How to Really Make a Game.

The craft, the process, and everything nobody warned you about. A visual journey through the hardest creative thing a human being can attempt.

Begin ↓

Games are the hardest creative thing a human being can attempt. This is the argument — built across ten acts, sourced, and written for anyone who makes games, loves games, or is thinking about making one.

Ten acts. Full-source. Scroll all the way through.

A note before you start: This isn't a guide to coding a game. It's about creating a game as an idea — as a creative work, a product, and a design challenge. The concepts here apply whether you're a solo developer, a team lead, a producer, or someone who's never opened an engine. The principles don't change.

ByJosh, Indieformer
Acts10

Act One

The Claim

Why games are the hardest creative thing a human being can attempt

The provocation

Games are the hardest creative thing a human being can attempt.

Not the most expensive. Not the most technically complex. The hardest — in terms of breadth and depth of creative discipline required simultaneously.

The credibility

"I didn't always believe that."

This comes from someone with a background in film and television — a world that genuinely respects craft — who got close to game development and had to rethink what craft actually means. The complexity doesn't announce itself. It accumulates.

The Medium That Outgrew Everything Context
Scale Games now generate more revenue than film and recorded music combined — the dominant entertainment medium on the planet. Most people still think making a game is like making an app.
$0B Global game revenue, 2024
Statista ↗
0B Active players worldwide
Newzoo ↗
0 New Steam games in 2025
Sensor Tower 2026 ↗
This is the landscape. This is what anyone making a game walks into.The question isn’t just “can you make it” — it’s “can you make something that earns its place in this.”
Act Two

What Someone Signs Up For

Eight disciplines. Not sequentially — simultaneously, from day one

2
The Full Scope Eight jobs in one
Eight jobs, one person When someone decides to make a game, they're not signing up for one job. They're signing up for eight — simultaneously, from day one. In film, you have whole departments for each. In indie game development, one person often holds all eight at once.
DESIGN
Rules · Systems · Balance · Feedback loops
CODE
Engine · Logic · Performance · Platform ports
NARRATIVE
Story · World · Character · Tone
PRODUCTION
Scope · Timeline · Budget · Risk
ART
Visual identity · UI · Animation · Direction
MARKETING
Positioning · Wishlist · Community · Press
AUDIO
Music · SFX · Adaptive audio · Mix
BUSINESS
Publishing · Legal · Revenue · Sustainability
What games actually cost to build
Solo / game jam
$0 – $10K
Small indie team
$10K – $100K
Mid indie studio
$100K – $1M
AA studio
$1M – $25M
AAA blockbuster
$50M – $300M+

The indie advantage: ship at a fraction of the cost. The indie constraint: no second chances if you misjudge scope.
Ranges are aggregate industry estimates. Developer funding data: GDC State of the Industry 2025 ↗

How developers fund their games — 2025
Self-funding
56%
Publishing deal
28%
Govt grants
15%
Venture capital
15%
Friends & family
14%
Crowdfunding
11%

Self-funding is the dominant route — 82% of indie developers report putting their own money in, compared to 40% of AA and 29% of AAA studios. Most developers pursue more than one source.
Source: GDC State of the Industry 2025, p.30 ↗

Each of these disciplines is a real career in its own right — eight of them, compressed into a single project.
Act Three

Everything Film Does,
Games Do Too

The shared foundation — and why it already makes games extraordinary

Clair Obscur: Expedition 33

Clair Obscur: Expedition 33 — character design, visual direction, and world-building as a unified language. 2025.

3
Everything Film Does, Games Do Too The shared foundation
Narrative Story structure, pacing, dramatic tension — the craft of making people care. Every game with a story needs a writer who understands narrative at a fundamental level. And unlike film, the player controls the pace.
Character People don't connect to plots. They connect to people. Creating characters who feel real is as hard in games as anywhere — harder, because you may need to write a character the player inhabits rather than observes.
Visual Design Every frame is a choice. Colour theory, composition, atmosphere — all of it communicates before dialogue starts. Games need this across hundreds or thousands of environments, at consistent standard, built by a small team.
Sound & Music The invisible layer that tells you how to feel before you consciously know why. Games need all of this — and it must respond dynamically to what the player does. Hades swaps music stems mid-bar as you move between biomes; Celeste layers melodies in as the emotional stakes rise. Adaptive audio is a discipline in its own right.
World-Building The world has to feel like it existed before the story started. Environmental storytelling — objects, architecture, NPC placement — communicates without a word. Players feel the absence of depth even when they can't name it.
Games must do all of this. At professional level. That alone makes it one of the most demanding creative endeavours there is. But they don't stop here.
Act Four

Everything Film
Doesn't Have to Do

Where the argument turns — the disciplines no other medium demands

4
What Only Games Require The ceiling film doesn't have
Rules & Systems Every game is fundamentally a ruleset. The rules define what's possible and what's interesting. Bad rules create exploits — players find the path of least resistance and the whole thing collapses. Good rules feel inevitable, like they couldn't have been any other way. See Koster, A Theory of Fun.
Economy & Resources What is scarce in your game? What does the player earn, spend, lose? Economy design underpins almost every meaningful decision a player makes. Get it wrong and the player never has to make interesting choices. A game without interesting choices is just a movie you can pause.
Difficulty Curve Easy enough to learn. Hard enough to stay interesting. Threading that needle across your entire game, for players of vastly different skill levels you've never met, is genuinely one of the hardest design problems there is. Dark Souls solves it by making the wall explicit — the contract says "this is hard." Celeste solves it with an assist mode that doesn't break the achievement curve. Most games solve it by accident and lose players in Act 2. See Csikszentmihalyi's flow theory ↗.

Raph Koster, A Theory of Fun

The implication You're not building an experience. You're building a possibility space for a stranger to inhabit. That redefines the job entirely — you're architecting a space where another human being constructs their own experience. No other medium does this.
Act Five

Designing for a Stranger

The hardest part — the player you'll never meet

You're not making a world.

You're making a relationship
with a stranger.

In film, the audience receives. You control what they see and when. In games, the audience acts. You have to design for every action they might take — including the ones you never anticipated — before the game ships.

5
Player Psychology Flow, failure, contract
Flow State Csikszentmihalyi's flow — the channel between boredom and anxiety where someone is completely absorbed. Think of a game where time disappeared. Building that, for strangers, at every skill level — is the job.
Failure Design Failure must feel fair (player understands why), recoverable (they believe they can do better), and instructive (each failure gives information). A death that feels random permanently loses a player.
Hades II — the player promise

The Player
Contract

Every game makes an unspoken promise about what kind of experience this is — and players feel it the moment they pick up the controller, even if they can't name it.

The moment you break that contract — even accidentally — the trust is gone. Players don't always know why they stopped playing. It's often this.

Mobile puzzle game adds unskippable ads at hour two — after two ad-free hours.
Competitive shooter adds paid stat-boosts after marketing as skill-based.
Narrative RPG spikes to punishing difficulty in Act 3. Story players lose trust.
Cozy farming game quietly patches resource rates to push premium currency.

None of these are crashes or bugs. They're betrayals.

Dark Souls has one of the most demanding contracts in gaming: you will die constantly, the game will not explain itself, nothing will be handed to you. Millions accepted that deal — because the contract was honest from the first five minutes, and it never changed.

Write your contract in one sentence before you build anything: "This game promises the player ___." Test every major decision against it.

Dark Souls — the contract kept
Dark Souls · FromSoftware · 2011
The game that built a genre on an unbroken promise.
Act Six

Before a Single Line of Code

Most games don't die in development — they die here

6
Pre-Production The foundations most teams skip
The Core Loop What does the player DO every 30 seconds? Not the story, not the world. The literal action. In Balatro, you play cards. In Minecraft, you mine and build. In Dark Souls, you read, attack, dodge, survive. One sentence. If you can't say it, you're not ready to build.Ask: why does that action feel good? The answer is what you build first.
Design Pillars 3–5 statements every decision gets measured against. "Is this fun?" is not a pillar. "The player should always feel like they caused the outcome, good or bad" — that's a pillar. It's a filter for every decision down the line. See "The Four One-Page Design Docs You Need" — GDC ↗.
The Prototype The fastest, ugliest possible version of your core loop. No art. No music. No story. Does this mechanic actually feel like something? Most teams skip this. Most teams regret it.The prototype isn't a demo. It's a question. "Does this feel good?" is all you're asking.
Scope What is — and explicitly is not — in this game. Scope creep is the single leading cause of indie game failure — ahead of bad design, bad marketing, bad art. Scope that was never defined can never be defended.
The games that ship aren’t the ones with the best ideas.They’re the ones with the clearest foundations.
Act Seven

Making the Thing

Build · Test · Ship — and why each is harder than it sounds

7
Production The three phases
Build

Vertical slice: one polished section that proves the game is achievable. It also lies about how much remains — it takes as long as you think the whole game will take.

Alpha → Beta: all features exist, nothing polished → content complete, bug-fixing begins. Both milestones get redefined constantly.

The first 10% feels like the first 90%.Plan accordingly.

Test

Playtest strangers. Watch in silence. When someone gets confused — don't explain. The moment you open your mouth, you've destroyed the data.

In the real world, you won't be there. What you're watching is the game without you. That's the only version that matters.

The thing you're most attached tois usually first to go.

Ship

Steam page from day one. Your wishlist count before launch is your most predictive metric. Reaching 5,000–7,000 wishlists gets you into Steam's "Popular Upcoming" tier — the first organic visibility gate.

Run a demo during Steam Next Fest. Since 2024, Valve's recommendation engine treats demo players as potential customers — your demo now actively pushes your game to wishlists.

First two weeks post-launch determine long-tail visibility.The algorithm rewards launches, not quality over time.

86%
Word of Mouth  ·  86% of devs use it ↗
Content before launch.
TikTok and YouTube Shorts are where breakouts happen. A 30-second clip of something unexpected reaches more people than three months of press releases. R.E.P.O., Lethal Company, Balatro — all content before they were launches. You're not marketing a game. You're making content that happens to be about one.
26%
Live Streamers  ·  26% most-used channel ↗
Fit over follower count.
A streamer with 500 dedicated viewers in your genre is worth more than 50,000 who aren't interested. Give keys 3–4 weeks early, no pressure. The best coverage comes from streamers who discovered you — not who were briefed. Target by library, not audience size.
74%
Real-Time Community  ·  74% use Discord / Slack
Your early players are already advocates.
A Discord with 200 people who love your game before launch is worth more than any press hit. They post clips, bring friends, defend it in comment sections. Build the community during production — not after launch. The third-most used developer marketing channel in 2025 for exactly this reason.
Inbound Offers  ·  Mostly predatory
Slow down when they come to you.
Key aggregator sites charge for "visibility" to audiences that don't buy. Any offer where you pay to reach an audience you could reach yourself with a good trailer isn't a deal worth taking. Genuine publishing interest comes with advance money on the table. Everything else is the appearance of momentum.
The gap between a great prototype and a shippable game is where most projects die. Not from bad ideas — from underestimated distance between the two.
Act Eight

The Part Nobody Puts
in the Trailer

The industry is changing — and the numbers prove it

The layoffs were not a moment. They were a signal. The studio model that defined the industry for 30 years — large teams, long development cycles, publisher-backed budgets — is under structural pressure. What we're watching is a reorganisation already underway. And the numbers from 2024 make that visible in a way that's hard to dismiss.

0%

of developers reported working more than 50 hours in a single week — up from 46% the year before

Crunch has become the default, not the exception. 43% exceeded 40 hours per week on average — a figure that has risen every year since GDC began tracking it. Before the reviews, before the revenue, before anyone outside knows the game exists.

GDC State of the Industry 2025 ↗
0%

of developers were laid off in the past year — the third consecutive year of mass cuts

That's 11% of 3,000+ respondents surveyed by GDC in 2025 — professionals across every discipline and studio size. A pattern repeating every year from 2022, accelerating each time. The GDC 2025 report also found that 41% of developers were impacted by layoffs in some form — directly or through colleagues, studio closures, or cancelled projects.

GDC State of the Industry 2025 ↗
0%

of developers work solo today — up from 18% the year before

The industry is fragmenting. 56% are self-funding their own games. Indie studios now account for 32% of all respondents, while AAA studios have shrunk from 18% to 15% of the workforce — not because solo is safer, but because it's increasingly the only viable path.

GDC State of the Industry 2025 ↗
From the GDC State of the Industry Survey, published March 2025 (covering 2024): 52% work at studios using generative AI tools. 36% use it personally — up from 31% in 2023. Only 13% say it's had a positive impact — down from 21% in 2023. 30% say it's been actively harmful — up from 18%. 41% were affected by industry layoffs in 2024. 56% self-funded their games. Read the full report ↗
The Trough Every project has a point where the gap between your vision and what currently exists feels insurmountable. The prototype is embarrassing. The finish line is invisible. Every developer who has ever shipped a game went through this. Jonathan Blow nearly ran out of money making Braid. Eric Barone spent four years on Stardew Valley while working a minimum wage job. It is not a sign to stop — it's what separates the games that ship from the ones that don't.
Scope as Self-Care Setting a realistic scope is not a creative compromise. It is the only mechanism by which a game gets finished. An unfinished game helps no one — not you, not your potential players, not the people who were excited about it. Stardew Valley began as an attempt to recreate Harvest Moon with every feature the developer wanted. It shipped because he kept cutting until it was finishable by one person. That discipline is what you're building.
Community as Fuel The developers who build sustainable careers almost always have one thing in common: a genuine community around their work — players invested in the journey, not just consuming it. They carry you through the trough in ways no marketing budget can replicate.

Why players are choosing indie

The structural shift has a player-side explanation. What gamers are moving away from and what they're moving toward are not the same thing.

What players leave AAA for
What indie delivers instead
🪟

Polished graphics over everything

Better visuals with each release. Story and soul optional.

🎨

Unique artistic identity

A visual language you haven't seen before. Style as intention.

🔁

Safe franchise formulas

Sequel #7. Same loop. Known to be marketable.

Creative risk-taking

Unconventional mechanics. Storylines no committee approved.

💳

Battlepasses & paid DLC

$70 base. $30 season pass. $15 weapon skin. Repeat.

🎟️

Accessible, complete games

$15–25. Everything included. Built to be played, not monetised.

🏭

Corporate production lines

500 people building toward a release date, not a vision.

❤️

Developer passion — felt in the product

"You can feel that passion in the end product." — player, The Science Survey 2025

📣

One-way broadcast marketing

Announce. Hype. Ship. No relationship with the player.

🤝

Developer–community relationship

Alpha/beta feedback loops. Players invested in the journey, not just the product.

The Science Survey — The Rise of Indie Games, March 2025 ↗
Act Nine

The Numbers

The market has shifted — and the data tells you exactly how

9
The Odds The landscape has fundamentally shifted

More releases. Less discoverability. The same amount of player attention. The market has shifted significantly over three years — and understanding how it shifted is the prerequisite for making decisions that don't surprise you once you're inside it.

~0%

of indie games never break even on development costs

Every year, every genre, at every price point. The baseline for indie development is loss — and knowing that before you start changes how you build. In 2024, the median Steam release earned less than $400 lifetime.

Estimate derived from revenue distribution analysis — How to Market a Game (2022) ↗ · VG Insights 2024 ↗
0%

of all indie Steam revenue was captured by a handful of Triple I (high-budget indie blockbuster) titles in 2024 — up from 18% in 2020

In 2024, all other indie releases combined generated less revenue than Black Myth: Wukong on its own. The long tail of indie gaming still exists — but the money is consolidating at the top, and it's accelerating. Three years ago, 82% of indie revenue was distributed across the field. Today more than half goes to the handful of titles that crossed from indie into cultural event. This isn't bad luck. It's structure.

VG Insights — Global Indie Games Market Report 2024, p.13 ↗
0%

indie revenue growth on Steam in a single year — then the floor dropped

In 2023, indie revenue on Steam grew 77% year-over-year — the biggest single-year surge the category had ever seen. Then Black Myth: Wukong arrived in 2024 and, classified as indie, inflated the whole category. Then Steam hit $11.7 billion in 2025 — a record year — and indie revenue actually declined. AA and AAA grew. Boom. Concentration. Fragmentation. All within 36 months. The market didn't break. It restructured. And how it restructured is the entire story of what to do next.

Sensor Tower — State of Gaming 2026, p.34 ↗
What revenue concentration actually means The Triple I stat isn't a ceiling — it's a map. Revenue concentrating at the top tells you where the gravity is, not where you can't go. Lethal Company's developer had three prior games earning under $1M combined. Pocketpair shipped four titles before Palworld. The same VGI data shows second games earn 40% more on average than first games. The market rewards compounding. See VGI 2024 full report ↗.
And yet —

In 2025, the best-selling PC and console game by downloads was Battlefield 6 — one of the year's biggest AAA releases. The second and third best-selling titles were R.E.P.O. and PEAK. Two chaotic co-op indie games. Under $20. Tiny teams. Both made the Steam top 10 by monthly active users for the entire year.

The market isn't saturated.
Games that play it safe are.
Fresh ideas — executed well —
still break through.

R.E.P.O.
18M copies
PEAK
15M copies
Schedule I
8M copies

Sources: GDC State of the Industry 2025 (PDF) · VGI — Global Indie Games Market Report 2024 · Sensor Tower — State of Gaming 2026 · Shahrabi — The 2024 Indie Landscape

Act Ten

So Why Does
Anyone Do It?

The only answer that makes sense of everything above

PEAK — players climbing together

Games are the only creative medium where the audience becomes a participant.

Film makes you build everything.
Games make you build everything — plus the person who inhabits it.

10
The Payoff The co-authorship argument

You don't just make a world. You make a relationship with a stranger — one where they get to act, and your job is to make every possible action feel meaningful. Where their choices matter. Where their skill develops. Where they surprise themselves.

Every player who finishes your game made it with you. No other medium does this.

The co-authorship That's why holding all eight disciplines simultaneously, surviving the trough, and shipping deserves genuine, informed respect. And if you want to be one of those people: start. The only game that definitely never ships is the one that never starts.

The industry is changing.
The question is whether you're part of the change.

Those stats in Acts 8 and 9 aren't discouragement. They're a description of a structural shift that's already happening — and that the people building today are actively shaping. The studio model that defined the industry for 30 years is contracting. The gate is open. The people walking through it right now are writing the next chapter.

01
Industry Signal
The signals point forward.
Indie now represents 48% of all Steam revenue — equal to AA and AAA for the first time ever. That's a structural rebalancing of an entire industry. The layoffs at major studios, the consolidation, the record Steam revenues — all the same story: the old model contracting, the new model still finding its shape. You get to be part of finding it.
02
The Only Variable You Control
Persistence is the lever.
Second games earn 40% more than first games. Third-plus games earn 24% more again. Lethal Company came after three games that earned under $1M combined. Palworld came after four. These aren't lottery tickets — they're the compound interest of craft, audience, and iteration. The industry rewards people who stay long enough to learn from their misses. The barrier to entry is not talent. It's endurance.
03
The Invitation
The hardest thing is also the most meaningful.
You've seen the disciplines. The trough. The reality of the market. None of it is hidden from you now. Being genuinely part of this industry — making things that didn't exist before — is both the hardest creative act there is and the most singular. You build the world and the person who inhabits it. No other medium does that. The data doesn't make it impossible. It makes it meaningful.

Start with the core loop. One sentence. Everything else follows from that.

And if you want to understand what's worth playing before it gets discovered by everyone else — that's what Indieformer is for.

The only game that never shipsis the one that never starts.