There's a test I run on every track before I let myself fall in love with it. I pull up the session, mute everything except one instrument and the vocal, and play it from the top. Just the chords and the voice. No drums, no bass, no reverb tail doing emotional work the song hasn't earned.
If it still moves me, I've got something. If it falls flat, I don't have a song. I have a really good-sounding loop with a melody sitting on top of it.
That distinction took me years to hear, and once I heard it I couldn't un-hear it.
Production is great at hiding a weak song
Modern production is absurdly good. A stock plugin chain can make almost anything sound expensive. Sidechain the pad, drench the vocal in a long plate, drop a sub that rattles the room, and a four-note idea suddenly feels like it has stakes. The low end hits, the high end sparkles, the arrangement breathes in all the right places.
None of that is songwriting. It's atmosphere. And atmosphere is seductive because it gives you the feeling of having written something without the harder work of actually writing it.
I've shipped tracks I knew, somewhere underneath, were thin. The production was carrying them. They did fine. But they didn't stick, not for listeners, and honestly not for me. Six months later I couldn't hum them. The good ones, the ones people still mention, all pass the one-instrument test. Every time.
Why this trap is worse than ever
If you came up as a producer first and a writer second (which is most of us now), your instinct when a track feels flat is to produce harder. Add a layer. Try a new vocal chain. Automate a filter sweep into the chorus. Reach for the thing you're good at.
Sometimes that's right. Often it's avoidance. You're decorating a room with no foundation under it.
The uncomfortable truth is that production skill and songwriting skill are not the same muscle, and being elite at one can let you coast on a weakness in the other for a long time. The DAW will happily help you avoid the question "is this actually a good song" for your entire career if you let it.
What the one-instrument test actually checks
When you strip a track to a piano or a guitar and a voice, you're exposing the parts that have to be load-bearing:
The melody. Is the topline memorable on its own, or does it only work because the beat is telling you where the energy is? A strong melody has a shape you can sing walking down the street with no music playing.
The chords and how they move. A lush pad can disguise a progression that doesn't go anywhere. Naked, you hear whether the harmony has any tension and release, or whether it's just four nice chords on a loop.
The lyric and the phrasing. Reverb and delay smear over clumsy phrasing. Dry and exposed, you hear every syllable that doesn't sit right, every line that's filler.
The arrangement of the actual song (verse, lift, chorus) as opposed to the arrangement of the production. Does the chorus feel like a chorus when it's just voice and chords, or does it only feel like one because the drums get bigger?
If the song holds up naked, production becomes what it's supposed to be: amplification. You're making a strong thing stronger instead of using sound design to fake conviction the writing doesn't have.
How I work it back into the process
I'm not saying write everything on an acoustic guitar by candlelight. I produce. I love producing. But I've changed the order of operations.
Sketch the song before I dress it. Even if I started from a beat, at some point early I force myself to play the core idea on one instrument and sing the melody. If I can't, I keep writing before I keep producing.
Keep a "naked bounce." When a track is mostly arranged, I'll mute down to voice and one harmonic instrument and bounce thirty seconds of just that. Listening back on my phone the next day tells me everything. If the naked version is boring, no mix is going to save it.
Separate the two jobs in my head. Writing day and producing day aren't the same workout. On a writing day the goal is a song that survives being stripped. On a producing day the goal is to make that song hit. Mixing them is how thin songs get expensive clothes and ship anyway.
The harder, better question
It's easy to ask "does this sound good." Your ears get a hit of dopamine, the meters look healthy, and you move on. The harder question is "is there a song here, or just a really good-sounding production of not much."
Most of the tracks I'm proudest of were boring to produce, because the writing was already doing the heavy lifting and I just had to stay out of its way. The ones that were the most fun to produce (endless layers, clever automation, hours of sound design) were usually the ones papering over a song that was never quite there.
Strip it down. If the voice and one instrument can carry it, you've written something. Everything after that is just making it louder.