Game Onboarding Playtest: How to Test Your First 10 Minutes

Game onboarding is everything a new player goes through from launch to the moment they play on their own: the title screen, the first controls, the tutorial, the first win and the first real choice. Developers call this the first time user experience, or FTUE. You cannot judge it yourself, because you already know how the game works. The only way to see it is to watch people who have never played. This guide gives you a repeatable onboarding playtest: who to recruit, what to watch in the first 10 minutes, what to ask afterwards, and how to fix what you find.

Why onboarding needs its own playtest

Most playtests try to answer "is this fun?" An onboarding playtest asks a narrower question: "can a new player get to the fun without help?" Those are different problems with different fixes.

A great core loop does not matter if players quit before they reach it. Plants vs. Zombies creator George Fan said at GDC 2012 that if the tutorial hadn't worked, the rest of the game "wouldn't have mattered" for getting casual players through it. Demos make this even sharper. A Steam demo or itch.io build is often a player's only first impression, so the first few minutes carry the whole decision to wishlist or close the window. If you are preparing a demo, our guide to Steam Next Fest demo feedback covers the timing side.

What to test: the onboarding moments

Write out your first 10 minutes as a list of moments. A version adapted from the first-10-minutes audit RoClicks recommends for Roblox games works for most genres:

1. Launch to first input. Splash screens, logos, menus, settings prompts, account or language pickers.
2. First control. The first moment the player moves, clicks or acts.
3. First feedback. The first time the game tells them an action worked or failed.
4. Each tutorial step. Every prompt, popup or forced action.
5. First win. The first goal completed, enemy beaten or puzzle solved.
6. First free choice. The first decision they make with no prompt telling them what to do.
7. First loss. The first death, fail state or setback, and what happens next.

For each moment, note what the player has to know to get through it. Every item on that list is something your onboarding has to teach or something players have to guess.

Who to recruit for an onboarding test

The single rule: testers must be new to the game. UserTesting's FTUE research guide suggests screening with a question as blunt as "Have you played [your game's title]?" Anyone who has seen an earlier build, watched a devlog or read your design doc is out.

Beyond that:

• Match the audience. People who play your genre bring the right expectations. A strategy fan and a casual mobile player will stumble in different places, and you want the stumbles your real players will have.
• Fresh testers every round. Once someone has learned your controls, they can never un-learn them. Re-use them for later-game tests, not onboarding.
• Small groups, several rounds. A handful of new players per round is usually enough to find the biggest blockers. Fix them, then test again with new people. Our guide on how many playtesters you need covers numbers by goal.

Finding a steady supply of fresh players is the hard part. This is where posting a playtest publicly helps: on Get Playtests, players browse and pick the games they want to play, so each one arrives without any prior exposure. Posting is free and needs no credit card. You can post your game and ask reviewers to record their first session.

How to run the onboarding playtest

Set up

• Fresh start every time. Wipe saves, reset settings, clear any unlocks. If the game remembers a returning player, your test is broken.
• The real first launch. Start from the actual launch screen, not a debug scene.
• Record everything. Screen plus voice if they agree. You will miss things live.
• One line of brief. "Play this like you just downloaded it. I won't help. Please say what you're thinking out loud." That's it. Do not explain the controls or the goal. That is what you are testing.

Watch, don't help

The hardest part of an onboarding test is staying quiet. Every hint you give hides a problem your real players will hit alone. If a tester gets stuck, count to ten, then ask "What are you trying to do?" rather than pointing at the answer. Egmatic's playtest script puts it simply: prompt attention, never give answers.

For a step-by-step session script with timings, see how to run a playtest. For tests you can't watch in person, remote playtesting covers recorded and live options.

What to log

Keep a timestamped log of these signals:

• Time to first meaningful input. How long before they are actually playing?
• Pauses. Any moment the player stops and looks around.
• Misreads. They try something the game does not support, or do the opposite of what a prompt said.
• Skipped text. Tutorial popups they close without reading. If it happens to most testers, the text is not doing its job.
• Backtracking. Going back into menus or retracing steps to figure out what they missed.
• First self-directed action. When they stop following prompts and start making their own decisions.
• Exit point. If they quit, exactly where and what had just happened.

The dev.to write-up "A practical first-minute playtest for browser games" suggests timing three moments in particular: the first meaningful input, the moment the player recognizes what happened, and the first decision that changes their next attempt. Those three timestamps tell you whether your opening teaches cause and effect fast enough.

Questions to ask after the first 10 minutes

Ask after play, never during, so you test the game and not your explanations. Keep it about understanding, not approval.

• What did you think the goal was?
• What told you an action had worked or failed?
• What did you want to try next?
• Was there a moment you weren't sure what to do? What happened right before it?
• Was there anything on screen you didn't understand or ignored?
• Did you ever feel the game was teaching you too slowly?
• If you were at home, would you keep playing right now? Why or why not?

The first three come from the dev.to first-minute playtest guide, and they are the most useful because a wrong answer points straight at the moment to fix. If a player says the goal was "collect stuff" and your goal is "defend the gate," your onboarding did not communicate it. For a bigger question bank, see playtest survey questions.

Common onboarding problems and how to fix them

These are the patterns that show up again and again, with fixes from people who have shipped successful onboarding.

Too much text

Players skip walls of text. George Fan's rule for Plants vs. Zombies was a maximum of eight words on screen at once. Rewrite each prompt until it is a short instruction ("Click a seed packet to pick it up") instead of an explanation.

Reading instead of doing

Fan's second tip was that players learn best by doing, not reading. If a tester skips a tutorial popup and still gets it right, that popup was not needed. If they read it and still get it wrong, the action needs to be taught by a guided first try, not a description.

Teaching everything at once

Apple's onboarding guidance for games says to teach one step at a time and let players show competency before moving on to advanced objectives. Plants vs. Zombies held back peripheral systems, with money only arriving after the first 10 levels. If your testers forget the first mechanic by the time the third is explained, spread them out.

Slow start

Apple also recommends avoiding splash screens, menus, agreements and disclaimers when a new player first opens the game, and presenting non-essentials like ratings prompts and notifications after onboarding. Time from launch to first input is one of the easiest numbers to improve.

No safe space to fail

If a new player's first death sends them back a long way or punishes them hard, many quit. Watch what testers do right after their first loss. If they sigh and close the game, soften the early failure.

Babying experienced players

Fan used adaptive messaging: hints only appeared for players who were doing the wrong thing, so players who got it right could feel smart. Apple suggests letting players skip tutorials if they already know the basics. If genre veterans in your test look bored, give them a way to move faster.

The first free choice is a trap

Look one step past the tutorial. The first decision a player makes alone should not lead to a likely loss. If testers make their first independent choice and immediately get punished, the onboarding ended too early or the difficulty spikes too soon.

After the test: fix one thing at a time

1. Rank the problems. List every moment more than one tester struggled with, earliest first. Early problems hide later ones, because people who quit at minute two never see minute five.
2. Fix the earliest blocker first. Change one thing, such as a prompt, a cue's timing or the order of two steps.
3. Retest with new players. The same people will breeze through because they already know it.
4. Track the numbers across rounds. Time to first input, how many testers reached the first free choice, and where people quit. If those improve, your onboarding is improving.

For live games, Apple suggests comparing retention between day one and day seven to judge onboarding changes. Before launch, round-over-round playtest results are your version of that. When you are ready to test whether players come back for a second session, how to know if your game is fun covers return signals.

Onboarding playtest checklist

• Testers have never played or seen the game.
• Testers play the genre.
• Saves wiped and launched from the real start screen.
• Session recorded with voice.
• One-line brief, no control or goal explanation.
• Timestamped log of pauses, misreads, skipped text, first free action and exit point.
• Debrief questions asked after play, about understanding.
• Problems ranked earliest first.
• One fix per round, retested with fresh players.

FAQ

What does FTUE stand for in games?

First time user experience. It covers a new player's journey from launching the game to playing comfortably on their own, including tutorials, menus and the first session.

How long should an onboarding playtest be?

Long enough to reach the first free choice and the first loss. For most games that's 10 to 20 minutes. Stop sooner if your test question is only about the opening.

Should I explain the controls before an onboarding test?

No. The test is whether your game explains them. Give one line of brief and let the game do the rest.

Can I test onboarding remotely?

Yes. Ask for a screen recording with commentary from the first launch, and send the debrief questions after the recording ends. Unmoderated recordings are often more honest, because nobody is watching the player struggle.