How Long Does It Take to Write a Song?

Writing a song takes about 8 to 30 hours of real working time, spread across 1 to 6 weeks. The sketch, meaning the chords and the shape of a melody, often shows up in 20 minutes. Getting from that sketch to something worth paying a studio to record is the part nobody puts a number on, and it is mostly lyrics.

I got asked this on a call last month by an artist who had been on the same song since February. They wanted to know if that was normal, and underneath the question was the real one, which was whether taking that long meant they were bad at this. It does not. But the way they were spending the months was costing them something, and that is worth talking about honestly.

After 20 years of making records (19 Billboard Top 20 albums, a catalog past 3.3B+ streams), here is where the hours actually go.

The Real Breakdown

Stage	Typical time	What is happening
Idea to workable sketch	20 min – 2 hrs	Chords, tempo, a melodic contour, a title or a hook phrase
First full draft	2 – 6 hrs	All sections exist. Verse, chorus, bridge, an arrangement in rough
Lyrics to finished	3 – 20 hrs	Almost always the long pole, spread across days
Rewriting and cutting	2 – 10 hrs	Second verse, the bridge nobody likes, trimming
Total to a song worth recording	8 – 30 hrs over 1 – 6 weeks

The spread in that bottom row is enormous and it is honest. A song that arrives whole, and they do arrive whole, can be finished in an afternoon. Most of mine do, and I will get to that. The five-week second verse is real too. I see it across the desk far more than I see it in my own writing, and the next two sections are about why.

How It Goes for Me

How long is a piece of string? The Beatles probably wrote "Help!" in forty seconds. I am not claiming that, but my own songs sit much closer to that end of the table than to the six-month end.

It usually starts with singing a melody. Then I will fit some words into it, then I will strum the guitar around that. Sometimes the words and the tune show up at the same time. Either way the song comes together within half an hour to four hours of writing. Not the record. The song. Everything after that is arranging and producing, which is a different clock and a different bill, and I get to it below.

I am giving you that as a second data point next to the table, not as a standard. The table is what I see across the artists who come through my room. The four hours is what happens in mine, and I think the melody-first order is most of the reason. The words have a shape to fit into before I write them, so they never get the chance to eat the calendar.

Why Lyrics Eat the Calendar

Music can be judged in the moment. You play the chord change, and you know. Your body knows before your brain does. There is no waiting period.

Lyrics do not work like that. A line written at 1am feels like the truest thing you have ever said, and then you read it on Thursday morning and it is a greeting card. That gap only shows up with distance, which means lyrics need calendar time rather than more hours in one sitting. This is the single most common thing I see artists get wrong about their own process: they try to solve a lyric by staying up later, when the actual tool is putting it down and coming back on Tuesday.

So when I say 3 to 20 hours for lyrics, those hours are deliberately spread thin. Four sessions of ninety minutes across two weeks will beat one eight-hour marathon almost every time, and it is not close.

The Uncomfortable Thing About the Song You Are On Right Now

Here is something I can bring to this question that a songwriting blog cannot, because we keep a database of streaming snapshots built for our Sonic Analyzer.

Across 118,291 artists where we can see both the artist's most-streamed track and their most recent release, only 1.3% have their biggest song be their newest one. For the other 98.7%, the track carrying their catalog is behind them by a median of 5.3 years. Restrict it to artists who actually released something in the last twelve months, which is 57,084 of them, and the number gets starker rather than softer: 0.5%, with the biggest song a median of 5.4 years back.

One honest caveat, because I would rather you trust the next number I give you. Some of that gap is just accumulation. A song released in March has had less time to gather plays than one released in 2019, and that alone would tilt the result. So do not read this as "new songs fail."

Read it as this instead: you cannot tell which one it is while you are writing it. Not from inside the room. The song that turns out to matter is chosen years later by people you have not met, and it is very rarely the one you were most sure about. I have had records I fought for that went nowhere and throwaways that paid rent for a decade.

Which is the actual argument against spending seven months on one song. Not that slow is bad. That betting seven months of your writing life on your own in-the-moment judgment of which song is the good one is a bet the data says you lose.

Where Writing Ends and the Bill Starts

Writing is free. Everything after it is not, and artists routinely blur the two into one timeline and then get surprised.

Once the song is written, producing it takes roughly 40 studio hours over 2 to 4 weeks, which is arranging, tracking, editing, mixing and mastering. Stack that on the writing and a single realistically runs 6 to 10 weeks from blank page to distributed master.

The reason this matters financially is that unfinished writing is the most expensive thing you can bring into a studio. If you arrive with a second verse you are still unsure about, you will solve it on the clock, at producer rates, in a room where the meter is running and your judgment is worse because you are being watched. Finish the writing on your couch where it costs nothing. That one habit has saved my clients more money than any gear recommendation I have ever given.

How to Write Faster Without Writing Worse

• Set a deadline that is not about quality. "Draft done Friday" works. "Done when it is great" is not a deadline, it is a mood.
• Separate writing from editing, physically. Different session, ideally different day. Doing both at once is why the second verse takes five weeks.
• Finish bad songs. The skill you are building is finishing, and you cannot practice it on the song you have decided is precious.
• Write more of them. Volume is the only lever that reliably works, precisely because you cannot pick the winner in advance.
• Know when it is done. If you are moving words sideways rather than forward, it is done. I wrote about that judgment call in is my song good enough to release.

And if the songs are getting finished but nothing is sticking after release, the problem has usually moved downstream of the writing. That is a different diagnosis, and I covered it in why your song gets streams but no fans.

The Bottom Line

Budget 8 to 30 hours over 1 to 6 weeks for a song worth recording, expect the lyrics to take two thirds of it, and spread those hours across days rather than stacking them. A song written in one afternoon is not cheating. A song that took six months is not virtuous. The only thing that separates writers who eventually land one is how many they finish, because at roughly $3,000 to $4,000 per million streams, the math never rewards the perfect single song. It rewards the catalog that had enough entries for one of them to catch.

Once the writing is finished and you want to know what recording it properly actually costs, the rate calculator gives you a number calibrated to where you actually are as an artist, not a range built for somebody else's career. Run it here, it takes about a minute.