Half of every indie game we playtested in 2026 shipped with a broken tutorial.
That is not hyperbole. We Playtest Games runs professional playtesting for indie developers. Across the games we processed this year, 505 of them were independently flagged for poor onboarding by at least two of our three AI classifiers. That is the same conservative, cross-checked methodology used throughout the book. It is roughly 46% of everything we tested. The classifiers produced 1,306 distinct observations of onboarding failure across those games, every single one anchored in a real thing a real playtester said during a real session.
That makes poor onboarding the single most common failure in indie games right now. Not performance. Not difficulty. Not balance. The first fifteen minutes, where you teach someone what your game is.
This post is the long-form version of Chapter 1 of our book Top 20 Indie Game Development Mistakes: What Your Indie Game Is Getting Wrong (And How To Fix It). If you like this, the full book covers the twenty most common mistakes in the same depth.
Why onboarding keeps failing
You know your game. You made it. You have pressed E to interact ten thousand times. You cannot remember what it felt like to not know the controls, because you never didn’t know them.
This is called the curse of knowledge, and it quietly ruins more tutorials than any other factor. When you look at your own onboarding, you skim past it, because to you it contains no new information. You forget that the player is missing literally everything: the controls, the goal, the loop, the UI, and the world rules.
The second cause is a fundamental misunderstanding about what a tutorial is. Many developers treat onboarding as a one-time information dump. A wall of text. A controls screen. A popup that says “Press SHIFT to sprint.” The player reads it once, nods, and never performs the action under guidance. Five minutes later they are stuck against a wall because they forgot sprint existed.
A tutorial that tells you how to play is not a tutorial. It is a manual. Manuals are fine. They are not tutorials. A tutorial makes you do the thing.
The third cause is structural. You wrote the tutorial early in development when the game was simpler. New systems got added. The tutorial never got updated to cover them. By launch, your tutorial teaches a game that no longer exists.
The patterns we saw, over and over
When you aggregate hundreds of playtest transcripts, the failure modes are not infinite. They cluster into a handful of specific, recognizable shapes. Here are the ones we saw again and again.
Pattern 1: Drop the player in with no tutorial at all
The most common failure. The game just starts. No prompt. No intro. No controls screen. The player is alone.
The sentence we heard over and over was some variant of “I have no idea what I’m supposed to do.” Sometimes more resigned: “There’s been no tutorial, and I guess there isn’t going to be one since I’ve already started playing and shooting enemies.” Sometimes more blunt: “I didn’t see any controls, teach me sensei.” In one session the tester said, “There was no introduction, they just dropped me into the game like I would know the controls.” In another: “I was dropped in here instantly with no instructions and everything, only the Steam page.” In a third: “I guess there’s no tutorial or anything, so I’m gonna have to figure this out myself.”
The assumption behind this pattern is usually “it’s intuitive, players will figure it out.” They will not. In our data, the games that did best were the ones that assumed the opposite: players know nothing, and every bit of knowledge needs to be handed to them explicitly.
Pattern 2: The wall of text
The game does have a tutorial. It is a static screen with nine bullet points. The player reads it once, dismisses it, and remembers none of it.
Actual quotes from testers staring at these screens. “Okay, that’s a lot to remember. I really don’t think I can remember all of that.” “20 pages of tutorial with just text is a bit too much to ask from a deck builder game.” “I read the instructions, but it was just word-bombing at me.” “There will be a lot of reading here.” One player told us the opening instructions were “very forgettable” because the tutorial was “not properly fed progressively”, and that dragging cards into their hand (a core interaction) was never specified at all.
There is a reason this fails. Reading and playing use different mental modes. You cannot absorb a list of nine key bindings through a screen you just want to dismiss to get to the game. You need to do each one, in a low-stakes context, immediately after being shown it. Fifteen seconds of guided practice beats fifteen minutes of reading.
Pattern 3: Core abilities the player never knew existed
This is the pattern that hurts the most, because it means you shipped content that the player literally cannot access. Dash, double jump, stealth, drift, block, parry, heavy attack, sprint. Abilities that are in the game but never in the tutorial.
A sampling of what this sounded like in our transcripts. “We’re 10 minutes into gameplay and I only just figured out I have a dash.” One tester, 34 minutes into a fighting game, after finally working out they could dodge vertically: “If I knew about this sooner it would have helped. Should probably make a tutorial where you explain each move, especially the dodging up and down.” Another, 27 minutes into a racing game, after discovering the drift button: “Oh my god, the only button I wasn’t pressing was space. Probably because the game didn’t also have a tutorial.” “I had no idea stealth was a feature until I stumbled into it.” “I didn’t know about the slam/pound move until I went into the rebind menu.” “I randomly click on my keyboard and it’s an emoji system. You guys didn’t tell us about the emoji system.” “I wish there was some warning saying that I can double jump or triple jump at all. It’s not said in the tutorial.”
When abilities ship in the game but are never surfaced in the game, they may as well not exist. The loading screen tip does not count. The rebind menu does not count. The Steam page does not count. If the player does not use the ability under guidance during the tutorial, assume they will never use it.
Pattern 4: Tutorials that tell but never let you do
The game shows you a prompt. The prompt has text. You dismiss the prompt. You never performed the action. This is surprisingly common and surprisingly lethal.
The playtesters who spotted this articulated the fix better than we can. “When you’re explaining ‘move with WASD,’ let the player actually move. Right now the tutorial tells you things but doesn’t let you do them at the same time.” “The tutorial is just written text. It doesn’t let me actually try anything. I wish I could test it during a tutorial.” “The tutorial should explain each game mechanic as you go and not state everything at once then throw the player in.” “You could have just showed me that interaction and I would have understood immediately instead of all that text.”
The fix here is not to add more text. It is to remove text and replace it with guided practice. Show the prompt. Do not let the game advance until the player performs the action. Dismiss the prompt. Move to the next one.
Pattern 5: The tutorial only covers combat, not the loop
This is the failure mode for strategy games, sims, and roguelikes. The tutorial teaches you how to fight a battle. Then the real game is about managing an economy, running a factory, or planning a campaign. None of which were mentioned.
“The tutorial was fine great, and then it just threw me into this campaign which I have no clue what’s going on.” “Tutorial does not extend past battle so players are left with no explanation on what their goals are.” “I know that there was some kind of tutorial at the beginning, but not really explained the gameplay loop.” In one session, the tester finished combat training, entered story mode, and immediately hit a mechanic that had not been introduced at all: “Suddenly right on the story we have to battle without proper tutorial. Why do I suddenly have to retire my card? There is no explanation. I’m completely confused here.”
The fix is boring but necessary: cover the full loop. If your game has a run, a day cycle, a base, a research tree, or a meta-progression, every one of those systems needs the same guided introduction the combat got. Otherwise you have taught the player half a game.
Pattern 6: Broken tutorials
Some tutorials simply do not work. The prompt will not dismiss. The NPC will not trigger. The tutorial soft-locks the player. This is the worst possible first impression, because the player cannot even reach the game you wanted them to see.
We logged tutorials that hung on a loading screen for seven minutes with no cancel button, forcing the player to relaunch the game. Tutorials that trapped the player in an immovable state until they reloaded the tab. Tutorial boxes that refused to dismiss no matter what the player did: “Whatever I do, it doesn’t go away.” Tutorials where the player picked up the starting weapon and could not fire it. Tutorials that vanished mid-lesson: “You dropped the tutorial. Now I don’t know what you want me to do anymore.” Tutorials broken badly enough that the player gave up on learning entirely: “I wish the tutorial didn’t break. I think I know enough. YOLO.”
One playtester summed up the stakes: “If somebody cannot complete a tutorial they will probably quit the game straight away, they will not give it another chance.”
Test your tutorial on a fresh install. On a slow machine. With no controller plugged in. With a controller plugged in. The tutorial is the one piece of your game that every single player experiences. It must be rock solid.
Pattern 7: The tutorial and the real game play differently
The tutorial is in a sanitized environment. The real game is not. The player learns a rule in the tutorial that is false outside of it.
We saw this in several specific shapes. A safety cushion in the tutorial removed from the real game: “The tutorial was fine with me hitting the walls with three but then that obviously misled me to going into the full game.” A tutorial level stripped of the hazards that define the real game: “I think how this game is presented in tutorial doesn’t really represent the game how it actually is. The tutorial should show everything you can meet in the game.” A prologue that removed control from the player entirely and taught nothing about how the game actually felt to play: “At the beginning, it felt like a visual story where I wasn’t actually doing anything.” A tutorial that taught one weapon and then sprang a completely different one on the player with no introduction: “You’ve suddenly gone from shooting in the first one to using a sword in the second one.”
If the tutorial lies, even by omission, you are setting the player up to fail the first encounter of the real game, and to blame the game for what they perceive as unfairness.
Pattern 8: Unrecoverable tutorials
The tutorial runs once. The player misses something. They cannot get it back.
“I had to skip the tutorial and there’s no way for me to get back into it.” “On running the game for a second time, I don’t see any information on what I should be doing here.” “All these little tutorials that pop up should be re-accessible from the menu or some other way, just a list of all tutorials so anyone can reread them.” In one session, the player accidentally exited the tutorial, spent sixteen minutes trying to get back into it, and finally said, “That’s what I needed to begin with.” In another, the player finished an entire session and only realized at the end: “I didn’t even see that there was a tutorial. Maybe I should have started with that.”
At minimum, put a “How to Play” entry in the pause menu. It is an afternoon of work. It is the single highest-leverage change you can make to your onboarding.
How to spot this in your own game
Hand the controller to someone who has not played your game. Do not speak. Do not guide. Set a timer.
After five minutes, ask them: what is the goal of this game? What can your character do? How do you know when you have won or lost?
If they cannot answer, your onboarding is broken. It does not matter how polished the rest of the game is. You have lost them already.
A few more specific checks.
- Does the player perform every core action at least once, under guidance, before the tutorial ends? Not “can they, theoretically?” Did they, in a real playtest, actually do it?
- Can the player re-open the tutorial from the pause menu?
- If the tutorial is skipped, can the player still succeed? If they did not skip it, can they still get stuck?
- Are there any abilities or mechanics your player might reasonably use but that are never shown on screen? For each one: where is it introduced? With what prompt? On what trigger? Write the answer down. If you cannot, it is not introduced.
- Is there a path through your tutorial where it soft-locks? Where it breaks? Where the player accidentally dismisses it and cannot get back? Test each one.
The fix
Teach by doing, not reading. Replace static text walls with contextual prompts that fire when the player actually needs them. The prompt appears. The player performs the action. The prompt goes away. If the player does not perform the action, the game waits.
Introduce one mechanic at a time. Give each new verb a quiet space to be used before the next one arrives. If your combat system has five abilities, do not teach all five in the first fight. Teach one. Give the player an encounter that rewards using it. Then layer the next one on top. One playtester put it cleanly: “You might have ramped up the tools too quickly which didn’t give me a chance to understand how each one worked.”
Explicitly teach every input, including the obvious ones. WASD. Sprint. Interact. Crouch. Jump. We saw dozens of cases where testers genuinely did not know they could sprint, or did not realize the game used WASD instead of arrow keys, or could not figure out which key opened the inventory. “Obvious” is a word that has meaning only to people who have already played the game.
Make the tutorial recoverable. Add a “How to Play” entry to the pause menu that lets the player re-read any lesson they missed. Make sure the tutorial cannot soft-lock. Make sure skipping it does not leave the player permanently unable to progress. Make sure the tutorial can run a second time on a fresh save, because one tester told us that “on running the game for a second time, I don’t see any information on what I should be doing here.”
Teach the full loop, not just the controls. If your game is a roguelike, the player needs to understand the run, the meta-progression, and the win condition. If your game is a sim, they need to understand the day cycle, the economy, and what ends the game. Controls alone are not enough.
Make sure the tutorial and the real game play the same way. Do not make the tutorial easier in ways the real game is not. Do not demonstrate a mechanic in a sanitized environment that is missing the hazards it is designed to counter. A misleading tutorial is worse than no tutorial, because it actively teaches the player the wrong thing.
Test the tutorial on a real stranger. Not your partner. Not your friend who has already watched you play. Someone who has never seen your game. Watch them silently. Bite the inside of your cheek and do not explain. Every time you have the urge to say something aloud (“oh, you have to press E there”), write that moment down. That is a missing tutorial prompt.
What “fixed” looks like
A first-time player, five minutes into your game, can describe the goal in one sentence. They can list the main things their character can do. They have used the core verbs of the game under no pressure, and they know what those verbs feel like. When the real game begins and the difficulty ramps up, they are not scrambling to remember which button dashes. They are playing.
The tutorial has disappeared from their awareness. They no longer feel like they are being taught. They feel like they are playing your game.
That is the bar.
Get the full book
This is one of twenty chapters from our book Top 20 Indie Game Development Mistakes: What Your Indie Game Is Getting Wrong (And How To Fix It). The other nineteen mistakes (unclear objectives, stale hints, broken navigation, overtuned difficulty spikes, audio problems, and more) are covered at the same depth, with the same evidence base of real playtests and real players.
You can download the full book as a free PDF at weplaytestgames.com/book.
If you want to see your own game in our data with a passing grade on every one of these twenty checks, that is what we do. A one-hour professional playtest costs $20, and you get the transcript, the video, and the report. Ship something you would not be embarrassed to hand to a stranger.