Songwriters - what's your standard procedure?

Something to add: I try to cultivate a sense of naturalism. What feels correct, rather than being all whizzy with math and strange bar lengths and stuff. It's all about the groove for me.
 
a bit of an odd one here...

My bassist tells me he has "Synesthesia", and writes most music to whatever numbers he sees in a colour..some shit like that. He also tells me when he sees my name and myself, he sees me in the colour yellow.

But, he does most of the writing for my band, and some of the stuff is amazing. Very talented lad he is.

I also normally think of riffs in my head, then play them when I get a guitar in my hands.
 
Great thread!

Ever since I've been writing on my own I've been having massive trouble writing anything resembling a song. I will usually write a very cool riff or two and leave it there. I don't ever finish shit. And when I have three or four riffs it goes into riff salad and I have no clear verse or chorus.

+1
This EXACT same thing happens to me (I also happen to play BM).

jamming on a cool drum loop

That's a first for me, also. If I get lazy and don't get to write a drum part for the guitar riff I just came up with, the riff will become an 'orphan' from drums for a very long time. One good thing about it though is that if I pick it up later, I'll have a 'fresh' set of ears and it will be a little as if another person would be 'pitching in', making things more interesting generally (I don't have a band I do everything myself). But of course if I write both things at the same time, it will be much more 'tight', albeit not as complex or unexpected.

I used to write a riff at a time and then began to orchestrate it (I write orchestral stuff). That makes matters worse, because not only did I have different riffs w different tempos/keys/drum patterns, but also different sets of instruments. It's all nice and great until the time comes when I have to join at least 2 of these suckers. I'd say to myself something like 'ok now you two riffs will stay together wether you like it or not' and then, if it proves absolutely impossible (after a shitload of time wasted) I end up creating a small bridge of ridiculous complexity to somehow try to 'frankestein' the two parts together. Most times I will end up with a messy indescribable turd, but the thing is, the very few times it actually does work...

The new approach I'm trying since a few months ago is writing only guitars and drums, hand in hand, in a single tempo and a single key, trying to avoid the temptation of dwelling into a single riff as much as possible. But so far that's not cutting it.

In black (orchestral at least) the greatest songs (or at least the ones I like most) are the ones that have so many elements blended together that you can spend literally years listening to them (as a fan) and you will always discover more and more details because of a somewhat 'unified complexity'. The unifying part is what breaks my balls; the 'complexity' part I have it more than narrowed down :lol:

No GP for me, as I have a SERIOUS case of short memory, and so if I don't hit record and play the riff instantly, the riff will be gone FOREVER (man I've lost such GREAT ideas for not recording them at the moment). If I start figuring out the notes with a friggin mouse, the muse will be gone after a bar and a half..

I've learned to run and record ideas or hum them/write them/memorize them etc with the same urge as if I had diarrhea or something
 
and its rarely the case that the GP sounds good and the real song sounds shit, its the other way round 95% of the time, actually in my band we usually say, if the GP already sounds good, then the real song can only sound better

agreed 100%. Sometimes I hear the song in Gp and I´m like "I don´t know dude, it sounds kinda lame" and then we record it or rehearse and it sounds great. Now if in Gp I´m like "dude this riff is so badass" in real life it only sounds better
 
For me it is different each time, but the same few things seem to happen alot

I'll use a riff I've had sitting around for a few years develop it with new sections and find that either the old riff or the new stuff is way better than the other and have to can the song

Or I'll develop an old idea with little problems only to realise the song hasn't taken vocals into consideration and I'm left with an awesome instrumental with a small section of vocals

Or I write an entirely new song which takes vocals into consideration, I find the right key for my voice and record a song that would sound great if I was a good singer.

One thing I'm sure of is it is much easier writing with a computer and being able to critique an idea with all the instruments playing. Sometimes a killer riff doesn't work with a band, and many times I find a lame generic riff on its own drives the song once drums,bass and vocals come along. Just jamming on the guitar for me doesn't give me the entire picture, I'm good at imagining the other elements, but it is so much easier hearing them as you write, being able to tweak a kick pattern and then jam to it..... I suppose you could do this with a band in a rehearsal room, I just never have
 
Everything I write is done alone. I get somnething i like, might be a drum loop, guitar riff, a vocal line with lyrics from a dream (it's happened, and how i fired up cubase and recorded it before forgetting it is a miracle), whatever. Anyway i get something i love and work from there.

Musically I dont have to work so that might not be practical.
 
For me songwriting has to start with a hook, once you have even that ten second riff and a couple of meaningful lines then you can work from that. The initial idea could be a bridge or a chorus or verse, it matters not, usually i won't know what part it is until the song has progressed. I try to modulate into different tonic centres so that the song is changing gear as it progresses, but also repeats itself so as to create familiarity. I try to keep to fairly obvious arrangements more often than not as they usually work best imo. I always try to think what a song sounds to someone else on first spin.

If you come back to a song six or twelve months later there are always improvements that you wouldn't have got at the time, more often than not i'm trimming stuff out of it rather than adding to it. Less is more, most of the classic maiden and priest songs are based around no more than half a dozen riffs.

I think a lot of songwriters here might be getting too hung up on the fine technical details of the recording, that can come later once the song is coming together. If the song idea is good you can chuck a blank tape in a tape recorder and hit record, that's how most classic anthems came together. If the basic song idea is naff no amount of turd polishing or amazing musicianship will make the song work.

Some bands write 'songs' as a finished product, then hand them over to the vocalist to write lyrics and melodies. Personally I think that words and riffs have to evolve together right from the start. Also a good drummer will bring your song to life in a way that would never happen with programmed drums, it is really inspiring how ideas evolve once they get to rehersal stage.
 
I usually work on several pieces at once over a period of time . They pretty much all start with a drum loop and a killer riff that grows into a short sequence of riffs . When ever I get stuck I know there's plenty of other older unfinished stuff that I can be getting on with adding to but generally the strongest tunes keep me hooked . Much of my song structure comes from the actual process of cut and paste editing the rest is done on the fly .
TBH I don't have one system that works all the time but I always have a clipboard of ideas (loose riffs) that I use to import ideas to and from various song projects .
I never seem to have a blank canvas , that kind of situation normally implies that you feel under pressure to produce something that you haven't prepared you self to do so that's why I always record everything even if its not immediately relevant .
Finally after a song is near to being finished I home in on all the parts that seem too long and cut them down sometimes drastically (trust me this works ) it cuts out all the dead wood and breaths life back into the track.