The first point to make is that I'm most definitely not a music theory guru - I'm just a player who believes in the usefulness of the knowledge, and is on the way to learning as much as I can. So take any details with a pinch of salt, and no guarantees.
The first thing that struck me was that without any information about the melody or the rhythm I couldn't really tell much about the song. Hundreds of different songs can (and do) use the same progressions. The difference is in what you do with them, including how your harmonies work. When I played it through it sounded nothing like your version when I eventually heard it.
The second thing was that everything seemed to keep ending on D chords, which is rather unusual (as Gecko pointed out). You said that the song was in G, but what do you understand by that?
When I first looked at theory I thought that it might mean that the melody used just notes from
the GMajor scale, and the chords were those built from that scale.- i.e G Am Bm C D etc. But that's not really so, as a quick look through a few songbooks soon proves. It's a starting point, but it's certainly not a fence.
The true definition of a key is along the lines of
"The quality of a musical passage or composition that causes it to be sensed as gravitating towards a particular note, called the key note or the tonic". (Grove Dictionary of Music). You also see talk of "tonal centres". All it means is that if you use a particular key then the name note (and by extension the main associated chord) exerts a strong influence. You can accept that influence and use it, or you can deliberately resist it to create more tension. You can even change key in the middle of a piece.
So, if that's in GMajor, then it would be trying to pull towards that G, including trying to end on it. A quick flick through one of my songbooks (100 popular songs type of thing) shows page after page of well known hits following that path - ending on the same note as the key signature (i.e. the tonic). It's not in any way compulsory or necessary, but it's VERY common.
Dozens of songs in G in the book, ending on a G, but also one or two songs that appeared to be 'in G' - ( i.e. the staff showed only one sharp, F# ) - actually ended on an E however. Why E? Well, one possible reason is every Major key has what's called a "relative minor" - which is a minor key that uses
exactly the same notes in its scale. Count along any Major scale and it's the 6th note. So for C it's relative minor is Am, and for G it's Em. The same chords can be built from the scale but the difference is in how they perfom when you use them around a different 'centre'. The I, IV, V progression will therefore also use different chords (from the same overall set) when it's in GMajor than it would in Em, which obviously sounds totally different. So a song that looks like it's 'in G' but that ends on an E might well actually be in Eminor. The way to tell is presumably to look and see how it's structured (and of course what it actually sounds like!
)
Just to further confuse things, plenty of chords show up in several keys. For example, there's only one note difference between the key of G (which has an F#) and the key of D (which has an F# plus a C#). So some of the commonly used basic chords used in G and D will be the same. Depending on the direction you take the music in, just looking at chords can be quite misleading, especially as there's nothing wrong with using chords that don't belong to the set of 'usual suspects' or doing things like using a minor where you might 'expect' a Major - or even both together in the same song. What theory does is give you some great hints on how to handle all that to achieve the effects you want.
But can't you get there by just trying lots and lots of options, until you find one you like, and building up a bank of 'ear theory' instead of 'book theory'????
Well..... I certainly hope so... because I do both... I just think that theory saves me time, and also opens up possibilities that I probably would never have stumbled on. Works for me.
What I am starting to get out of it (at pretty primitive level still) includes suggestions about likely chord substitutions to take the song away from the usual paths, and information about building interesting harmonies under the melody (or the other way around). I tend to experiment first - 'by ear' if you like, and then when I get tired of working through dozens of possible chords I check out the 'theory' - which is basically (as VHS said) just a collection of advice about what many thousands of other previous musicians have found to work. It seems especially useful for those times when I've tried something fairly randomly and it's
nearly there but still clunks. Changing a note or two in a chord, can make all the difference. Theory can help identify which ones to change, what to use instead, or simply what to add.
)
