Ten bedroom mix mistakes, and the listening test that catches each one

There are already plenty of lists of mixing mistakes. The problem with most of them is that a beginner reads "don't over-compress" and has no way to tell whether they did. So this guide is built the other way round: each mistake comes with a test you can run on your own song, in your own room, in a few minutes. Hear it first, then fix it.

Our sources are two well-used teaching pieces, by Marc Roehm and Jochen Weyer in Sound On Sound (August 2021) and by Nick Messitte for iZotope (August 2022). Where a number appears below, it comes from them. The tests are ours. They are listening habits, not lab methods.

1. Mixing without a plan

The mistake. Opening the session and tweaking whatever catches your ear. Messitte's fix is a "punch list": a written list of specific changes for each element, such as drums, bass and vocals, before you touch anything. Roehm and Weyer make the same point about balance: decide the hierarchy of the song first.

The test. Before you start, write one sentence: "The most important thing in this song is ___." If you can't finish it, you aren't ready to mix. Come back to the sentence at the end and check that the mix agrees with it.

2. A bad balance between the main parts

The mistake. Roehm and Weyer list balance as the first big mistake: a weak or overloud low end, and a vocal that is either too loud or buried. They suggest the kick and bass make "equal contributions".

The test. Turn your monitors down until you can barely hear the song. At very low volume only the loudest elements survive. If the snare is all you can hear, or the vocal drops out, your balance is off. Then mute the kick and bass in turn. If the song barely changes when one goes, it is too quiet. If the song falls apart when the other goes, it may be carrying too much.

3. Gain staging that leaves no room

The mistake. Tracks clip, or plug-ins are driven harder than you meant. Roehm and Weyer suggest starting track levels around -18 dBFS and keeping master peaks below -3 dBFS in the loudest part of the song. Messitte notes that analogue-modelled plug-ins react to the level going in, and that his own rough mix peaked at +3.5 dBFS before he fixed it.

The test. Play the loudest section and watch the master meter. If it touches 0 dBFS, pull every fader down by the same amount, or use a trim at the top of each channel, rather than putting a limiter on the master to hide it. Then listen to whether your saturation and compressors still sound the same. If they sound cleaner, they were being driven by level, not by your choices.

4. Too much bass and treble

The mistake. The "smile curve". Messitte says boosting both ends tires your ears and makes the mix translate poorly. Roehm and Weyer warn about too much top end, and about cutting fundamentals too hard with high-pass filters.

The test. Listen for twenty minutes, take a break, then play the chorus again. If it suddenly sounds harsh or boomy, your ears had adjusted to it. Then compare it at the same volume against a commercial track in your genre (see mistake 10). Matching the volume matters: a louder track almost always sounds better at first.

5. A muddy low end

The mistake. Kick, bass, low synths and pads all fighting below 100 Hz. Roehm and Weyer give a simple map: the kick is the only thing with significant content below 50 Hz, and the bass is most prominent between 60 and 100 Hz. Your genre may break this on purpose, but you should know when you are breaking it.

The test. Solo only the low instruments, briefly, and sweep a narrow EQ boost slowly through the low end to find where each one lives. If two parts boom at the same spot, give one of them priority there. Then go straight back to the full mix. Messitte warns against long stretches in solo, because you lose sight of the whole.

6. Over-compressing

The mistake. Compression that flattens the hit of drums and the life of a vocal. Roehm and Weyer advise keeping gain reduction "in the single-digit range", and say caution is needed when you approach double digits.

The test. Match the compressor's output level to its input, then toggle it on and off. A louder signal will fool you, so the levels must match. If the "off" version has more punch and you prefer it, you have gone too far. Watch the gain-reduction meter as well. If it never comes back to zero, the compressor never lets go.

7. Too much reverb, or the wrong kind

The mistake. Roehm and Weyer list reverbs and delays that are too loud, too bright or too small, and suggest cutting the high end of reverb and delay returns so they sit behind the dry sound. Messitte suggests limiting a whole mix to "only four or five reverbs".

The test. Mute every reverb and delay return, then bring them back one at a time. If the dry mix suddenly sounds clearer and you miss nothing, the effect wasn't needed. Count your reverb sends. If every track has its own, try sending several parts to one shared space.

8. Phase problems you can't see

The mistake. Messitte flags polarity and phase, mostly in multi-mic recordings such as drums. Bedroom producers meet a second version: stereo-widening tricks. Sound On Sound's Matt Houghton explains that Haas-style delays, which work in a roughly 5 to 35 ms range, can disappear or change in tone and level when the mix is summed to mono.

The test. Put a mono switch on your master and flip it during the chorus. If a part vanishes, thins out or changes tone, it has a phase problem. Phones, smart speakers and many venue systems play in mono, or close to it. For multi-mic sources, Messitte's method is to flip the polarity of one mic against another and keep whichever setting sounds fuller and reads higher on the meter.

9. Fixing everything with plug-ins

The mistake. Messitte calls stacking plug-ins without a reason "by far the biggest mistake", and adds a related one: trying to turn a sound into something it isn't. He describes a student who cut down to two plug-ins and one send effect, and whose mix improved.

The test. For each plug-in on a channel, say out loud what job it is doing. If you can't, bypass it at matched level and see whether you miss it. If a sound needs six plug-ins to work, the real fix is often a different sample, a different part or a re-recorded take.

10. Mixing without a reference

The mistake. Messitte's tenth point. Without a reference your ears drift, and you make decisions that only work in your room.

The test. Choose two released tracks you like in the same genre. Turn them down until they sound as loud as your mix, because they will be mastered louder. Then switch between them and your mix every few minutes. Compare four things in turn: vocal level, how loud the low end is, brightness and width. Don't try to copy the reference. Use it to notice when your mix has drifted.

A word on loudness

Many bedroom mixes go wrong because the producer is chasing commercial loudness before the mix is finished. Spotify normalises playback to -14 LUFS and turns louder masters down. So a mix pushed louder at the end won't play louder there, and it may just be more squashed. Our loudness guide sets out what each platform documents.

What we think

The ten mistakes come down to two habits. The first is trusting one listening position. Low volume, mono, a break and a level-matched reference are four different ways of hearing the same track, and each catches something the others miss. None of them costs anything.

The second is reaching for tools before ears. Most problems on this list come from adding something: another plug-in, another reverb, more compression. The quickest fix is often to remove it. We would rather a beginner learn to bypass well than learn ten new plug-ins.

What this list can't do is let you hear the difference. Mixing is learned by ear, and the best next step is to run these tests on one of your own songs, keep a before-and-after bounce, and listen to them a day later. Many of these mistakes only show up once you've had a break.