Time Oliver spends per post: about 60 seconds to add music and publish

Those aren’t vanity metrics. People watch the slideshows, download his apps, start trials, and subscribe.

What Apps Is He Promoting?

Oliver built two iOS apps:

Snugly - an app that lets you photograph any room and see it redesigned in different styles using AI.

Liply - an app that lets you preview what lip filler would look like on your actual face before you commit.

He launched these right before starting his job at RevenueCat. Without Larry, these apps wouldn’t be getting promoted at all. He doesn’t have the time.

The Content Format: TikTok Slideshows

TikTok photo carousels are blowing up right now. According to TikTok’s own data, slideshows get 2.9x more comments, 1.9x more likes, and 2.6x more shares compared to video. The algorithm is actively pushing photo content in 2025.

Every slideshow Larry creates has:

• 6 slides exactly (TikTok’s sweet spot)

• Text overlay on slide 1 with the hook

• A story-style caption that mentions the app naturally

• Max 5 hashtags

How Larry Generates the Images

Larry uses gpt-image-1.5 through OpenAI’s API. Oliver chose this model for two reasons:

1. It’s what his app uses. Snugly generates room designs with the same model, so the TikTok images match exactly what users see when they download. No bait and switch. The marketing IS the product.

2. It looks real. When you include “iPhone photo” and “realistic lighting” in the prompt, the images genuinely look like someone took a photo on their phone. Not AI art. Not renders. Actual photos.

The Prompt Engineering (This Is Where It Gets Good)

The challenge with room transformations: you need the SAME room across all 6 slides, just in different styles. If the window moves or the bed changes size between slides, the whole thing looks fake.

Oliver’s solution: lock the architecture.

Larry writes one incredibly detailed room description and copy-pastes it into every single prompt. Room dimensions, window count and position, door location, camera angle, furniture size, ceiling height, floor type. All locked.

The only thing that changes between slides is the style - wall colour, bedding, decor, lighting fixtures.

Here’s a real example of what a prompt looks like:

iPhone photo of a small UK rental kitchen. Narrow galley style kitchen, roughly 2.5m x 4m. Shot from the doorway at the near end, looking straight down the length. Countertops along the right wall with base cabinets and wall cabinets above. Small window on the far wall, centered, single pane, white UPVC frame, about 80cm wide. Left wall bare except for a small fridge freezer near the far end. Vinyl flooring. White ceiling, fluorescent strip light. Natural phone camera quality, realistic lighting. Portrait orientation. Beautiful modern country style. Sage green painted shaker cabinets with brass cup handles. Solid oak butcher block countertop. White metro tile splashback in herringbone. Small herb pots on the windowsill…

The bold part is the only thing that changes. Everything else stays identical across all 6 slides.

How the Posts Actually Get Published

Larry posts everything through Postiz - a social media scheduling tool with an API. He uploads every slideshow with privacy_level: "SELF_ONLY" which means it lands in Oliver’s TikTok drafts folder.

Why drafts? Because music is everything on TikTok.

Adding a trending sound massively boosts reach. But you can’t add music via the API, and trending sounds change constantly. So the workflow is:

1. Larry generates images, adds text overlays, writes the caption

2. Larry uploads everything to TikTok as a draft via Postiz

3. Larry sends Oliver the caption in a WhatsApp message

4. Oliver opens TikTok, picks a trending sound, pastes the caption, hits publish

Oliver’s part takes about 60 seconds. Larry’s part takes 15–30 minutes.

That’s the magic. Larry does 95% of the work. Oliver just adds the finishing touch that can’t be automated yet.

How Larry Learns and Gets Better

This is where most people’s AI setups fall short.

Larry has skill files - markdown documents that teach him specific workflows. His TikTok skill file is over 500 lines long. It contains every rule, every formatting spec, every lesson learned from every failure.

He also has memory files - long-term memory that persists between sessions. Every post, every view count, every insight gets logged. When Oliver asks him to brainstorm hooks, Larry isn’t guessing. He’s referencing actual performance data.

Larry also has access to Oliver’s RevenueCat analytics, so he can see daily MRR changes, subscriber counts, and churn - important metrics for knowing how well the marketing is actually converting.

Every failure becomes a rule. Every success becomes a formula. The system compounds.

The Failures (Before It Worked)

They tried local generation first

Oliver’s PC has an NVIDIA 2070 Super, so naturally they tried generating images locally with Stable Diffusion. Free generation. No API costs.

It wasn’t good enough. Room transformations need photorealistic output that looks like someone actually took a phone photo. Stable Diffusion kept producing that slightly uncanny AI look. The gap between local generation and gpt-image-1.5 was massive.

The API costs turned out to be tiny anyway - about $0.50 per post, $0.25 with Batch API.

Images that looked terrible

Early on, Larry was generating rooms at 1536x1024 (landscape) instead of 1024x1536 (portrait). Black bars on every video. Killed engagement.

He was also using vague prompts. Rooms looked different on every slide. Windows would move. Beds would change size. The transformations felt fake because you could tell it wasn’t the same room.

Text that was unreadable

Text overlays were too small. Positioned too high, hidden behind TikTok’s status bar. The canvas rendering was compressing text horizontally because lines were too long. Everything looked squashed.

They’d post something and wonder why it got 200 views. Then Oliver would look at it on his phone and realize you literally couldn’t read the hook.

Hooks that nobody cared about

Their first hooks were all self-focused:

• “Why does my flat look like a student loan” → 905 views

• “See your room in 12+ styles before you commit” → 879 views

• “The difference between $500 and $5000 taste” → 2,671 views

Dead. All of them.

They were talking about themselves. Their problems. Their app’s features. Nobody cared.

The Breakthrough: Hooks That Actually Work

Then they tried: “My landlord said I can’t change anything so I showed her what AI thinks it could look like”

234,000 views.

That one post got more views than everything else combined.

They tried it again with “I showed my mum what AI thinks our living room could be.” - 167,000 views.

Again with “My landlord wouldn’t let me decorate until I showed her these.” - 147,000 views.

The formula became clear:

[Another person] + [conflict or doubt] → showed them AI → they changed their mind

Every post that follows this formula clears 50K minimum. Most clear 100K. Everything else struggles to break 10K.

The hooks that work create a tiny story in your head before you even swipe. You picture the landlord’s face. You picture the mum being impressed. It’s not about the app - it’s about the human moment.

How to Set This Up Yourself

Here’s the step-by-step breakdown:

1. Get a machine running Linux

Any old computer, a Raspberry Pi, or a cheap VPS. A Mac Mini if you’re feeling fancy. Install Ubuntu if you’re not sure what to pick.

2. Install OpenClaw

It’s open source and free. Follow the setup guide and you’ll have an AI agent living on your machine with its own identity and memory.

3. Get an image generation API key

Oliver uses OpenAI. Sign up at platform.openai.com. You’ll use gpt-image-1.5 for image generation. Expect to spend about $0.50 per slideshow, or $0.25 with Batch API.

4. Sign up for Postiz

This connects your agent to TikTok. It has an API that lets you upload slideshows as drafts.

5. Write your skill files

This is the most important step. Create markdown files that teach your agent exactly how to do the job:

• Image sizes and formats (1024x1536 portrait, always)

• Prompt templates with locked architecture descriptions

• Text overlay rules (font size, positioning, line length)

• Caption formulas and hashtag strategy

• Hook formats that work in your niche

• A failure log so the agent never repeats mistakes

Write them like you’re training a new team member who’s incredibly capable but has zero context. Be obsessively specific. Include examples. Document every mistake.

6. Start posting and iterating

Your first posts will probably be bad. That’s fine. Log what went wrong, update the skill files, keep going. The system gets smarter with every post.

The Real Unlock

The agent is only as good as its memory.

Larry didn’t start good. His first posts were embarrassing - wrong image sizes, unreadable text, hooks that nobody clicked. But every failure became a rule. Every success became a formula.

He compounds. And now he’s genuinely better at creating viral TikTok slideshows than Oliver is.

That’s the real unlock. Not the AI itself. The system you build around it.

Follow Oliver’s Journey

Oliver is building Snugly and Liply in public. You can follow him at @oliverhenry on X.

Larry has his own X account too: @LarryClawerence.

My Take

This is probably the most detailed, practical breakdown I’ve seen of someone actually using an AI agent to do real marketing work. Not “I asked ChatGPT for captions.” An actual autonomous system that generates images, writes copy, posts content, and learns from its mistakes.

The $0.50 per post cost is wild. The 60 seconds of Oliver’s time per post is even wilder.

Is this the future of indie app marketing? Maybe. Worth experimenting with at least.

Follow me on X: https://x.com/anulagarwal

Say Hi to me on LinkedIn: https://www.linkedin.com/in/anulagarwal/

Millions (maybe billions) of people scroll reels on various platforms everyday, that is where they live - and that is where you should market.

Stop chasing gaming influencers/streamers and become your own - that is what these guys did.

Here are a few case studies of indie devs, who kept sharing videos about their games on social platforms and achieved success.

1st case study: Bopl Battle

Johan Gronvall sold 500,000+ copies of his game, Bopl Battle, on Steam, with 0 efforts on paid marketing and UA.

Check out his video here, where he talks about his whole process and what worked:

His videos get millions of views on YouTube & TikTok, and sold 500,000+ copies of his game!

He can now ‘predict’ which of his videos will do well.

2nd case study: Virago

Virago is a narrative, serious mobile game made by Sunday Akor. He leverages TikTok and YouTube, and his game reached the Top #3 free games in the USA.

I had a quick call with Sunday, and his biggest skill is storytelling, which he integrates into his videos and hence reaching millions of viewers that download the game.

3rd case study: Highway Racer Pro

Highway Racer Pro is a racing mobile game that went viral on TikTok and reached #1 spot in the racing category. Their videos show straightforward gameplay without any fancy editing.

The developer behind Highway Racer Pro is a big studio and not any indie dev - they made $100,000+ post-launch, all through the support of TikTok.

New era of marketing: Short-form videos

Learning to create videos that sell or reach a wider audience is an underrated skill today.

If you have a perspective that can convert any game or app idea into an engaging video, you win. You can reach viewers in different parts of the world and sell your idea or product to them.

It works, I have tried it for my game: Squishy Cats. I don’t have crazy million views, but my videos convert. People watch them and go the app store to search and download.

[If you have a high-quality game/app and need help with short-form marketing, reply here!]

A few tips on what worked for me on TikTok:

What to do:

✅ Authenticity wins

✅ Share behind the scenes - talk about your journey, progress

✅ Have interesting hooks and opening sentences + visuals

✅ Warm up your account before posting

✅ Keep an eye on analytics and iterate

✅ Keep watching different creator and seeing what works

✅ Experiment with formats: voiceovers, captions, footage style etc.

What not to do:

❌ Don’t overpost, max 2-3 videos/day

❌ Have basic narration that goes like: “Download my game pls”

❌ Don’t repost old videos or other channel videos (algorithm is very smart)

❌ Don’t clickbait

❌ If from India, don’t use VPN to login - your account will be banned

This is the golden era for indie-devs with AI tools + leveraging social media, all you need to do is build and ship :D

Join my game-dev discord, where we talk about:

• Web-Games

• AI+Games

• Apps

• Short form marketing

Until next time,

Anoool

When I landed my first job (Game Designer) in the games industry 6 years ago, I was making just about $350 per month.

Fast forward to a few days - I earned $2,200 for only 2 hours of game development work.

Let me share exactly how that happened, along with the long and winding path that led me here.

My Journey in Game Development

I’m a 27-year-old indie game developer from Vadodara, Gujarat - a tier-2 city in India.

My journey into development started about 13 years ago, creating websites and Android apps. Games weren’t even on my radar initially.

About 6 years back, I pivoted into game development. Like many others, I took a job at a small game studio, excited but unsure of what to expect.

Over the next few years, I moved through a series of positions (Game Developer, Publishing Manager, Game Lead/Producer), each ending with layoffs.

My last full-time position was as a Game Producer in Germany in 2022.

I moved back home to India, determined to build something of my own.

Since late 2022, I’ve focused entirely on making browser games - simple, engaging games that run directly in a web browser without any downloads.

To date, I’ve developed over 15+ browser games, mostly solo, but occasionally working with small outsourced teams.

One standout success has been my “Watermelon Game”, which not only hit the top 3 on Coolmath Games but also won the Game of the Year award last year.

Overall, my web games have reached more than 30 million players worldwide - an incredible milestone considering how uncertain everything seemed just a couple of years ago.

I started creating content around this on LinkedIn & YouTube and got invited to give a small talk on web-games! My first ever offline talk, I was super nervous 😬

The Two Models for Making Money in Browser Games

To understand exactly how I made $2,200 in just two hours, you need to know how browser game developers typically earn:

• Revenue Share Model: You integrate ads into your game, and revenue from these ads is split between you and the hosting website. It can be lucrative but fluctuates based on user engagement and traffic.

• License Fee Model: You sell a license for your game to a platform upfront. You integrate their logo and SDK and hand over the build. You receive a fixed amount regardless of how the game performs.

You can also check out this detailed video guide I made on browser games.

Stay tuned for more updates, insights, and behind-the-scenes details by joining my Discord server.

How I Made $2,200 for 2 Hours of Work

A few months ago, I proactively reached out to a number of browser game platforms via LinkedIn and Email, showcasing my portfolio of games.

This is something I regularly do - I keep pitching to platforms to increase my games’ exposure.

Surprisingly, AFTER MONTHS, the founder of one of these platforms replied, expressing interest in two of my games.

After a brief discussion, we agreed on a fee of $2,200.

The tasks required were straightforward: integrate their logo, make a few minor adjustments, and deliver the build. Despite Unity’s notoriously slow WebGL build and testing cycles, the entire process took me around two hours.

To my delight, the founder paid upfront via PayPal, demonstrating not only professionalism but genuine respect for my time. Just like that, my rate effectively became $1,100/hour - beyond anything I’d ever imagined.

Key Takeaways from This Experience

If you’re an indie developer or aspiring to become one, here’s what I’ve learned:

• Be proactive: Don’t wait for opportunities to find you. Regularly pitch your games to different platforms.

• Understand your market: Recognize the difference between revenue-share and licensing models, and choose wisely based on your goals.

• Build relationships: Connect genuinely with people in the industry. Networking on platforms like LinkedIn can lead to surprisingly lucrative opportunities.

• Keep things simple: Sometimes minor adjustments are enough to close a valuable deal. Focus on providing polished, adaptable builds to prospective buyers.

What’s Next?

My next big step is my indie mobile game, Squishy Cats - a physics-based puzzle game featuring jelly-like cats that squeeze through obstacles to cuddle up with their partners.

Until next time,

Anul

Okay, this is actually quite crazy — I was able to recreate all game art assets from some of the top games you see on mobile stores, and it literally took me less than five minutes using ChatGPT. Why? Because ChatGPT’s latest image generation model lets you replicate or create any art style you can think of effortlessly.

Here’s exactly how you can do it too.

What You Need

First things first — you’ll need to be on the paid version of ChatGPT since this image generation feature isn’t out for the free model yet.

But honestly, it’s worth it.

You can also watch video version of this article:

Step-by-Step Example:

Sticker Game by Lion Studios

Take this sticker game, super popular on mobile. You need tons of different art assets, characters, and environment objects, right? All I did was take a screenshot, pasted it into ChatGPT, and typed a simple prompt like, “recreate this image style.” Seriously, you don’t even need to be great at writing prompts — just basic words are enough.

It took around 2 to 3 minutes (I sped up the video), and boom — you get a near-perfect replica of the original image. Then, you can easily clean out the background and extract individual assets using Photoshop, Photopea, or whatever your go-to tool is.

Angry Birds Example

Next, I tried Angry Birds. Same drill — I took a screenshot, pasted it into ChatGPT, and asked it to extract objects like birds and elements. Initially, I didn’t get the exact style because of graphical overlays covering the image, so I just took a cleaner screenshot, and ChatGPT nailed it this time.

You get a clean object sheet ready for your edits. You can even ask ChatGPT to add specific textures or art styles. I tried asking for a shaded texture, and it came out pretty cool.

Candy Crush with a Twist

Candy Crush is huge and visually appealing, right? So, this time, instead of a direct replica, I asked ChatGPT to modify the texture into something rope-like. Again, within a couple of minutes, I got assets in a totally unique style — including that cute little monster character in the center.

Funky Among Us Characters

For Among Us, I wanted those jelly-like imposters, but with some added flair. I asked ChatGPT to give them funky hats and some texture. Again, simple prompt — fast results. It delivered characters perfect for mockups, wireframes, or whatever creative ideas you have.

Customized Uno Cards

Uno is another super popular card game, so I tried creating my own style of Uno cards. ChatGPT managed to generate a complete card design — numbers, texts, stats, everything. Just select the image, clean the background in Photopea or Photoshop, and you have custom card assets ready to import into Unity or wherever you need.

Cheat Sheet for Writing Good Prompts

• Be Specific: Clearly state what you want to replicate or create.

• Mention the Art Style: Cartoon, realistic, sketch, pixel art — specify the style.

• Request Adjustments: Add texture, shading, or color variations.

• Clarify Purpose: Game asset, character design, environment object.

• Use Reference Images: Always provide a visual reference for the style.

• Keep It Simple: Short prompts can often yield great results.

Wrap Up

Honestly, this method lets you have your entire art asset set ready in less than an hour if you plan it well. Whether it’s for your actual game, mockups, ideation, or visual infographics, ChatGPT makes this process incredibly simple and fast.

Join my game-dev discord: https://discord.gg/4PFDRsShtX

That’s it for now! I’ll be posting more videos soon on creative ways to use ChatGPT!

See you soon!

neal.fun

He uses Vue.js to build these games. Vue is a javascript framework library.

His games keep going viral.

He makes quirky, funky & ‘unusual’ games - the kind of games that are not trending, or wanted by the popular web-platforms.

But since he does not work with web-platforms, how does he reach the players, and bring them to play his games?

Leveraging Twitter/X

He has a massive following on Twitter/X and this is where he posts updates about his new games.

Infinite Craft

A year ago, he posted about his game - Infinite Craft.

This game concept is not new - there are games like Little Alchemy which follow the same gameplay, but Neal provided an infinite merging option for hte players!

This bought him 21 million impressions + plenty of engagement 🤯

He went ‘viral’ and people jumped at the chance to play his game. People were searching for his game all across Google.

His follow up posts on the game, did decently well on engagement and views.

His web-platform, neal.fun, got an instant SEO boost and now receives ~2 million players+ every month.

And majority of them come from the US!

This is not a small number for a solo-developer, many web-platforms like Lagged or GamePix cannot achieve this with a decent team.

It was also one of the most Googled games in 2024!

Mobile Release

The game was also released on IOS & Android - again, gaining massive traction.

I have followed Neal for a while on Twitter/X and his game updates always intrigue me, I love how he comes up with new, funky, and unusual ideas.

I guess this is how he has attracted such a huge player base on the platform!

Do check out his other games - most of these games are like toys and don’t feel like the actual games (mobile/steam), hence these games naturally attract audience online!

A solo developer gaining such massive traction is an inspiring story, just proves how solo-devs can easily make a good living by developing games.

[AD]
Join my game-dev Discord, with 1250+ members discussing:

Web-games, AI + Games, App Dev & getting feedback for their games!

Hop in here

Best,

Anul

[AD] 
Join my game-dev Discord, with 1250+ members discussing:
Web-games, AI + Games, App Dev & getting feedback for their games!
Hop in here

2024 was an exciting year for me in the web game space - I was making some good $$$, but I quickly noticed a challenge: my games would lose traction shortly after launching on a web platform.

This is a problem with most of the game devs building for the web: games will die out quickly!

I started looking for a more consistent revenue stream, and that’s when things got interesting.

I connected with Amit Dubey from GameBee Studio—a pro at publishing TV games. With Amit’s guidance (and Bosco D’mello’s expertise), I successfully ported one of my games to TV, and in the process of porting another one.

What are TV games anyway?

Games playable on TV, via the use of Remote control (arrow keys)

This is a niche but growing market!

• Platforms: You can upload your games to:

  • LG WebOS

  • Samsung Tizen

  • Amazon Fire TV

  • Roku

  • Jio STB

What’s Different About TV Games?

• Input: On TV, you only have the remote and arrow keys.
  ❌ Drawing games? Not the best fit.
  ✅ Racing games? Perfect!
  
  ❌ Match-3 games? Not the best fit.
  ✅ Space Shooters? Perfect!

• UI & Controls: Setting up navigation with arrow keys is super straightforward in Unity. So you need to ensure every button can be accessed by pressing the arrow keys.

• Monetization: Instead of the ad-dominated revenue in web games, we integrated in-app purchases (IAPs) and subscriptions for a steadier income. This was a first time for me to focus on IAPs & subscriptions instead on solely ad-based model.

GameBee is spending some $$$ on User Acquisition to get users, check the data/analytics, and ensure that the game will be profitable in the coming weeks.

Plus, I’m porting my Happy Fruit Game (20 million+ players) from the web to TV—with a fresh local multiplayer mode in the works!

The local multiplayer is basically a Pass n Play mode - so they can use the same screen but compete for highscore!

By March, I’ll have more data on how these TV games perform and how much $$$ they can generate for an indie developer like me.

Venturing into Apps

I am focused more on building apps in 2025.

I see many Indie Developers crack the app distribution through short-form content via TikTok & Instagram, and to be honest, it does not seem too difficult.

More on this in the next editions!

Keep building,

Keep shipping!

Cheers,

Anul

1 year ago, I made the Watermelon Game and distributed it over web-platforms.

The game has been in Top 3 on Desktop for this platform since then, and holds the #1 spot on mobile.

It also has ~350,000 ratings at 4.6/5!

Last week, the game won the Game Of The Year award on the Coolmath Games platform which gets millions of players, from the USA (and others), every month!

You can checkout the awards list here.

The best part is that I developed this game solo, using ChatGPT for code & art!

What helped this game achieve this award is:

🌟 Constant seasonal theme updates (Halloween/Christmas)

🌟 Optimization for mobile

🤔 Making games work for mobile with Unity is tricky, but I figured it out before Unity 6

Next, I plan to distribute this game over TV platforms: I think it can be a perfect fit for the TV audience who want to play games!

TV platforms like LG WebOS, Samsung Tizen & Amazon Fire TV are growing market for games.

2024 was all about web-games as a sustainable source of revenue for my bootstrapped, solo game studio - Momo Games.

2025 will be about making real $$$ from building games.

Stay tuned on my YouTube, for my game-dev journey in 2025!

YouTube

Merry Christmas and Happy Holidays! 🎅🏻

Best,

Anul

Hello 👋

Have you ever dreamed of creating and earning money from your own video games?

But here’s the catch: many people don’t have the time to build their game projects.

Whether you're working in marketing, finance, as a barista, a doctor, or even a cab driver, chances are you want to create side projects while managing a full-time job. Sadly, time is limited.

A friend of mine faced this exact problem. He wanted to make games on the side, but thought it would take too long and didn’t have enough time to spare.

So, today, I’m here to show you how you can become a super programmer, save tons of time building your side game projects, validate your ideas early, and ship games fast! Plus, I'll share how you can earn money by distributing your completed games to millions of players online!

How to Become a Super Programmer

Goal: Create quick prototypes and reduce time spent on repetitive tasks.
Aim: Test your ideas and see if players enjoy them.

But you might not know where to start if you're a beginner.

So, here are four effective ways to achieve this:

1. Create Custom Templates

Your first game might not be a hit. However, enjoying the game-making process is crucial—the most fun part is coding the gameplay. So, why not make it easier?

Every game project needs:

• Assets: Fonts, sounds, and UI elements.

• Scripts: Managing the UI and game logic.

• Managers: Handling audio, levels, and more.

Thus, create custom templates with these common parts to save time on each new project. No need to start from scratch every time!

Example: Have a pre-built menu system in your template. Just customize it to fit each game’s unique style.

In Unity, setup your scenes, assets, scripts, and resources. Export it as a Unity Package or keep it updated in a Github repository.

Whenever you wish to start a new project, clone this repo!

2. Use Tools Like ChatGPT

I’ve been writing code manually for 11 years—in Notepad, on my phone, you name it. I have deep experience in logic and programming.

Yet, coding can be time-consuming and repetitive. That’s why I use ChatGPT to write complex code, fix bugs, and solve problems.

With ChatGPT, I can create custom-level editors, complex systems, and algorithms. So, any feature I think of can be added without feeling overwhelmed.

Some might worry that using AI is cheating. However, it’s just a tool to help you work faster and focus on the fun parts of making games.

Break down your idea in tech terms to ChatGPT (use o1/subscription) and ask it to help you implement in Unity.

3. Outsource Projects

If your ideas are small and you can invest some money, hire a freelancer to develop the game for you. This way, you can focus on designing the idea and working with the freelancer to bring it to life.

Here’s how:

• Find a Reliable Freelancer: Use Upwork, Freelancer, or Fiverr.

• Explain Your Idea Clearly: Provide drawings or detailed descriptions.

• Set a Budget and Timeline: Agree on costs and deadlines.

• Communicate Regularly: Ensure the project is on track.

This means you can bring your vision to life without handling all the coding yourself. However, clear communication and trust are essential.

4. Choose Manageable Ideas

Understand your strengths and limitations.

You could create a big game like Stardew Valley, spending four years on it—but that’s not practical with limited time.

So, make small, manageable games and release them one by one, like the simple games you see on the screen.

Consider:

• Your Ability to Take Risks

• Your Technical Skills

• Your Experience

Choose an idea that fits your current situation. If you’re working full-time, you might only have 2-3 hours a day to work on a game. Therefore, pick a small but exciting project.

5. Validate Your Ideas Faster

Making a game is one thing, but how do you know if people will like it? That’s why you need to test your ideas.

Here’s how:

• Paid Ads: Spend a little on ads to check interest. There is no end to how much you can spend, not the best option for small teams.

• Publisher Testing: Partner with a game publisher for audience testing. But the publishers are looking for certain styles of games that can be scaled on mobile.

• Social Platforms: Share your game on social media for feedback. Long term game.

• Poki Playtesting: Use platforms like Poki to let players try your game.

  • This is the best tool, to validate your ideas as early as possible.

  • The tool shares your game with people on their platform and records the screen, so you can check how the players are playing the game.

  • Export a web-build, upload on the tool, start the playtest session.

  • Wait for the recordings to come in (10 at once) - watch the recordings, analyse them.

  • All of it can be done within an hour!

But remember, feedback might be mixed. So, focus on helpful comments to improve your game.

Conclusion

You can become a super programmer and build games quickly, even with little time. Yet, it requires smart strategies:

• Create custom templates to save time on repetitive tasks.

• Use tools like ChatGPT to help with coding.

• Delegate projects to a freelancer!

• Choose ideas that are manageable given your time and resources.

• Validate your ideas early to ensure you’re on the right track.

Now that you have a game ready, how can you earn money from it?

Check out my other video to learn how you can earn money while sharing your games with millions of players.

So, enjoy making games, see if players like them, and maybe even earn some money—all without spending endless hours.

Start today! Use these tips to build your side projects faster and smarter. Who knows? Your game might be the next big hit!

Stay Connected:

Join my game-dev Discord (750+ members) where we discuss web-games, ai+games and everything related to making games!

Keep Building,

Keep Shipping!

Anul

Every week, I will insights from my journey and how I am earning $$$ from different sources as a solo developer/solopreneur.

Right now I am earning by making web-games, but soon I am adding more sources of income, which I will talk about soon!

Making your games playable on any device (mobile/PC/tablet) by designing adaptable UI & camera is a problem many game devs face.
[insights from discussions in my Discord]

Today, I'm sharing five tips to help you make your games playable on any device:

1. Anchor UI Elements to Screen Corners

Ensure your UI elements stay in place regardless of screen size or resolution by anchoring them to the appropriate corners.

Assign anchors based on each element's position so they adjust seamlessly when the screen resizes. This provides a consistent and adaptable UI experience for all players.

2. Make UI Elements Resizable

Allow your UI elements to resize dynamically with the screen by using anchor and stretch settings based on height and width.

This maintains proportionality and usability across different devices, ensuring buttons and menus remain user-friendly whether on a smartphone, tablet, or desktop.

I also have a video version of these tips here:

3. Toggle UI Elements Based on Orientation

Optimize your game's interface for both portrait and landscape modes by enabling or disabling specific UI elements based on the device's orientation.

Detect the orientation using Screen.width and Screen.height, and adjust your UI accordingly to keep the interface clean and intuitive.

4. Switch Cameras Based on Orientation

Enhance the visual experience by using different camera setups for portrait and landscape modes.

Implement separate cameras for each orientation, allowing you to zoom in or out and adjust backgrounds to suit the mode. This makes your game more engaging and visually appealing.

5. Adapt Camera Settings Based on Screen Size

Maintain consistent gameplay views across all devices by adjusting camera variables like field of view (FOV), size, or distance based on the screen's aspect ratio.

By detecting the ratio of height to width, you can ensure important elements remain in view, providing a seamless experience for all players.

Bonus Tips:

• Consider Control Adaptability: Ensure your game controls are flexible for touchscreens, mice, and keyboards to enhance the gaming experience across different platforms.

• Web Distribution: Publishing your game on the web can help you reach millions of players easily. However, make sure your game is accessible and enjoyable on various devices to maximize its potential reach. You can watch my video here!

Implementing these strategies will make your games more accessible and enjoyable for everyone, ultimately expanding your player base and enhancing your game's success.

Until next time,

Anul

I transitioned from building mobile games to web games a year ago - with no idea of the players that play these games or what platforms are out there.

Mobile games are good but very risky today - Working with a publisher does not guarantee success either. Cracking the CPI/LTV combo is a hassle.

Web games are good small money for every game you make - if you have good + original + engaging gameplay ideas, you can make 5-fig $$$ (although don’t go by my word and self-explore!).

There is plenty of good small money that can be made by distributing good games on the web.

But what platforms are out there and how do you reach out to them? What games to make?

I made a video covering the basics of building and publishing games on the web on my YouTube channel 🫡

Watch the video, it covers:
🌟 Choosing a game engine
🌟 Coming up with ideas for web games
🌟 Connecting with web-platform publishers (a small list of these platforms)

Also, do let me know the feedback on the video in the comments of the video! It will help a lot 🙏

I have a discord server with 500+ game-devs discussing web games, ai+games, and making games in general, you can join here:

https://discord.gg/35JAZwsB

If you want feedback on your games or have any questions: hop in and ask away!

Until next time,

Anul

I made ~12 web games in the past 8 months — I developed most of them by myself, and some projects I outsourced to freelancers!

I am a Game Developer and I build puzzle/arcade/sports games using Unity.

I work with web-platforms like GameDistribution, GamePix & Coolmath Games. I am expanding my games to new platforms, like Telegram, soon!

Web game platforms are the easiest way to get players to your game and earn revenue. I use ChatGPT to write most of the game code and upload my games on these platforms.

How I made $2,000

Now, an upcoming web games platform wanted some new games for its platform and asked us to license some of our previous games to them.

We decided together on 4 games (and more later) to ship to their platform, at a license fee of $500/game (less than our standard license fees).

The platform is in its early stages hence I did not negotiate on the price and rather focused on a long-term partnership!

They requested to remove the splashscreen logo and allow SiteLock (so you cannot play on other websites) for 2 domains.

With these changes and QA testing, I spent 2 hours/game (exporting from Unity took 30% of my time :( ) to make this happen.

So instead of working on a new project, I am still making money from my old games, without having to write new code!

But this is rare and not a recurring/repeatable source of revenue -> the golden rule for me is to focus on recurring sources of revenue instead of one offs!

My watermelon game still earns me about ~$300/month!

Making games using Unity is quite easy now for me. I heavily use ChatGPT for complex systems and algorithms, it has cut down my development time by more than 50%.

For the rest of 2024, my focus is on marketing them in the best, unconventional ways possible 🚀 Like using YouTube and promoting organic traffic for my games!

(If you have a platform and want some games for your players -> feel free to reach out!)

Conclusion

Making games and earning money is satisfying. You just have to keep identifying opportunities!

Stay tuned as I experiment more with gen-AI tools for game dev, share more unconventional ways to earn money by making games!

Subscribe now

You can connect with me on LinkedIn where I talk about gen-AI tools and game industry trends.

What do I develop? Websites, and apps before — now I make games and earn a good living ($8,000 - $9,000/month)

I have developed since I was 15 (out of passion and curiosity). I started making simple websites using PHP, JS, and MySQL (+HTML/CSS). I used to finish projects for 12th grade students and get some small fees in return.

In 2014, I was 16 and switched to Android Apps. I made utility apps — music players, photography, launchers etc. Made around $1000!

I joined a design university in Pune which included gaming as a course. 3 years later I went to Scotland for Masters, all the while still making apps on the side as a passion. I did not earn a lot, as I was busy upskilling and enjoying my Uni life!

Right before COVID, I returned to Pune for a job, and during COVID I ‘quiet quit’ my work and started building games on the side. The job was paying me ~$400/month [fresher].

I got fired eventually, and set up my small remote team to make games. I hired a couple of game devs and artists. I partnered up with publishers who would fund our hypercasual mobile games. I did this from 2020–2022. I earned about $60,000 in total over these 2 years.

I got a good job offer from a gaming company in Germany in 2022 — got fired again, and I came back to India. I saved about $22,000 in 10 months while I worked there.

From 2023 to now, I am again making games — mostly as a solo developer but also hiring some freelance people time to time.

I make web games for platforms like Coolmath games, Gamepix, and GameDistribution. I also take on freelance projects. They pay well, per game, for someone living in a country like India where costs are low. Over the past few months I am earning $8,000-$9,000 every month (growing steadily)

Each game takes around 2–3 weeks to develop, test and ship to these platforms. I will also put my games on JioGames soon.

I use tools like ChatGPT to write most of the game code for me. Check out this video where I earned $10,000 from a game I made using ChatGPT.

I upskilled in writing and I am very active on LinkedIn (~10k followers). I get my clients and partnerships from there.

Writing allows to me talk about my development process in a fun way and gets more people to reach out to me.

I am more like a ‘solopreneur’/’indie’ game-dev who wants to leverage social media.

All of this to say, focus on upskilling and keep at it.

Developing apps/products is my passion, I have been fired from all 3 jobs I did but that does not mean there is no alternative path. I can never fit in any job structure because I work at a very flexible yet high pace.

Subscribe now

At some point, I may switch back to apps! I miss finding problems in my daily life and creating solutions around it!

Happy to answer any questions!