The Reverse Outline: A Scene-by-Scene Guide
What to actually write for each scene, what patterns to look for across the full manuscript, and what to do before you open the draft to revise.
The concept is simple: read your draft, write down what you find, build the outline retroactively. The part writers usually get wrong is what to write down.
The most common failure mode of a reverse outline isn't structural or mechanical. It's that writers fill it with descriptions of what they intended to write rather than what's on the page. The outline looks complete. It tells you nothing useful. You close the document and go back to revising by feel, which is where you started.
This is the guide to not doing that.
Three things to track for every scene
You don't need a complicated format. You need to track three things per scene — things that, taken together, will surface problems you cannot see while rereading your manuscript normally. Work scene by scene, not chapter by chapter. Scenes within the same chapter can have completely different functions, and collapsing them loses the resolution you need.
Scene event
What literally happens. Not the theme. Not the subtext. Not what the scene is "really about." Observable action: who does what, who says what, what changes hands, what is decided.
The test for whether you've written this correctly: could you describe it to someone who hasn't read the book without using your knowledge of what you intended?
Too vague: Mara and her father have a difficult conversation about family secrets.
Specific enough: Mara confronts her father about the letter. He denies writing it. She doesn't believe him but leaves without pressing further.
The first entry describes the category of scene. The second describes what actually happens — and from it you can already see a question worth asking: Mara retreats without pressing. Is that a choice that costs her something, or is the scene going soft?
Protagonist agency
Who drives this scene? And specifically: is your protagonist deciding something, or watching others decide?
Write "Active," "Passive," or a brief note on how it's mixed. "Active (initiates confrontation, then retreats)" tells you more than a checkbox.
This column exists because one of the most common things a reverse outline surfaces is a protagonist who has quietly become a spectator in their own story. It happens gradually — a scene where someone else makes the key decision, a scene where your hero mostly observes — and it's almost impossible to feel from inside the draft. For discovery writers, who build character through forward motion, this is often the most important column in the whole document. Written down across fifty scenes, the pattern becomes impossible to ignore.
Value change
What is different at the end of this scene from the beginning? Name the poles: Mara had certainty, now she has doubt. Daniel trusted Anna, now he doesn't. The investigation was stalled, now it has a lead.
If nothing changed, write: No change.
That entry is not a failure of your outline. It is the outline doing its job. Scenes where nothing changes are the scenes draining momentum from the rest of the manuscript. You need to know they're there. You may decide they're worth keeping for other reasons. But you cannot make that decision if you haven't named the problem.
What you'll find
The patterns that emerge from these three columns, read across a full manuscript, are the structural story your draft has been trying to tell you.
The most common finding: a protagonist who is active in the first act and increasingly passive by the middle. They stop initiating and start reacting. The forward pressure that made the early chapters work quietly disappears, and the draft starts to feel like things are happening to a person rather than a person making things happen.
The second most common: clusters of "no change" entries in the same section of manuscript — usually the middle. This is rarely about the individual scenes. It's about what the draft needed to do in that stretch and didn't. The scenes aren't broken. The structure around them is.
You'll also find threads that vanish. A character whose interiority goes quiet for forty pages. A promise made in chapter three that never gets paid off. A confrontation set up and then never arrived at. None of these are visible in a normal reread because you know they're coming — or you know they were supposed to come. The outline shows you what's actually on the page.
What to do with it before you open the manuscript
The mistake writers make after finishing a reverse outline is opening the manuscript immediately. Don't.
Read the outline. Find the patterns. Then write the revision plan — what to cut, what to combine, what to add, where to move things — before you touch a single scene. Test changes on the outline first. Ask: if I cut this passive scene in chapter six, what does chapter seven now need to do? Work that out in the outline. It takes twenty minutes. Working it out in the manuscript, by trial and error, takes days.
The revision plan doesn't have to be exhaustive. It has to exist. Going into the manuscript with a list of three structural things you know need to change is better than going in with a vague sense that something is wrong. The outline gives you the list.
The thing that kills a reverse outline
Every editor and author who writes seriously about this technique names the same failure mode: describing the scene you intended to write rather than the scene that exists.
It looks like this: This scene seems slow, but it's building toward the confrontation in chapter twelve. That may be true. It doesn't belong in the outline. What belongs in the outline is what's on the page. If what's on the page is slow, write "slow." The interpretation belongs in the revision plan, not the diagnosis.
If you soften the entries, you build a document that reflects your intentions rather than your draft. That document will not help you revise. It will confirm what you already believed about the manuscript, which is the problem you were trying to solve.
The whole value of the exercise is that it makes the draft legible from the outside. That only works if you read it honestly.
The conceptual case for why this works is a separate piece. If you've read it and you're ready to build the outline, you have what you need above. If you'd rather start with a generated version and work from there — adding to it, pushing back on it, using it as a foundation — discowriter builds the first draft of the outline from your manuscript automatically.