Every sound team ships its work into a hostile environment: a phone that is probably on silent, in a room where the player is probably not alone, with a settings screen whose most-visited row is the one that turns the whole mix off. Players do not mute games because they hate sound. They mute games because most game audio takes more than it gives — it repeats until it grates, it startles at the wrong moment, it announces to a quiet train carriage that you are playing a game, and in exchange it tells you almost nothing you could not see. Designing audio that players keep on is therefore not a mixing problem first. It is a contract with three clauses: the game must work in silence, the sound must never punish, and loudness must be spent like money. Break any clause and the player’s thumb finds the toggle.
Why the mute switch wins by default
On a console or a PC, the player has arranged the room around the game. On a phone, the game has been smuggled into someone else’s room — a commute, an office lull, the last twenty minutes before sleep next to a partner who is already asleep. In those settings the social cost of any sound is real and immediate, and the polite default is silence. That is the honest starting point for mobile sound design: you are not scoring a film for a captive audience, you are asking for a privilege the context has already revoked. A mix that assumes it will be heard is designing for the minority case.
Even when the context allows sound, three failure modes do the muting for you. Repetition is the big one: a coin blip that was charming the first fifty times becomes sandpaper by the five-hundredth, because game sounds fire at a frequency no film or song ever asks a sound to survive. Startle is the second: title music that blasts at full volume the instant the app opens, or a stinger mixed hotter than everything around it, teaches the player that the safest volume is zero. And emptiness is the quiet third: sound that decorates but never informs. If the audio tells you nothing the screen does not already say, muting it costs nothing — and players are very good at noticing what costs nothing.
Clause one: the game must speak fluent silence
Before any craft about what sound to make, there is a structural rule about what sound may carry: never let audio be the only channel for information the player needs. If the incoming attack is signalled only by a whoosh, the muted player takes hidden damage and experiences it as unfairness. If the “correct move” feedback is only a chime, the muted player is playing a different, worse game. Every load-bearing cue needs a visual or haptic twin — the flash, the shake, the tick of a counter — so that the muted game is the same game, just quieter. This is the exact logic of the greyscale test we described in designing for colorblind players, transposed to another sense: colour may reinforce but never solely carry, and sound may reinforce but never solely carry. Run the mute test the way you would run the greyscale test — play a full session with the volume at zero and note every moment you were less informed. Each one is a bug.
There is one honest exception: games where sound is the mechanic. A rhythm game muted is not a quieter rhythm game; it is a broken one. That is a legitimate design — but it is a promise the store page and the first run must make explicitly, because a player who discovers mid-commute that the game is unplayable on silent will not blame themselves. For everything else — puzzles, strategy, arcade action — the standard is stricter: sound is a reinforcement layer on a game that is already complete without it. We hold our own games to that line as a stated design intention in our accessibility guide, for the players who cannot hear the mix as much as for the ones who choose not to.
Clause two: never punish with sound
The fastest way to earn a permanent mute is to make the player flinch once. The offenders are boringly consistent. Music at full blast on the first frame of the first launch, before the player has had any chance to set a volume. A mix mastered for headphones that clips and shrieks through a phone’s tiny speaker. A failure buzzer that treats a wrong answer like an alarm — punishing the player twice for one mistake. And the sound that plays when the player explicitly asked for silence: a game that ignores the phone’s ringer switch has announced that its preferences outrank yours, and the player will correct that arrangement exactly once. The platforms already encode the polite behaviour — iOS lets a game’s audio session respect the silent switch, and Android expects apps to request and yield audio focus rather than talk over whatever was playing — and honouring those conventions is table stakes, not polish.
Two more courtesies belong in this clause. First, a game should go quiet when it leaves the foreground — our games stop their audio on backgrounding rather than pretending to be music apps, because a game that keeps making noise after you switch away is a game making noise nobody asked for. Second, an honest limit from our own house: our free tiers are ad-supported, and ad creatives bring their own audio that we do not author. We can control our mix; we control the ads’ far less. That is a real seam in the experience, and pretending otherwise would be exactly the kind of claim this blog exists to avoid.
Clause three: spend loudness like a budget
The craft heart of unmutable audio is a single trade-off: the loudness a sound is allowed to spend should be inversely proportional to how often it fires. A sound the player hears every few seconds — a tile placement, a pencil mark, a footstep — must sit near the floor of the mix: short, soft, and low in drama, because it will be heard thousands of times. A sound the player hears a few times a session — clearing a stage, a streak milestone — can afford brightness. And the full fanfare, the one that spikes the mix and demands attention, must be rationed to genuinely rare moments, or it stops meaning anything and starts costing patience. Most fatiguing mixes are not too loud overall; they are flat — every event shouting at the same volume, so nothing is information and everything is noise.
The budget buys you the other half of the repetition problem too. A sound that fires constantly must not only be quiet — it must vary. The standard toolkit is old and effective: randomise the pitch a little on each play, rotate through a handful of recorded takes (round-robin) instead of replaying one file, and let the mix duck — briefly lower the music — when an important effect needs the foreground. None of this is audible as a technique; that is the point. The player never thinks “nice pitch randomisation.” They just fail to get tired of a sound they have heard two thousand times, and the toggle stays untouched.
What earns a place in the mix at all
With the contract in place, the question for each individual sound becomes blunt: what does this sound know that the player wants to know? The best game sounds are information first — they confirm the input landed, they distinguish the good outcome from the bad one by timbre, they mark state changes the eye might miss because it was looking elsewhere. Timing does most of the work: a modest sound landing on the exact frame of impact reads as more satisfying than a rich one landing a beat late, which is the same lesson as the rest of the feedback stack we walked through in the anatomy of ‘juice’ — audio is one layer of that stack, and it obeys the same rule that layers amplify meaning or they are noise.
Music deserves its own honest sentence. Looping background music is the single most-muted element in games, because it is the element most likely to overstay — a thirty-second loop over a forty-minute session is a promise of irritation. The disciplined options are all respectable: write longer and sparser than feels natural, thin the arrangement so it sits under thought rather than on top of it, or — the choice quiet puzzle games keep making — ship very little music at all and let a calm ambience and clean effects carry the room tone. Silence is a mixing decision, and for a game a player lives in daily, it is often the generous one.
The honest close: the review you never hear
Nobody writes a store review praising a game they did not mute. That is what makes this corner of the craft strange: done well, it produces no visible reaction at all — just a toggle that never gets flipped, a session that stays a little more legible, a mix that pays a small dividend of information and charges almost nothing in fatigue. The three clauses are cheap to honour and expensive to ignore: build the game to speak fluent silence, never make the speaker a weapon, and spend loudness like a budget with the fanfares saved for moments that deserve them. Sound players keep on is sound that pays rent. The rest of the WizusLabs Engineering blog digs into the neighbouring layers of the same craft — the feedback stack, the first minute, the screens everyone can read.
Notes
This is a first-hand craft piece, and it deliberately contains no usage statistics. In particular, we have not cited the widely circulated figures about what share of mobile players play muted: the numbers that make the rounds come from vendor studies whose methodology we cannot verify, and quoting a number we cannot stand behind would undercut the point of the piece. Where the essay describes player behaviour, it is describing a phenomenon we design around, not a measured rate. Our own games are free and ad-supported, with a single optional Pro purchase that removes ads — ad creatives carry audio we do not author, which is the honest limit named above.
Sources
- Apple Developer Documentation — AVAudioSession (audio session categories, including behaviour that respects the device’s silent switch).
- Android Developers — Manage audio focus (the convention that apps request and yield audio focus rather than play over other audio).
These support only the platform-behaviour statements; the post makes no statistical claims.
Keep reading: all posts on the WizusLabs Engineering blog.