Well, here's the bad news: There's no set way. Sometimes people start with a basic chord progression, sometimes they start with a bassline, sometimes they think of a melody, and sometimes they start building up a drum loop and work on top of that! It's all about what inspires them at the moment.

Here's the good news: Every single song you've ever loved is an open book waiting for you to read. Referencing tracks isn't just listening to them and dragging them into their D.A.W.s, it's about a whole new type of analytical listening. You're no longer a fan, you're a student.  I've attached a little PDF that has a good list of questions to take with you every time you listen to a track from now on. Start answering the questions as you listen to the song over and over. Do you notice any similarities between the tracks? Are there any trends that seem to emerge? Are certain sounds louder than others? What sounds did they think was the MOST important to hear (the sounds that feel like they're on top of everything else)?

The reason why the top-level producers often hesitate to talk about this kind of stuff is because no one wants to hear "It's a lot of work." But don't worry, because there's plenty of resources online to jump start you.

I'd watch this Q&A session with NGHTMRE and this SeamlessR playlist and see if it helps you any. The MOST important thing to remember is that when you're first starting out it is quantity, not necessarily quality, that you're looking for. The only way to truly know what sounds good is to make a lot of stuff that sounds bad, after all. Every time you force yourself to make a song, you've learned something new so that the next time, that part isn't so hard.

Don't give up!

EDIT: Attaching a picture of the PDF so people can see it without having to download

I found that creating your own song on top of an acapella always helps because you're forced to follow the flow of the vocals. Once you get good at doing that, you can start to make your own arrangements.

I've never thought of doing something like that, great idea!

Well, here's the bad news: There's no set way. Sometimes people start with a basic chord progression, sometimes they start with a bassline, sometimes they think of a melody, and sometimes they start building up a drum loop and work on top of that! It's all about what inspires them at the moment.

Here's the good news: Every single song you've ever loved is an open book waiting for you to read. Referencing tracks isn't just listening to them and dragging them into their D.A.W.s, it's about a whole new type of analytical listening. You're no longer a fan, you're a student.  I've attached a little PDF that has a good list of questions to take with you every time you listen to a track from now on. Start answering the questions as you listen to the song over and over. Do you notice any similarities between the tracks? Are there any trends that seem to emerge? Are certain sounds louder than others? What sounds did they think was the MOST important to hear (the sounds that feel like they're on top of everything else)?

The reason why the top-level producers often hesitate to talk about this kind of stuff is because no one wants to hear "It's a lot of work." But don't worry, because there's plenty of resources online to jump start you.

I'd watch this Q&A session with NGHTMRE and this SeamlessR playlist and see if it helps you any. The MOST important thing to remember is that when you're first starting out it is quantity, not necessarily quality, that you're looking for. The only way to truly know what sounds good is to make a lot of stuff that sounds bad, after all. Every time you force yourself to make a song, you've learned something new so that the next time, that part isn't so hard.

Don't give up!

EDIT: Attaching a picture of the PDF so people can see it without having to download

thanks i wont give up!

This isn't the most fun method, but you will learn a LOT if you do this.

Take a few songs that you really like and write out what is happening.  For example you might write something like "16 bar intro, hats go whole time, kick comes in at bar 9, blah blah blah."  You can do this for the duration of the tune.  Not only do you have a detailed road map, but you can separate it into larger sections like intro, drop, bridge, etc.  Then you can compare these different song forms to find any commonalities!  You'll be coming up with your own song forms in no time

To be honest, in my case anyway I never EVER plan out what comes next. the ideas just flow and the sections for naturally. I usually just begin projects with simple melodies and progressions and the ideas just evolve and evolve and so do the sections. I feel like forcing yourself and planning sections out can be frustrating and stressful. Just sit there and let your song evolve.