Do you have a strategy when you start writing a manuscript? Many people do as there is an order that makes the writing process a lot easier and will help you write a logical and consistent manuscript.
Here is the manuscript writing order many authors use, and I will explain below why this order works well.
1. Tables & Figures
2. Results & Figure Legends
Start with preparing the Tables and Figures. While preparing these, it will quickly become apparent whether additional experiments are needed. It will also facilitate discussion with co-authors or other interested people and if you need to present your data at a (department) meeting you have these ready.
Now, in the Results, carefully describe the data shown in the Tables and Figures. Use roughly one section (one or more paragraphs) per Figure or Table. Start each section with one or two sentences reminding the reader why a particular experiment was performed, and what the experiment involved. Give each section a relevant subheading that summarizes in one sentence the main result. Write the Figure Legends next: each legend should explain the experiment performed, which samples were included, etc and should have a short title that gives away the main conclusion of the Figure. Try to make sure that the reader can understand the research by just looking at the Figures and Legends. And, vice versa, try to make sure the reader can understand the outcome of the experiments in the Results section without having to look up the Figures.
After preparing the Figures and Tables and writing the Results you know which experiments are included in the manuscript so you can write the Materials and Methods. It does not make sense to write the Methods earlier as often many experiments performed along the way are not included in the manuscript. Again, start each paragraph by explaining in one sentence the rationale for performing that experiment.
Now that it is clear where the results are heading you can write the Introduction. First, describe the broader problem and gradually zoom into the specific aspect of the problem you worked on and what your research question was. Then outline in a few sentences what approach you used to answer the question.
In the Discussion start with a short paragraph highlighting your main finding(s). Subsequently discuss the conclusions from all experiments and how they tie in with the existing literature and how the existing literature modifies, confirms, extends, or contradicts the conclusions. Also, discuss the pitfalls of your research. At the end of the discussion come back to the broader picture you sketched in the introduction to explain to the reader how the results will aid in solving this problem. Some journals demand you add a Conclusion section at the end. If they do not, you can add one anyway, just without the label.
Now that the main text of the manuscript is complete write the Abstract. And, last but not least, think about a good Title that gives away the main finding. The Abstract and Title are based on all other sections so it is best to write these last.
General tips for manuscript writing:
Add a working title to the manuscript from the start but make sure it is preceded by ´Working title:´ so you do not forget to rewrite it at the end.
Save every version of the manuscript; you may need to go back to an earlier version to retrieve text you threw out, or to check comments in the margins.
While writing, automatically add the References using a reference management program, this prevents you from wasting time in the end because you do not remember where you read certain information and need to reread many articles to find it.
From the start keep the order of the sections in your manuscript as follows: Title page, Abstract, Introduction, Materials & Methods, Results, Discussion, References, Tables, Legends to Figures (yes, on a page separate from the Figures), and Figures. This is the order in which the journal usually wants you to submit them. I have seen many articles where the author kept the Figures and Tables in the Results section. This often resulted in a Results section that was not properly describing the data in the Figures and Tables (because the writer would all the time see the data, thus removing the need for proper description), and that was too meager.