There's a specific breed of frustration that only hits after weeks of preparation. The one-sentence summary is done. The paragraph summary, done. Character charts, four-page outline -- all done. The entire story lives in meticulous documents, organized, cross-referenced, ready. And then the blank page opens. Cursor blinks. Nothing comes.
Not because the story isn't there. It is. But between a four-page outline and an actual first draft, there's a gap that no amount of planning automatically closes. The outline says "protagonist enters the new world." That's a fine structural note. It is not a scene. It doesn't tell the writer what the first line of that scene sounds like, how long it runs, where to break for the next one.
The Snowflake Method's final four steps exist precisely to bridge this gap -- to transform planning into a blueprint so granular that sitting down each morning becomes execution, not invention from scratch.
Step Seven: Expand Your Character Charts -- Let Each Character Speak
"I don't know what my characters will do until I let them tell me themselves." Anne Lamott wrote that in Bird by Bird. She wasn't being mystical. She was describing something every writer who has spent real time with a character eventually discovers -- the moment a character develops their own voice, the story cracks open in directions the author never predicted.
By step seven, six steps of foundation are already in place: one-sentence summary, paragraph skeleton, character charts, four-page outline. But all of these share a blind spot.
They're written from the author's perspective. Omniscient. Top-down. God's-eye view. Characters in these documents are objects being described, not people doing the talking.
Step seven asks one thing: let each major character write a one-page storyline in first person. Not the author saying "he experienced this." The character opening their mouth: "This is my story."
The difference? Try it once and the answer arrives immediately. From the protagonist's perspective, a love story might read: "I met her, fell in love, lost her, fell apart." Four verbs, a clean line. But shift to the love interest's point of view and the same timeline becomes: "Someone hired me to play a role. I didn't expect to fall for the person following me. I wanted to tell him the truth but couldn't open my mouth. All I wanted was for him to love the real me -- not the phantom someone manufactured."
Same plot. Two entirely different emotional corridors. And the point where those corridors intersect? That's often the most dramatically charged scene in the entire book.
Many writers discover three things during this step: conflicts surface that were never planned; transition scenes are needed between characters whose motivations don't quite connect; and certain character decisions that looked logical from above make no sense from the character's own vantage point.
Catching these problems now costs a few paragraphs of revision. Catching them during the first draft costs chapters.
In Slima's File Tree, create a character-storylines folder. Inside, one file per character -- protagonist-storyline.md, antagonist-storyline.md, and so on. Switch between them freely. Compare side by side. When a storyline stalls, press Cmd+Shift+A (Mac) or Ctrl+Shift+A (Windows) to open the AI Chat Panel. Paste the character chart and your initial draft of the storyline, then ask the AI to rewrite it in that character's voice.
The AI won't replace authorial judgment. But it can help locate a character's speaking rhythm -- that subtle, specific tonal signature that belongs to this person and no one else.
Step Eight: Scene List -- Cut Your Outline into Executable Units
An outline is a map. A scene list is the daily marching route.
That distinction determines whether writing the first draft feels like "moving with direction" or "getting lost again every morning." The outline shows the full landscape of the story, but the full landscape is too large to serve as a daily work order. The scene list slices that big map into the smallest narrative unit: the scene.
What is a scene? A continuous stretch of action happening at a specific time, in a specific place. Time changes -- new scene. Location changes -- new scene. That simple.
An 80,000-word novel runs roughly fifty to a hundred scenes. That number isn't a rule, just an empirical benchmark. The point is -- the scene list lets the writer "see" the entire story's architecture for the first time, like looking down at a city street grid from altitude.
The film industry does the same thing with breakdown sheets. Each scene gets a number, a location, a time of day, a cast list, a props list. The entire crew uses that sheet to know what to prepare for each shooting day. A novelist's scene list doesn't need that level of granularity, but the core logic is identical: convert the fuzzy word "story" into concrete to-do items.
Build a scene list in Slima using table format:
| Scene | Chapter | POV | Location | Time | Summary |
|-------|---------|-----|----------|------|---------|
| 1 | 1 | Protagonist | Rooftop | Night | During a chase, a colleague falls; protagonist hangs on the edge |
| 2 | 2 | Protagonist | Friend's apartment | Day | Protagonist explains their fear and recovery plan |
| ... | ... | ... | ... | ... | ... |
Once it's built, spend ten minutes scanning from top to bottom. Three consecutive scenes at the same location? Pacing might be stuck. Five scenes of nothing but dialogue? Missing the breathing room that physical action provides. Ten scenes of nonstop high-intensity conflict? The reader will be exhausted.
The scene list's greatest value isn't documentation -- it's exposure. Passages that "felt fine" at the outline stage reveal their structural imbalances the moment they're broken into scenes.
Use the AI Chat Panel to check rhythm:
Here is my scene list. Please analyze rhythm issues: Are there too many consecutive slow or fast scenes? Are there scenes with unclear purpose? Is POV switching too frequent?
[paste scene list]
Step Nine: Scene Descriptions -- A Micro-Outline for Each Scene
The scene list answers "what scenes exist." Scene descriptions answer "how to write each scene."
For every scene, write roughly a hundred words of description. Not many words -- but they must precisely answer four questions.
What happens? Concrete action. "The protagonist follows the target to the museum, watches her stare at a portrait for twenty minutes." Not "the protagonist investigates" -- that's too abstract, useless when the draft-writing moment arrives.
What's the purpose? This scene's function inside the story machine. Does it advance the plot? Reveal character? Build atmosphere? Plant foreshadowing? A scene can serve multiple purposes simultaneously, but it needs at least one. If no purpose can be found, the scene probably shouldn't exist.
Where's the conflict? Every scene's energy source. External conflict -- a chase, an argument, a standoff. Internal conflict -- hesitation, fear, moral struggle. A scene without conflict is a car without an engine. It takes up space. It goes nowhere.
How does the situation change? When the scene ends, is the world different from how it began? If everything is identical, the scene didn't push the story forward. Good scenes make something better, or worse. Stasis is not an option.
In Slima, use Split Window (Cmd+ or Ctrl+) while writing scene descriptions. Scene list on the left, description file on the right. Side by side, context stays intact.
After finishing a scene description, hand it to the AI Chat Panel for a structural check:
Please check if this scene description fully answers four core questions: What happens, what's the purpose, where's the conflict, how does the situation change. If anything is missing, point it out and suggest additions:
[paste scene description]
Step Ten: Write the First Draft -- Preparation Finally Becomes Action
E.L. Doctorow once said: "Writing a novel is like driving at night. The headlights only illuminate a few dozen meters ahead, but you can drive the entire way like that."
The Snowflake Method's contribution is this: before setting out, it spreads the satellite map of the whole road across the table.
By step ten, the inventory is staggering. Story core (one sentence). Structural skeleton (one paragraph plus four-page outline). Character souls (summaries and charts). Each character's subjective lens (first-person storylines). Execution blueprint (scene list plus scene descriptions).
This is Hitchcock-style preparation. He once said: "The film is finished. I just haven't shot it yet." After nine steps, the story is already complete inside the writer's head. The first draft is simply the act of writing it down.
Open the first scene description. What happens -- known. Purpose -- known. Conflict -- known. How the scene ends -- known. No need to "figure out" what comes next. Just expand the blueprint into full prose.
That's why the Snowflake Method accelerates writing speed. The agony of staring at a blank page with zero ideas -- eliminated. The morning routine becomes: open scene list, locate today's scene, open description, write. No room for hesitation.
But -- thorough preparation doesn't mean zero surprises.
While actually writing, characters will say dialogue that never appeared in any plan. Details will grow out of sentences on their own. A plotline that seemed straight will suddenly curve. These surprises aren't loss of control. They're the best part of writing. The outline is the map. But walking the road, the roadside scenery is always new.
What if the outline turns out to be wrong mid-draft?
Change it.
The outline isn't a prison. It's scaffolding. Scaffolding exists to support construction, not to dictate the building's shape. A better path appears -- take it. No outline is worth sacrificing a better story for.
In Slima, commit a Snapshot via Version Control (Cmd+Shift+G or Ctrl+Shift+G) after completing each scene. Two benefits: first, the ability to rewind to any scene's state at any time; second, each commit is a psychological milestone -- the feeling of "one more done" that propels the next scene. Want to try a different approach? Create a new Branch to experiment. The main line stays untouched. If the experiment fails, delete the branch. If it works, merge it back.
The Snowflake Method's Flexibility
Randy Ingermanson never claimed this was the only correct path.
Some writers start their first draft after step four. The four-page outline is enough -- they figure out the rest as they go. Some skip the scene list entirely, jumping straight from outline to scene descriptions. Some finish all ten steps in two weeks. Some take two months.
None of these are mistakes.
The key isn't rigid compliance with each step. It's understanding what each step is for. Once the function is clear, the writer knows when to skip and when skipping would be costly. Character motivations unclear? Skipping step seven will cause problems. Plot structure rock-solid? The scene list might be simplified.
The steps aren't one-directional either. Writing scene descriptions and spotting a plot hole? Go back to the four-page outline and repair it. Writing the draft and feeling character motivation fall flat? Return to the character chart and dig deeper. The Snowflake Method isn't a straight line. It's a spiral -- at any stage, return to the core, adjust, then expand outward again.
Ingermanson's suggested timeline: a few days to a few weeks for steps one through nine, a few months for the first draft. But that's a reference, not a mandate. Writing rhythm is something only the writer's own body knows the answer to.
Moving Forward from Here
The Snowflake Method's core, in a single sentence: from simple to complex, expanding step by step.
One sentence becomes a paragraph. A paragraph becomes a four-page outline. The outline becomes a scene list. The scene list becomes scene descriptions. Scene descriptions become the first draft. Each step more concrete, richer, closer to the finished work than the one before it.
Like a snowflake growing from a crystal nucleus, branching outward into increasingly intricate structures. The story does the same. The core idea is that nucleus. Each step is a new layer of branches.
This series has walked through the complete Snowflake Method process. From story core to skeleton. From character summaries to charts. From expanded summary to four-page outline. From scene list to scene descriptions. Every step pushed the story from "idea" closer to "work."
The planning is over now.
Open Slima. Open the first scene description. Set a daily word count target in Writing Goals. Let Writing Streak track consecutive writing days. These aren't flashy features -- they're the tools that turn "sit down and write every day" into a habit.
Start writing. Don't wait for the outline to be perfect -- a perfect outline doesn't exist. "Good enough" is the signal to begin.
The story is waiting.