Hey guys, I'm wondering how you guys write songs. From the first time you pick up the guitar to when you record the song and every step in between. Thanks!
Aren't you the guy who spammed the Pit with his horrible FB page ad?

On a different note, I'll answer your post...
I write most stuff beforehand. Sometimes, I may have a riff I use as inspiration, but by the time it's on paper...it's no longer the original riff. (It usually becomes some mutated horrible thing with an odd sense of beauty.) Then, I refine what I've written. I swear I use more pages of sheet music than anyone I know. It's quite fun!
Thank you for your help! And I wasn't spamming. Its actually a page me and a friend made to help out local acts. Sometimes kindness can be misconstrued as being a form of unwanted internet ad. Networking with other musicians if you will.
Quote by JeffreyBDePuy
Thank you for your help! And I wasn't spamming. Its actually a page me and a friend made to help out local acts. Sometimes kindness can be misconstrued as being a form of unwanted internet ad. Networking with other musicians if you will.

Ok. But what you did was against the rules, technically.

One other thing that helps me is to have a progression in mind, before I start writing. The lead guitar (or other "lead" instruments) may end up doing something really interesting, but the base progression may be something simple like I-IV-V. Not always, but it works sometimes.
yeah, I didn't know about the rule. But now I do so I can avoid it lol live n learn. Now my question other question is, How do you connect everything?
Quote by JeffreyBDePuy
Hey guys, I'm wondering how you guys write songs. From the first time you pick up the guitar to when you record the song and every step in between. Thanks!

It's often a long process for me. Sometimes I just jam out but if I'm writing a song there is a lot of trial and error and messing around that goes into it.

It is a lot to explain and I'm going to miss out a lot of my process as it can vary a great deal. But I'll explain some of the approaches and processes that I have gone through recently in writing some songs.

Sometimes I will learn about or read something that inspires me to try a specific idea.

Example - I was reading something about harmonizing melody and chord substitution
Introduction to the subtle secrets of re-harmonisation
It should be obvious by now that even when a song appears to be defined by a simple melody, the underlying harmony should never be taken for granted. The chords of a song are not necessarily 'implied' by a tune, as novice musicians might assume. The same melody can be 'dressed-up' and take on a whole new identity purely by harmonising it in a different way.

The Beatles confirm that they understand this implicitly in those songs where they not only harmonise a melody - but re-harmonise it. This acknowledges that there is not necessarily one 'best' way to harmonise a melody line, and for any particular
pitch in a given melody a songwriter can change the supporting chord.

Take a look at 'Yes It Is', a song perhaps most famous for its painstakingly constructed three-part harmony. However, here we have just extracted what is the arguably the strongest individual melody line:
"Yes It Is" - The Beatles

The extract opens with a two-bar phrase whose melodic pitches are repeated identically to form the first 4-bar section. Notice that the sound of the crucial second syllable of 'to-night (which we hear at 0.15) is subtly different from the one we have already heard
(first at 0.08). The Beatles have switched the chord under that B note.

This B pitch is the familiar 5th of the parent scale of E major and, when sung over the F#m7, it appears as a delicately haunting 4th above the minor chord. However, second time around, the same B is re-harmonised by a steelier D major, over which it is now heard as a more luscious 6th.
...excerpt from The Songwriting Secrets of the Beatles by Dominic Pedler - Appendix 3 Harmonization pp 745-746

I've read this book years ago and loved it. I know all this stuff already but I often go back to different books or ideas to find a little idea to kick things off. So I read this appendix and thought -hmmm I want to try that.

So I took a few words ("It's only love") from another song and taking them out of context I thought about how different inflections of those same words can have entirely different meanings.

I settled on the message I wanted to portray and the kind of inflection the words would be spoken to convey that meaning. I spoke them and then sang them keeping the same inflection and tone to retain the intended meaning. From that I derived a simple melodic line.

I then harmonized the melodic line by matching the stronger melody notes to chords that contain those tones. The melody ended on G and I used Em (note G is the third of Em).

Then I repeated the melody for the second line and this time around I changed the last chord to a C major (G is the fifth in C Major).

I liked the effect this had and the next line (melody and harmony) just came without any real thought or effort. But I wanted to play with the idea some more so I repeated the new melodic line and reharmonized the last note. -This was a freakin nightmare. Finding just the right chord took me forever and then it lead me to altering the second guess other chords in the same line. But whatever. I eventually just settled for a chord progression that I still felt wasn't quite right but considered it a work in progress.

I played it over and over though trying out different ideas in that second section and even though I never got that section quite right the playing over and over led to development of new ideas and other parts to fill out the song a bit more and different lyric variations came to me as I was doing it - sometimes by accident sometimes on purpose. Eventually I looked at it and thought that structurally it was enough to be a complete song. I thought "That'll do for now, it's pretty much done but needs some tinkering with fresh ears."

I liked it when I wrote it but when I listened back the next day - it hit the scrap heap. It sounded pretty awful. Despite that it was not a total loss as there were certain ideas that worked well and that I can use again in a new context.

For another song (and a lot of songs) I would just pick up the guitar and start messing around with a particular chord or set of chords and through adding and removing notes I would build melodic movement within or over the chord. -Most recently this involved starting with Am and using direct chord substitutions like Asus4 Asus2 Am7 and Am(b6)I would create little melodic ideas over the chord. I would then go into an Em chord and continue the development of the melodic idea.

After vamping on these two chords I got a sound I liked. I toyed with a adding a C major chord in there, and played around with a moving bass, but abandoned these ideas and settled on just the Am and Em with minor variations on the melody. But playing it too much became too static and it needed some dynamicism.

I don't really plan out my meter or rhythm at all. I don't count the number of bars or anything like that, it's all done on feel. So after "enough" of the Am Em repetition I went to a G D7 C B7 Em It's not diatonic but I started with a cycle of fifths D G C B7 Em but just played around with it a bit and settled on the former.

The song is far from complete at this stage but I really liked what I had done so far. I took what I had and roughly recorded a it through a couple times as a scratch track to work on other parts expecting that they would then influence other parts as well as alterations to the original part itself.

Then I looped it and started trying to find some melodic ideas by just jamming along over the top with complete freedom and lack of judgment - kind of like a brainstorming session.

Then I would put on my critical ears and listen back. I would hear a few ideas and think hmm that sounds okay, that's alright, that's absolute dog balls, that's awful, that's alright...etc.

I'd repeat this process a few times. Until I hear it. I don't always here it and sometimes settle for less but occasionally I'll hear it and you know it when you do. Then you have your idea for that part and can polish that idea and grow out from it.

Other times I'll just jam on some chords with some mates and we'll sing corny lines and tell stories. This sometimes leads to song ideas. Sometimes I jam out with myself. I record a track then another then another layering track upon track upon track adding something new each time. It can be a lot of fun and generate some interesting ideas.

Sometimes I just write lyrics - on my computer or in a note book or wherever. Or I'll be driving and I'll get an idea so I pull over turn on the voice recorder on my phone then hit the road again and just work through ideas verbally or by singing them.

Other times I work without any lyrical ideas or chords and just build melodies. I tend to keep it simple. Slow notes that would work as a vocal melody. I'll start with a diatonic or pentatonic scale and just work out something that sounds good. Then I'll start building the rest of the song around the melody.

I guess the common thread with all the different ways that it's always on my instrument and there is always a period of just trying out a shitload of variations on an idea over and over and over.

There are things I don't do - I never start with a set structure or form in mind.

I never start with sheet music or writing the music like crazysam described. -I have to have an instrument in my hand and physically play or sing the ideas as they come to me. I usually record rather than notate my ideas, either on my phone or on the computer.

I never decide on a specific time signature or meter in advance. I do sometimes deliberately change the time signature when I feel I need change and I'll try it out. But I never set out to write a riff in 7/4 time or whatever - if the riff is in that time I usually don't think about it until I have recorded it and realize the measure bars in 4/4 don't line up.
Quote by JeffreyBDePuy
yeah, I didn't know about the rule. But now I do so I can avoid it lol live n learn. Now my question other question is, How do you connect everything?

This can, depending, be an easy thing or a hard thing.

Sometimes, I write really complex prog songs. In that case, I usually have to write a few "connecting" riffs. Because, otherwise, it would sound too disjointed.

Other times, I just write simple rock songs. Those are easy to connect, because the riffs are usually in the same key. Or in keys that naturally lead to each other. So, it just works out that it all connects.
I've written songs in just about every way I could imagine. Some real quick, and some over years of refinement. But the general rule for me, is hook, verse, bridge. Generally chords first, and then melody, and then lyrics.

Different styles are different though. Some, I prefer to start with a beat.

Sometimes I write straight in the DAW. Otherwise, I will record hook instrument, get bpm from that. Make beat for that, use beat as metronome, record full instrument I wrote with, guitar or piano, whatever it was, then a draft copy of vocals, and then usually bass, or whatever i feel like from there, Sometimes working section to section, other times one instrument the whole way through.

Like I said, I've written them so many different ways. The one constant though, is to sit down and do it. I might sit down and not feel this or that right away. But I will eventually find something I get into, and then the journey unfolds however it unfolds, depending on how each step inspires me to take the next one.
Last edited by fingrpikingood at Sep 17, 2014,
It seems to happen differently every time for me (the order of events, that is).

Usually it sparks out of a chord progression. I like my 'fancy' chords and Hendrix-style approaches to chord progressions so I usually have something like that going on. I like to imagine a chord progression and whatever colored chords I used as almost a back drop to the song. It's already given the song a mood/feel/vibe/landscape, whatever you wanna call it.

Then I sort of jam on the song for awhile and pick apart what's going on and where it seems to be wanting to go and write it down. I like to use big poster boards and lay one or two out on the floor and just write shit down with a permanent marker as I go. I'll write down the progression, figure out what kind of tempo would be good (moderate, fast, slow, etc), write out what's going on pulse/meter-wise, and jot down some basic notes on structure.

If I feel like It's going to take a 'Verse/Chorus/Bridge' sort of thing and I only have a 'Verse' progression, I'll start figuring out a few ways to go about putting together something for other parts and make decisions like whether to modulate to a different key, use the relative minor, use the whole 'soft/loud' idea, use a progression starting on a non-tonic chord, etc. at that point too. The idea is to get the shell of the song working.

Then I'll usually decide on a good tempo number and start throwing vocal ideas out there. I use a tape recorder, hold the metronome that's on my phone to the tape recorder, count off, and then sing a melodic idea. Then I play it back while playing my electric guitar unplugged to the click so I can hear how the melody sits on top of my progression. Then I'll try different things for hours until I'm completely happy with the melody.

I try to keep the melody within the 3-9 syllable range per phrase and usually only work on 8-10 bars at a time (so like a half-verse. It depends on the progression. I might work in 4s.). That makes it easier to throw lyrics on it later and to make changes if I have to. I'll also put together more ideas than I need with variations on how long the phrases are, where they start/stop, etc.

Also, I think John Lennon said that he takes other people's melodies and '****s them up'. I do the same thing. If I'm totally stumped on where to go, I'll try to find some songs with a similar progression or part of a progression and listen to how they approached it and in all likelihood 'borrow' notes.

Most people say to start with the melody, but I like to adjust along the way and I have a much easier time forming a rhythm/pulse to a tune by just playing with a chord progression and tinkering with it if need be.

Anyway, after that, I'll try to get some 'bum' lyrics going and get something coherent together in Ableton. Then it sits until I'm ready to put final lyrics to it. I'm no poet and it's something I'm still working at, so usually these guys sit and bake for a few weeks/months until I have 'something to say' or whatever the kids are saying these days

I would recommend reading some books on the subject. The one mentioned is awesome (just got done with that one a few months back). The one I started with was 'How Music Really Works'. It's not the most 'correct' book in the sense that some of the info isn't technically correct, but it gave me a foundation in looking and breaking down songs.

As much as there's a creative side to songwriting, there's also a technical side. Your process is your own, but understanding how to do things musically and understand what's going on and what your options are is vital to writing at a consistent quality level.
Last edited by mjones1992 at Sep 17, 2014,
Quote by mjones1992
As much as there's a creative side to songwriting, there's also a technical side. Your process is your own, but understanding how to do things musically and understand what's going on and what your options are is vital to writing at a consistent quality level.

Yes, this is where theory can help: it gives you options.
I write when I hear "it" in my head.

Sometimes it's just the melody. I record whatever it is I "have" and archive it. Sometimes for years, sometimes it happens right away. I get the basic melody and I "know" what chords go with it. Many times I have the melody and chords at the same moment, and from there I make it as raw as possible. Again, I record everything, and try and make sense of it later. I'll use words I think of that moment as "lyrics" and see where that stream of consciousness takes me. The only time I use theory is when I have exhausted all these ideas, and I feel like it needs something extra. Then I might start thinking more of trying an idea, and then going with it wherever it takes me, rather than me dictating what "it" is.

I refuse to write any other way. For me any other way feels unnatural.


Typically in three forms:

1. Stream of consciousness
Usually starts with a single chord that is expanded upon melodically, harmonically and instrumentally as soon as I think of what's coming next. These are usually written extremely fast and can range anywhere from 16bars to nearly a full song. Once it's finished I fix stuff that doesn't work.

2. Start with a specific context that I want to achieve.
Does what it says on the tin. I use what I know to achieve a specific musical goal. Could be to write in a specific form, or to indulge some chord progression that I like the sound of, or to implement some concept that I read about.

3. Copying others.
Basically just try to write a song in the style of someone else.

All the can be done either with or without an instrument in hand. Depends if I want to limit myself to my own skill level (or use instrument-specific technique) or not.
I just go with the flow. When I write I can't force something good to happen. It can start out with just a single riff, basic drum beat, or bass line. I have a few songs that were accidentally written meaning I was fooling around not expecting to come up with anything but it happened anyways and I saved it to my riff library. You might drive yourself crazy trying to sound like somebody else or over think the "technical aspect" but to each their own. I don't think there is a wrong way to write a song as long as you are being creative and expressing yourself.
Oh another thing I do - in regards to vocal melody and lyrics.

I sometimes have a basis for the rest of the track (drums, guitars, bass, synths or whatever) but no vocal or melody line and no lyrics.

If that happens I will often hit record and vocalize/sing a melody while listening to the track. I sometimes sing words that make no sense and sometimes it's just vocalizing (no words or syllables, just humming/grunting/singing something that works).

That allows me to figure out basically how the vocal melody will go and I then repeat and the vocalizing seems to often just form into words.
Quote by fingrpikingood
Generally chords first, and then melody, and then lyrics.

Melody first is a great approach sometimes. Often when I go harmony first, my melodies sound a bit half-assed.
Quote by Elintasokas
Melody first is a great approach sometimes. Often when I go harmony first, my melodies sound a bit half-assed.

Ya, the melody is the main important part. If you chord first, then it is easy to have very complex chords, which limit your options for melody. A lot of great songs are very simple in chord progression, very basic. If you write just chords first, then you will likely not be satisfied with something simple, so you will make something more complicated, which can affect things like I mentioned. Writing melody and chords at the same time solves that as well.

I'm more of an instrumentalist/improviser that also writes songs, than a songwriter though.

My melodies may not be the strongest in the universe, they may sound half assed to some people, but I never create them half assed for sure. Even if I come up with the progression first, the melody is always the priority to me. Music to me, is always, in a state of construction. I will change anything to allow for a stronger melody. Once I find a melody I like, I will sort of stick to that, unless I find something I prefer, and everything else must adapt to that.

That's why I come up with the melody before the lyrics often, or at the same time maybe. And that's why when I arrange it, the first thing I record is the tempo, then the main harmony, and then a draft of lyrics. That way, I am certain I won't write a bass line, or whatever that gets in the way of my melody.

Melody first is definitely not a bad approach at all though. Maybe I'll write my next guitar song that way. Haven't written one like that in a while.
Whatever works for you is the right way, but it never hurts to try different approaches.

That original of yours, "Miles in your Shoes" is a pretty nice track. The guitar sounds great. The only problem with it is that the song gets a little repetitive by the end.