After finishing three novels, each over a million words, using Scrivener, I’ve developed a deep appreciation for its strengths—especially its document structure—but also a growing sense of friction in certain parts of my workflow.
One feature I absolutely love is the outliner editor. My typical setup is to keep the outliner open on the left, while placing the editor on the right in Scrivenings mode. This allows me to read through an entire chapter continuously, while still being able to jump to specific sections by clicking individual documents in the outliner.
Before I start writing, I usually spend time planning. I outline the chapter structure, break it down into smaller sections, and arrange pre-written materials under the current chapter. I reorder them to establish a logical flow and assign them to specific sections. Outside of dedicated writing time, I often capture ideas on my phone—quick notes, fragments, or inspirations—and later organize them on my computer. These ideas don’t always belong to the current chapter; sometimes they are for future parts of the story.
For me, all of these are essentially TO DO items. I typically write them as short titles, and if needed, expand them with a synopsis.
However, when it comes to turning ideas into actual prose, Scrivener’s document model starts to feel slightly inconvenient.
The issue is that these scattered ideas ultimately need to be integrated into a coherent narrative. This often requires merging documents. At the same time, narrative effects—such as character development, foreshadowing, or thematic reinforcement—are rarely isolated. A single passage might serve three to five different purposes simultaneously.
Because of this, I can’t simply write each “effect” in isolation and then stitch them together later. Instead, I have to implement them collectively within a single document. Once done, I summarize the achieved effects in the synopsis and delete the original placeholder documents.
Similarly, some narrative effects need to be repeated across multiple sections to reinforce the reader’s impression of a character or theme.
As a result, my workflow constantly involves:
- rewriting synopses
- moving documents around
- and deleting temporary documents
Many of these deleted documents never become part of the final text. They exist purely as intermediate structures to help me think and plan, but because their intended effects are interwoven into actual scenes, they never survive into the manuscript.
This made me realize something:
what I might actually need is a document model with built-in TO DO functionality.
If each narrative “effect” could exist as a TO DO item within the document structure:
- I wouldn’t need external tools to track them
- I wouldn’t need to create and delete placeholder documents
- I could simply check them off once implemented
Completed items could be automatically hidden or shown based on preference. Some TO DOs that don’t yet belong to a specific section could remain at the parent level, and only be moved into a section when I’m ready to write. In some cases, they might even temporarily exist as their own “logical section” before being merged into actual text. These TO DO items could also be easily moved across different documents and folders as the structure evolves.
In short, instead of managing documents, I would be managing intentions—and letting the writing naturally absorb them.
I’m curious if others who write long-form fiction in Scrivener have run into similar issues, or if you’ve found alternative workflows or tools that better support this kind of process.
