There is always room for more songs in the world, especially great ones! Have you written one? Or perhaps you've started one that needs help, or you've written hundreds and... well if you've written hundreds of songs, you probably don't need to keep reading. But for all other interested parties, here we go!

Step 1: Disclaimer; Disclosure

Step 1. Disclaimer; Disclosure
Disclaimer
First of all, keep in mind that nothing I say is etched in stone, or if it is, it shouldn't be. These are just some of the tricks I use, and everyone has their own methods. If anything I say doesn't ring true, then by all means, ignore it! There are stellar examples of quality songwriting which are exceptions to everything written here. For instance, the order of the steps is almost completely arbitrary; I've certainly written songs that didn't follow them. But I do suggest that you look at each step at some point in the process; I usually tend to notice things that could be improved in my songs when I look back
(see step #5).
Also, google "songwriting tips" and you'll have tons more to read, and probably of much better quality.

Disclosure
Here is where I'm "coming from" when it comes to songwriting:
I play guitar and write and sing. I believe that quality in songwriting lies in achieving a balance between (forgive the cliches) the heart and the head. Your inspiration comes from your guts; from emotion, and you render that raw emotion "deliverable" by filtering it through your brain. If the balance is off, the song is less than it could be. Pure emotion doesn't translate to the audience unless the performer (or songwriter), intentionally or unconsciously, makes that emotion accessible.

Hendrix's axe was a part of him, his emotion translates ... but what if he just hadn't practiced enough? Or he hadn't invested lots of time playing with new hardware, new sounds? Or if he forgot all of the words (Yeah yeah, I know... he DID forget the words a lot)? My point is, he was able to convey the emotion (guts) of his music, because he found a way to make it work on stage (brain). Jimi (at his best) found a balance, and for him it happened to be a bit farther toward the guts than the head.

Sting has done some truly great songwriting, but he constantly writes songs with too much brain and not enough guts. Too much meditation? Not enough? You be the judge.

I do believe that there is great value in a song that a huge number of people like. "Popularity" is either a sign of a truly bad song, or a truly great one: There will always be an audience for cookie-cutter pop drivel, but there will also always be an audience for a great song. Anybody who has been at a performance where the whole audience knows the words and is singing along, has felt the value of a "popular" song, and it's elitist to suggest that the moment a song becomes popular, its quality decreases. So I think that the best songs have something about them that "lets everybody in," or at least, lots of folks, rather than focusing on an extremely narrow audience. Everybody appreciates a great song, regardless of the genre, and I think music is about communication. So I think a song that is well-written, is one that more people "grok."

As a songwriter, I try to write songs that "don't need me." By that I mean, I want them to be able to stand on their own, to be of sufficient quality that any decent singer could sing them. Few of the truly timeless, transformative classic songs have "weak spots" that a singer must "sing around." In my opinion, it is the songwriter's job to take their inspiration and wrap it in a package that magnifies; that enhances. A "poorly wrapped" song conveys little or none of its original magic, no matter how genuine the inspiration. And a song with no "guts" is just a wrapper.

The craft of songwriting is like cutting a gem.

Bad songwriting hides or distracts from the inspiration, good songwriting amplifies it without overpowering it.
That being said, "Louie Louie" it a great song... so again, take what I say with a grain of salt.

Step 2: Inspiration

This is often the hardest part: what to write a song about? I think you're best off finding subjects that you truly know about, and care about; something you're feeling in your gut. Otherwise, your song comes straight out of your head, and some of the worst songs in history were flawlessly executed cerebral exercises; gutless wonders.
So, the "big breakup" is a no-brainer; you're emotionally crushed and have a deep well of torment to dip your songwriting bucket into. If, however, you haven't just been dumped or you aren't unrequited or generally grouchy about anything, you have to look elsewhere. I've gone WAY back to my high school days, or used a friend's story, or used books or movies that impressed me. Watch out for "protest songs" unless you are truly angry or moved and have something new to say; otherwise they end up sounding whiny.
(Forgive me as I inject my opinion about whiny protest songs: "Excuse Me Mister" by Ben Harper: GREAT groove, great vibe, great attitude, but no beef: it's blaming the "other guy" for everything, and it comes off as whiny to me. In contrast: "Play the Greed" by Dar Williams: practical advice, saying something new, taking responsibility, no whining. Definitely a "cerebral" song, but she believes it enough to lend it guts!)

Don't write if you're not thinking about anything. Or at least, don't get attached to what you're coming up with. Wait for the inspiration, or go get it somewhere; meet people, watch an old movie, eat Doritos. Then write.

Step 3: Babble

Now that you have a vague subject, it's time to holler gibberish. I play guitar, so generally I'll start playing while thinking about "the subject," and sing random verbiage. Eventually, an relevant word or phrase will fall together with some musical element that's splattering around; a noise the guitar is making, a piece of the random melody, etc. Then you can build on that. If you don't play an instrument, you may just want to write out some flow-of-consciousness text until something on the page "resonates."
I have countless pages of babble. Keep these forever! They may contain seeds of great songs that you didn't notice during the initial babble-session, that will sprout years from now.
Some folks might want to record this stage... you're welcome to give it a try if you don't mind the notion that your babbling has been recorded. I'd rather scribble.

Step 4: Build...

Now that you've found some "seeds" of a song, start putting words together. Look for rhymes. You're just fishing here, looking for direction. At some point, a few lines will attach themselves to each other, and your song will start telling you what its "rules" will be.
"But," you may exclaim," there are no rules to art, are there?"
"Sure," I reply, "but quit yelling, will you? I'm trying to type!"

There are no general rules, really: there are always hundreds of examples of great songs with any given rule being broken. But I submit that each song should have its own rules. You can make them up, of course, and you can break them. I just suggest that you do it for a reason. Do it because it helps the song, makes it better.
The "rules" can be structural: rhyme scheme, verse-chorus-verse sequence, iambic pentameter, never rhyming, always rhyming, no half-rhymes, only half-rhymes... or the rules can be more about content: no quotes from former presidents, only quotes from former presidents...
A song is stronger when it has rules. Because structure on some level is crucial to communicate the magic part.
I use high-speed, barely legible penmanship when I'm writing songs, because I want to get it on paper before it slides out of my head. I scribble out chords, or "hints" that will help me remember the music I'm fiddling with. I cross out verses that I don't like, put stars next to the ones I do. When I know I need more words in a verse but haven't thought of them yet, I let squiggly lines work as placeholders. This process will of course be different for everyone, but the point is, find a "building" process that works for you.

Step 5: Craft!

Lots of folks pride themselves on their songwriting speed, but they tend to be the same folks that have songs with parts that are boogered-up.

I agree that the stuff that "flows" out in half an hour is often some of the best stuff. But I'm also constantly hearing unfinished work. Here's why: most of the time, the inspiration happens and you end up with two great lines that rhyme, and crap in between that is only there to get from Fantastic Line #1 to Fantastic Line #2. Your job as a songwriter is to fix the crap without overshadowing or otherwise screwing up the inspiration, or even better, to magnify the inspiration. This is more craft than art, but it's crucial and it's the difference between timeless and transformative songs and just good songs.

I read somewhere an interview with Glen Frey, where he talks about how he does this "2 line at a time" thing, with those "placeholder" words to glue the good stuff together, but that Don Henley could always come in and put meaning into those in-between words, and take the song to new heights. Of course of the two, Henley has come out with more stuff that is annoyingly over thought, too cerebral. And we've all heard songs like that; too cerebral, no inspiration. Music written with all head and no heart, all thinking and no feeling.

Again, It's all about balance, just like anything else. Heart with head: you get the inspiration in your guts, but if you just leave it alone, then you're the only one who's going to really get it. If your performing (or recording) for others, you owe it to them to give them a way in, and that's done with your head. Sometimes you luck out, and the inspiration already translates to the listener, but you have to figure out when it does and when it doesn't, and you owe it to the song to make it as great as possible, right?

After you write a song and it's at its first "plateau" of completion, give it some time, then try to listen objectively, and see if you still like the song the way it is. Most of the time, the lame bits show themselves, and are easily fixed in a way that helps the song; a small word here and there, minor tweaks...

Another item to consider: balancing the general with the specific: For example, you're a guy and you wrote a song about a girl that broke your heart. Does the song suffer if you play the "pronoun game," making the song gender-neutral, so that a girl who just broke up would also relate? Or a guy who broke up with a guy? A girl with a girl? A guy with a beloved chicken? If the answer is "yes, the song would be worse off," then by all means, don't change that element. But if the song's power is left intact (or increased by singing straight to the person: saying "you broke my heart" instead of "she broke my heart"), then what's the harm in making the switch? You've widened the possible impact of the story, without messing with the emotional center of it.

Play the profanity card carefully! James Blunt really should not have done it in his ""You're Beautiful". He used the f-word when he could have just kept telling the story, and it detracts and distracts from the song. In contrast, Ani DiFranco's "Untouchable Face" uses the f-word like a million times, and it works. Know that "that word" you've included in your song will receive extra attention. Does that throw off your song, or make it better?

Let songs simmer for long periods of time before you stamp them "finished," no matter how quickly they were written in the first place. Get over the "I wrote it in 15 minutes" ego thing. If you wrote it in 15 minutes and it's good, imagine how great it would be if you went back and fixed the tiny things that detract from the power? If you kill the inspiration in that process, then you've blown it...

...But I think you've screwed up just as bad if you leave dumb stuff in the song. Tragic, isn't it: a good song with the potential to be great, that makes it into the world complete with a bunch of little obstacles to its greatness. It will fizzle, when it could have been one of "those songs."

Step 6: Now What?

Good question. And a bit beyond the scope of this instructable.
For me, I perform my songs live, record them, make CDs, give them away or sell them at my shows or in local stores. And no, I haven't quit my day job.

(My music pages are here; and here, videos are here, free mp3 downloads of everything I've ever recorded is here)

Another thing to do is try to market your songs. Good luck with that; it's a lot of work... but you'll get a paycheck now and then if somebody records it. If you don't play, look around for players willing to give your song a try. Hit your local open mic. Join a songwriting contest. Sing around the campfire.

But regardless of what you do with your song, work on it a little. Do a great job on it. Think about what music has meant to you; that one song that really hit you hard when you heard it on the radio or that movie soundtrack... why not shoot for that? Even if the world doesn't get to hear it, you'll know that you did good work, and that's fulfilling enough to be well worth the attempt. Right?