How to Write a Novel: The Snowflake Method

Writing a novel is easy. Writing a good novel is hard. That's just life. If it were easy, we'd all be
writing best-selling, prize-winning fiction.
Frankly, there are a thousand different people out there who can tell you how to write a novel. There
are a thousand different methods. The best one for you is the one that works for you.
In this article, I'd like to share with you what works for me. I've published six novels and won about a
dozen awards for my writing. I teach the craft of writing fiction at writing conferences all the time. One
of my most popular lectures is this one: How to write a novel using what I call the "Snowflake
Method."
This page is the most popular one on my web site, and gets over a thousand page views per day, so
you can guess that a lot of people find it useful. But you may not, and that's fine by me. Look it over,
decide what might work for you, and ignore the rest! If it makes you puke, I won't be insulted.
Different writers are different. If my methods get you rolling, I'll be happy. I'll make the best case I can
for my way of organizing things, but you are the final judge of what works best for you. Have fun and .
. . write your novel!
The Importance of Design
Good fiction doesn't just happen, it is designed. You can do the design work before or after you
write your novel. I've done it both ways and I strongly believe that doing it first is quicker and leads to
a better result. Design is hard work, so it's important to find a guiding principle early on. This article
will give you a powerful metaphor to guide your design.
Our fundamental question is this: How do you design a novel?
For a number of years, I was a software architect designing large software projects. I write novels the
same way I write software, using the "snowflake metaphor". OK, what's the snowflake metaphor?
Before you go further, take a look at this cool web site.

At the top of the page, you'll see a cute pattern known as a snowflake fractal. Don't tell anyone, but
this is an important mathematical object that's been widely studied. For our purposes, it's just a cool
sketch of a snowflake. If you scroll down that same web page a little, you'll see a box with a large
triangle in it and arrows underneath. If you press the right-arrow button repeatedly, you'll see the
steps used to create the snowflake. It doesn't look much like a snowflake at first, but after a few
steps, it starts looking more and more like one, until it's done.
The first few steps look like this:
I claim that that's how you design a novel -- you start small, then build stuff up until it looks like a
story. Part of this is creative work, and I can't teach you how to do that. Not here, anyway. But part of
the work is just managing your creativity -- getting it organized into a well-structured novel. That's
what I'd like to teach you here.
If you're like most people, you spend a long time thinking about your novel before you ever start
writing. You may do some research. You daydream about how the story's going to work. You
brainstorm. You start hearing the voices of different characters. You think about what the book's
about -- the Deep Theme. This is an essential part of every book which I call "composting". It's an
informal process and every writer does it differently. I'm going to assume that you know how to
compost your story ideas and that you have already got a novel well-composted in your mind and
that you're ready to sit down and start writing that novel.
The Ten Steps of Design
But before you start writing, you need to get organized. You need to put all those wonderful ideas
down on paper in a form you can use. Why? Because your memory is fallible, and your creativity has
probably left a lot of holes in your story -- holes you need to fill in before you start writing your novel.
You need a design document. And you need to produce it using a process that doesn't kill your
desire to actually write the story. Here is my ten-step process for writing a design document. I use
this process for writing my novels, and I hope it will help you.
Step 1) Take an hour and write a one-sentence summary of your novel. Something like this: "A rogue
physicist travels back in time to kill the apostle Paul." (This is the summary for my first novel,
Transgression.) The sentence will serve you forever as a ten-second selling tool. This is the big
picture, the analog of that big starting triangle in the snowflake picture.
When you later write your book proposal, this sentence should appear very early in the proposal. It's
the hook that will sell your book to your editor, to your committee, to the sales force, to bookstore
owners, and ultimately to readers. So make the best one you can!
Some hints on what makes a good sentence:
 Shorter is better. Try for fewer than 15 words.
 No character names, please! Better to say "a handicapped trapeze artist" than "Jane Doe".
 Tie together the big picture and the personal picture. Which character has the most to lose in
this story? Now tell me what he or she wants to win.
 Read the one-line blurbs on the New York Times Bestseller list to learn how to do this. Writing a
one-sentence description is an art form.
Step 2) Take another hour and expand that sentence to a full paragraph describing the story setup,
major disasters, and ending of the novel. This is the analog of the second stage of the snowflake. I
like to structure a story as "three disasters plus an ending". Each of the disasters takes a quarter of
the book to develop and the ending takes the final quarter. I don't know if this is the ideal structure,
it's just my personal taste.
If you believe in the Three-Act structure, then the first disaster corresponds to the end of Act 1. The
second disaster is the mid-point of Act 2. The third disaster is the end of Act 2, and forces Act 3
which wraps things up. It is OK to have the first disaster be caused by external circumstances, but I
think that the second and third disasters should be caused by the protagonist's attempts to "fix
things". Things just get worse and worse.
You can also use this paragraph in your proposal. Ideally, your paragraph will have about five
sentences. One sentence to give me the backdrop and story setup. Then one sentence each for your
three disasters. Then one more sentence to tell the ending. If this sounds suspiciously like back-
cover copy, it's because . . . that's what it is and that's where it's going to appear someday.
Step 3) The above gives you a high-level view of your novel. Now you need something similar for the
storylines of each of your characters. Characters are the most important part of any novel, and the
time you invest in designing them up front will pay off ten-fold when you start writing. For each of your
major characters, take an hour and write a one-page summary sheet that tells:
 The character's name
 A one-sentence summary of the character's storyline
 The character's motivation (what does he/she want abstractly?)
 The character's goal (what does he/she want concretely?)
 The character's conflict (what prevents him/her from reaching this goal?)
 The character's epiphany (what will he/she learn, how will he/she change?
 A one-paragraph summary of the character's storyline
An important point: You may find that you need to go back and revise your one-sentence summary
and/or your one-paragraph summary. Go ahead! This is good--it means your characters are teaching
you things about your story. It's always okay at any stage of the design process to go back and revise
earlier stages. In fact, it's not just okay--it's inevitable. And it's good. Any revisions you make now are
revisions you won't need to make later on to a clunky 400 page manuscript.
Another important point: It doesn't have to be perfect. The purpose of each step in the design
process is to advance you to the next step. Keep your forward momentum! You can always come
back later and fix it when you understand the story better. You will do this too, unless you're a lot
smarter than I am.
Step 4) By this stage, you should have a good idea of the large-scale structure of your novel, and
you have only spent a day or two. Well, truthfully, you may have spent as much as a week, but it
doesn't matter. If the story is broken, you know it now, rather than after investing 500 hours in a
rambling first draft. So now just keep growing the story. Take several hours and expand each
sentence of your summary paragraph into a full paragraph. All but the last paragraph should end in a
disaster. The final paragraph should tell how the book ends.
This is a lot of fun, and at the end of the exercise, you have a pretty decent one-page skeleton of
your novel. It's okay if you can't get it all onto one single-spaced page. What matters is that you are
growing the ideas that will go into your story. You are expanding the conflict. You should now have a
synopsis suitable for a proposal, although there is a better alternative for proposals . . .
Step 5) Take a day or two and write up a one-page description of each major character and a half-
page description of the other important characters. These "character synopses" should tell the story
from the point of view of each character. As always, feel free to cycle back to the earlier steps and
make revisions as you learn cool stuff about your characters. I usually enjoy this step the most and
lately, I have been putting the resulting "character synopses" into my proposals instead of a plot-
based synopsis. Editors love character synopses, because editors love character-based fiction.
Step 6) By now, you have a solid story and several story-threads, one for each character. Now take a
week and expand the one-page plot synopsis of the novel to a four-page synopsis. Basically, you will
again be expanding each paragraph from step (4) into a full page. This is a lot of fun, because you
are figuring out the high-level logic of the story and making strategic decisions. Here, you will
definitely want to cycle back and fix things in the earlier steps as you gain insight into the story and
new ideas whack you in the face.
Step 7) Take another week and expand your character descriptions into full-fledged character charts
detailing everything there is to know about each character. The standard stuff such as birthdate,
description, history, motivation, goal, etc. Most importantly, how will this character change by the end
of the novel? This is an expansion of your work in step (3), and it will teach you a lot about your
characters. You will probably go back and revise steps (1-6) as your characters become "real" to you
and begin making petulant demands on the story. This is good -- great fiction is character-driven.
Take as much time as you need to do this, because you're just saving time downstream. When you
have finished this process, (and it may take a full month of solid effort to get here), you have most of
what you need to write a proposal. If you are a published novelist, then you can write a proposal now
and sell your novel before you write it. If you're not yet published, then you'll need to write your entire
novel first before you can sell it. No, that's not fair, but life isn't fair and the world of fiction writing is
especially unfair.
Step 8) You may or may not take a hiatus here, waiting for the book to sell. At some point, you've got
to actually write the novel. Before you do that, there are a couple of things you can do to make that
traumatic first draft easier. The first thing to do is to take that four-page synopsis and make a list of all
the scenes that you'll need to turn the story into a novel. And the easiest way to make that list is . . .
with a spreadsheet.
For some reason, this is scary to a lot of writers. Oh the horror. Deal with it. You learned to use a
word-processor. Spreadsheets are easier. You need to make a list of scenes, and spreadsheets were
invented for making lists. If you need some tutoring, buy a book. There are a thousand out there, and
one of them will work for you. It should take you less than a day to learn the itty bit you need. It'll be
the most valuable day you ever spent. Do it.
Make a spreadsheet detailing the scenes that emerge from your four-page plot outline. Make just one
line for each scene. In one column, list the POV character. In another (wide) column, tell what
happens. If you want to get fancy, add more columns that tell you how many pages you expect to
write for the scene. A spreadsheet is ideal, because you can see the whole storyline at a glance, and
it's easy to move scenes around to reorder things.
My spreadsheets usually wind up being over 100 lines long, one line for each scene of the novel. As I
develop the story, I make new versions of my story spreadsheet. This is incredibly valuable for
analyzing a story. It can take a week to make a good spreadsheet. When you are done, you can add
a new column for chapter numbers and assign a chapter to each scene.
Step 9) (Optional. I don't do this step anymore.) Switch back to your word processor and begin
writing a narrative description of the story. Take each line of the spreadsheet and expand it to a multi-
paragraph description of the scene. Put in any cool lines of dialogue you think of, and sketch out the
essential conflict of that scene. If there's no conflict, you'll know it here and you should either add
conflict or scrub the scene.
I used to write either one or two pages per chapter, and I started each chapter on a new page. Then I
just printed it all out and put it in a loose-leaf notebook, so I could easily swap chapters around later
or revise chapters without messing up the others. This process usually took me a week and the end
result was a massive 50-page printed document that I would revise in red ink as I wrote the first draft.
All my good ideas when I woke up in the morning got hand-written in the margins of this document.
This, by the way, is a rather painless way of writing that dreaded detailed synopsis that all writers
seem to hate. But it's actually fun to develop, if you have done steps (1) through (8) first. When I did
this step, I never showed this synopsis to anyone, least of all to an editor -- it was for me alone. I liked
to think of it as the prototype first draft. Imagine writing a first draft in a week! Yes, you can do it and
it's well worth the time. But I'll be honest, I don't feel like I need this step anymore, so I don't do it
now.
Step 10) At this point, just sit down and start pounding out the real first draft of the novel. You will be
astounded at how fast the story flies out of your fingers at this stage. I have seen writers triple their
fiction writing speed overnight, while producing better quality first drafts than they usually produce on
a third draft.
You might think that all the creativity is chewed out of the story by this time. Well, no, not unless you
overdid your analysis when you wrote your Snowflake. This is supposed to be the fun part, because
there are many small-scale logic problems to work out here. How does Hero get out of that tree
surrounded by alligators and rescue Heroine who's in the burning rowboat? This is the time to figure
it out! But it's fun because you already know that the large-scale structure of the novel works. So you
only have to solve a limited set of problems, and so you can write relatively fast.
This stage is incredibly fun and exciting. I have heard many fiction writers complain about how hard
the first draft is. Invariably, that's because they have no clue what's coming next. Good grief! Life is
 too short to write like that! There is no reason to spend 500 hours writing a wandering first draft of
your novel when you can write a solid one in 150. Counting the 100 hours it takes to do the design
documents, you come out way ahead in time.
About midway through a first draft, I usually take a breather and fix all the broken parts of my design
documents. Yes, the design documents are not perfect. That's okay. The design documents are not
fixed in concrete, they are a living set of documents that grows as you develop your novel. If you are
doing your job right, at the end of the first draft you will laugh at what an amateurish piece of junk
your original design documents were. And you'll be thrilled at how deep your story has become.
<POver the years, I've taught the Snowflake method to hundreds of writers at conferences. I've also had
this article posted here on my web site for a long time, and the page has now been viewed over
1,500,000 times. I've heard from many, many writers. Some people love the Snowflake; some don't. My
attitude is that if it works for you, then use it. If only parts of it work for you, then use only those parts.
I write my own novels using the Snowflake method. Make no mistake -- it's a fair bit of work. For a
long time, I did it the hard way, using Microsoft Word to write the text and Microsoft Excel to manage
the list of scenes. Unfortunately, neither of those tools knows about the structure of fiction. Finally, I
realized that it would be a whole lot easier to work through the method if the tools were designed
specially for fiction.
So one day I decided to create that software. I wanted something that would automate every step
that could be automated. The result was a commercial software package I call Snowflake Pro. It
makes my own Snowflaking incredibly easier, and it's now doing the same for zillions of other writers