I put a question to a songwriting community on Reddit recently: what’s the part of writing you’ve never found a satisfying answer to? I expected chords, melody, structure. Almost every answer circled the same thing instead, and it wasn’t technique.
It was judgement. How do I know if this idea is worth writing about? How do I know if I’m actually improving? How do I know whether this verse and chorus belong together? How do I know if a lyric is vulnerable or just embarrassing? How do I know when a song is finished?
The more of those I read, the more I started to think we might have the hard part of songwriting backwards. Learning to write is difficult, but it’s the kind of difficult you can study your way out of. Learning to know when you’ve written something good is the part that never fully resolves.
Beginners have easier problems
Not easier in the sense that starting out is easy. It’s usually bewildering. But a beginner’s problems tend to be visible ones. The lyric doesn’t fit the melody. The verse runs eight minutes. Every line has seventeen syllables except the one with four. The chorus arrives with less energy than the thing that came before it.
You can point at problems like that, and because you can point at them, you can learn to fix them. You study structure. You find out that a chorus doesn’t become a chorus just because you sing it louder. You write enough flat second verses to realise that restating the first verse with new furniture isn’t character development. You improve.
But then the problems get harder to see. The lyric scans perfectly and still feels dead. The chorus lifts, but a little too predictably. You’ve got three lines that all work. The song is technically finished and you have no idea whether it’s finished or you’ve just run out of ideas. You solved the obvious problems. Welcome to the worse ones.
Good and bad stop being useful categories
As you improve, the decisions get finer, and the easy labels stop helping. Is this lyric simple or boring? Is the melody restrained or forgettable? Is the repetition hypnotic or lazy? Is that cracked vocal giving the song character, or is it just a rough take you’re attached to because it was the first one?
The honest answer is that it depends. A line can be painfully obvious in one song and devastating in another. A chorus can repeat the same phrase eight times and feel like a trance, while somewhere else two repeats have already outstayed their welcome.
There’s no law that a verse must be sixteen bars, or a title must land in the chorus, or a bridge must exist. There are conventions, and they’re worth knowing because they tell you what listeners have been trained to expect. But good songs break them constantly.
So the better you get, the less it helps to ask whether a choice is technically correct. The only question left is whether it works, and that one is much harder to answer.
More options aren’t better decisions
When you’re starting out, the first line you think of is often the line, not because it’s brilliant but because it’s the only one you have. With experience you can generate ten. You can try a different melody, move the chorus earlier, cut the pre-chorus, flip the point of view, rewrite the second verse, strip it all back, add a bridge, lose the bridge, put the bridge back because maybe you were right the first time.
That sounds like progress, and it is. But more options isn’t the same as better decisions. Skill widens the range of directions you could take a song, which means the question stops being what can I do here and becomes which of these should I do. That might be the real line between technique and judgement. Technique gives you options. Judgement tells you which one the song actually needs.
Getting better doesn’t guarantee better songs
You can objectively improve at songwriting. You can get better at melody, structure, rhyme, prosody, harmony, storytelling, tension and release. These aren’t imaginary skills, and you can study and drill every one of them until you have real control.
Unfortunately, none of that guarantees a better song. We’ve all heard technically accomplished writing that leaves us completely cold, and heard someone sing three chords into a bad microphone and felt something we couldn’t explain.
One writer likened it to the scene in Breaking Bad where Walt and Gretchen tally up the elements that make up the human body, account for every part by weight, and still come up fractionally short. Something is left over that the list can’t name. A soul.
Songs often have the same irritating problem. You can account for melody, harmony, rhythm, structure, lyrics, arrangement, production and performance, take the whole thing apart and explain what every piece is doing, and still not fully explain why one song moves you and the next one doesn’t. Maybe that missing thing was never a separate ingredient sitting between the second verse and the bridge. Maybe it only shows up in the relationship between the parts, and between the person singing and the person listening. Either way, technical improvement and emotional effect are not the same measurement. Craft buys you control over the song. It doesn’t buy you anyone caring.
Practice makes you better at whatever you practise
Another writer told me they’d spent three years generating ideas. Hundreds of loops, verses, fragments. Enormously productive, and almost nothing finished. Their own diagnosis was exact: “I basically trained myself to only write parts of songs.”
That’s the catch in the usual advice to just write more. Practice doesn’t make you better at songwriting in general. It makes you better at whatever you’re actually rehearsing. Write a hundred openings and you get good at openings. Start a new song every time the current one gets difficult and you get very good at starting songs and dodging hard decisions. Finishing is its own skill, rewriting is its own skill, cutting a line you love because the song doesn’t need it is its own skill. I’ve written separately about how to practise the part you’re actually trying to improve; the short version is that volume rehearses your habits, so it matters a great deal which habits you’re handing the reps to.
Taste shows up before the skill to match it
There’s an uncomfortable stage in most creative practice where your taste develops faster than your ability. You can hear that something isn’t good and you can’t yet say why, let alone fix it. That gap is painful, and it’s a good sign: your judgement is out ahead of your hands. Then your ability starts to catch up, and the problem changes shape again. Now you can write several versions that are all fine. Which one is better?
This is where the craft stops being about rules and starts being about an ear you trust. Not blind faith in every instinct, because your instincts are stuffed with habits, fears and safe choices. “Trust your gut” is useless on its own. The skill is knowing which parts of your gut have earned the trust, and that comes from taking songs apart, from noticing why a line stopped you, from getting feedback and clocking when three different people all drift off at exactly the same bar, from finishing things and living with the decisions. Taste is often described as what you like. But the more useful version is the ability to notice the difference between two choices and hear what each one does to the song.
The song doesn’t care how hard you worked
This might be the cruellest part. You can spend three weeks on a lyric and make it worse. You can write the best line in the song in thirty seconds while the kettle boils. You can spend hours repairing a verse and then realise the song was better without it.
Effort and quality have an unreliable relationship, and from inside the process they’re easy to confuse. The line you fought hardest for feels more valuable because you remember the fight. The demo you’ve heard three hundred times feels inevitable because you’ve forgotten what it sounded like before you knew every breath.
The song knows none of that. It doesn’t know which line took three weeks and which one fell out by accident, or that the bridge was hard to write. It only knows what’s actually there.
Judgement is partly the trick of hearing the song as though you weren’t the person who made it. You never fully can, but you can get closer: put it away for a week, play it for someone who doesn’t know what you were going for, and ask where their attention wandered, which line they remembered, what they thought it was about. Not whether they liked it.
Liking is almost useless information. You want to know what actually happened when the song met another brain.
The hardest skill is deciding
The longer I write songs, the less I think the job is mainly about inspiration. Inspiration hands you material. Craft hands you options. Judgement is what makes the song. It’s deciding this verse needs four fewer lines even though you love all four. It’s catching that the clever rhyme is pulling attention off the emotion. It’s hearing that the first vocal take has something the technically perfect one lost. It’s realising the chorus isn’t failing because it needs more production, it’s failing because the verse gave it nowhere to go. And sometimes it’s knowing when to stop.
There’s no final level where these decisions turn obvious. If anything, getting better hands you more possibilities to choose between and a sharper ear for everything that might be wrong. That’s why songwriting can get harder the better you get. Not because you’ve lost something, but because you can hear more. The goal was never a point where every decision feels certain. It’s enough craft, attention and self-knowledge to make good decisions while you’re still unsure.
Every rule eventually runs out, and you’re left with the song in front of you, a few things you could do next, and one question worth getting better at answering: what does this song actually need?
If you want another set of ears on that question, that’s worth a conversation.
Not in the way people expect. The obvious, visible problems get easier: a lyric that doesn’t fit the melody, a chorus with no lift. But improving gives you more possible directions for every song and a sharper ear for what might be wrong, so the decisions get finer and harder. You’re not losing skill. You can just hear more.
Yes. You can objectively improve at melody, structure, rhyme, prosody and arrangement and still write a song that leaves people cold. Technical improvement gives you more control over the song. It doesn’t guarantee the song will move anyone. Craft and emotional effect are not the same measurement.
Technique gives you options: more melodies, more structures, more ways to rewrite a verse. Judgement is knowing which of those options the song actually needs. As you improve, the bottleneck stops being what you can do and becomes which thing you should do, and that decision is the harder skill to develop.
Partly by learning to hear it as though you didn’t write it. Effort and quality have an unreliable relationship, so the line you fought hardest for isn’t automatically the best one. Put the song away for a week, play it for someone who doesn’t know your intentions, and ask where their attention wandered and which line they remembered, not whether they liked it.