Use your inbuilt and forgotten ability to explain your topic
Most of us find it very easy to tell our colleagues about our work, if it is in an informal short-notice environment. Bump into someone getting coffee, and if they say,
"What are you up to?", and you will likely instantly respond with, "Oh, you know that X happens? Well, we've found out why. It's because Y!".
If they enquire, "What makes you so sure?", you will say, "We got some Z, and did A, B and C, and found D. This means E. So then we did ... etc."
Typically they will say, "OK I suppose I believe it, but couldn't it still be P or Q?" And you would reply, "Yes, but P isn't likely because ..., and Q probably contributes only in a small way because of ...".
I am a Professor of Cardiology, so this is how it goes in Medicine, but it is likely to be similar in all academic writing.
Somehow, though, when we come to writing it down we are paralysed
Beat this paralysis by actually doing the interaction described above. Or even just imagining it. That's the main thrust of the paper. Once you have that in your head, set about the paper like this:
1. Write the Abstract first
In fact write the last line of the Abstract's Conclusion first
In situation W, X is largely caused by Y. Then precede this with whatever else your paper concludes.
Then write the Abstract's Intro
X, commonly found in W, is important because .... Its cause has been unclear. In this study we tested whether X is caused by Y, Y2, Y3 ..., using ....methods.
Then write the rest of the Abstract
All the key methods and results. Within the word limit (e.g. 250 words total for Abstract). You will come up with lots of stuff that doesn't fit. Don't worry: push them down to the end of the document because they will go into the main Results section.
2. Write the Results, and any associated Figures/Tables, next
Go through the Abstract's Results, and each statement you wrote there, write in fuller detail here, bringing in all the things that you didn't have space for.
3. Now write the Intro/Background of the paper
Only start writing it now, not before, otherwise you will fill it with general stuff about your topic. If you only write it now, you will focus it much better on "Why we are bothering to do this research." You don't have to review the whole topic. Only talk about the problems that you will solve in this paper.
4. Now write the Discussion of the paper
Keep it short. It won't end up short, but if you always try to keep it short, you will help prevent bloat.
In this study we found 1, 2, 3.
Then a section on finding 1. If you differ from other workers in your results, speculate on why. Ditto for 2 and 3.
Then discuss any implications for practical use.
Next, the Study Limitations section of the Discussion
What would people (puny ignorant fools!) immediately think was wrong with your reasoning? Put those ideas here, and then say why those doubts are unfounded. Then include any genuine weaknesses you can think of.
Finally, the Discussion's Conclusion
Reiterate the Abstract's conclusion, perhaps allowing yourself more words. In the Abstract, your conclusion had to seem like a valid conclusion from the information you presented in the Abstract itself. In the Discussion Conclusion, you can now draw upon all the results and all the reasoning in the Discussion, so you might be able to be more specific.
Congratulations!
You now have your first draft.
Refine
Find someone who has 10 minutes to spare. I suggest reading out the paper to them, watching their expression on their face. When they look puzzled, make a note on the paper. Do not give an on-the-spot explanation. You won't be there to do that when Reviewers or ultimate Readers of the paper are going through this experience.
If they become totally confused and disinterested at any stage, abandon the reading at that point. Otherwise continue reading to the bitter end.
Now go away and edit the paper to avoid that person being puzzled at those points.
Come back and try on the same person or different people. You need to (start with!) a lot of friends. Don't worry, you can do the same favour for them in return.
When you have got the paper to a stage which allows it to be read without causing bemusement/crying on behalf of the listener, try giving it on paper to people. Again, don't ask them to send written comments back. Watch them while they read it.
Check their understanding by taking the paper away from them and asking them to describe what it said. Don't give them hints.
Success
Once people seem to be able to read it and understand it, you are done! Time to submit, and let the Journal's Reviewers read it and send their thoughts.
