How to Write a Novel: A Working Novelist's Honest Guide
The worst chapter I ever wrote was chapter eleven of my second novel. I had been at the desk for nine months -- 74,000 words in, three acts sketched out on a whiteboard, the whole back half of the story still ahead of me. Chapter eleven was the pivot. The scene I had been picturing since before I started drafting: the moment everything the first half had been building toward finally cracked open and fell the wrong way. I had thought about it on walks. I had half-dreamed it. When I finally sat down to write it, I produced, over four consecutive mornings, eleven pages of the most mechanical, airless prose I had made in years. Every sentence did its job. None of it breathed.
I rewrote those pages six times over the following month. The seventh version was the one that went to my editor. It was the best chapter in the book.
The lie that derails most first novelists is that the process should feel inspired. That flow means you are on the right track, and struggle means you have taken a wrong turn somewhere. Stephen King, in On Writing, describes writing 2,000 words every day including his birthday and Christmas -- not because every day is good, but because the discipline is what makes the good days possible. Ursula K. Le Guin described first drafts as "putting it down, getting it wrong, trying again." Margaret Atwood has compared writing a novel to driving at night with headlights that only illuminate a few feet ahead -- you can make the whole journey that way, but you cannot see the destination while you are in it. None of them is describing a process that feels inspired. They are describing a process that is often hard, often discouraging, and sometimes profoundly unrewarding -- until it is not.
This is a complete guide to that process. Not the romantic version. The one that works.
Before You Start: The Readiness Question Nobody Asks
The version of this question most first-time novelists ask is: "Am I ready to write a novel?" This is the wrong question. There is no preparation sufficient to make the first novel feel natural. The novel is the preparation. You will learn things during the writing of your first novel that you could not have learned any other way -- things about scene structure, about character consistency, about how to handle the moment when your plot logic collapses three-quarters of the way through and you have to rebuild it from underneath while the story is still standing. You cannot learn these things from reading about them. You can only learn them by being inside them, confused, at two in the afternoon, trying to figure out what went wrong.
The real question is: "Do I have enough of an idea to start?" And the bar for that is lower than you think.
You need a character you want to spend a year with. You need a situation that generates questions you do not yet know the answers to. You need one scene -- not a full outline, not a complete arc, just one scene that feels alive when you picture it. Anne Lamott's permission slip, from Bird by Bird, is to write the "shitty first draft" -- not the good one, not the finished one, just the one that exists. The draft that teaches you what the novel is trying to be. That draft cannot exist until you write it. So you write it. The case for not waiting is this: every month you do not start is a month you do not learn.
The novel is the education.
When an Idea Is Novel-Shaped
Not every story idea is a novel. Some are short stories. Some are novellas. Some are essays. Part of the pre-writing work is figuring out which shape your idea actually is -- because choosing the wrong form will produce either a padded short story masquerading as a novel, or a compressed novel doing its best work inside a short story frame.
The rough test: a novel idea generates more questions than it answers. A short story idea answers one question, usually about a character and usually about one moment or discovery. Donna Tartt's The Secret History opens with the narrator telling you the murder has already happened, in the first paragraph. That is not a short story idea. It is a question that takes 600 pages to answer, because the real question is not who did it (you know) but how a group of people became capable of it, and how they live in the aftermath. Every chapter generates new sub-questions. That shape is the shape of a novel.
A story with a single, clean reversal is usually a short story. A story with a protagonist who must change over time, across multiple encounters with the same resistance, is a novel. If you cannot picture three distinct phases of change in your main character -- a beginning state, a different middle state, an end state that could not have been predicted from the beginning -- you may have a short story. That is fine. Write the short story. Let it become a novel if it needs to. Do not pad it into one because you want to write a novel.
The Pitch: Testing the Spine Before You Build the Body
Before you outline, before you set up your project structure, before you write the first scene, write the pitch. Not for agents or publishers -- for yourself. A single sentence: character, situation, what is at stake. Not theme, not genre, not "this will be like X meets Y." Who is this person, what is the impossible situation they are in, and what stands to be lost?
If you cannot write that sentence, you do not know what the novel is about yet. That is not a problem. It is useful information. Most writers do not know what a novel is about until they have written the first draft, and the pitch sentence clarifies over time. But trying to write it early, even badly, forces a useful crisis. "A woman discovers her sister is alive" is a situation but not a spine. "A woman discovers her sister is alive and must choose between her own rebuilt life and the truth" is closer. "A woman who has spent fifteen years becoming someone new discovers the sister she let the world think was dead, and must decide whether the person she is now can survive the return of the person she was" -- that is a spine, and it will hold a novel.
After the one-sentence pitch, write the one-paragraph version. Expand the stakes. Add the second act problem. Name what the character most wants and what stands between her and it. Write it badly. Then make it better. Then write the novel.
The Plan: How Much Structure Is Actually Enough
The outliner versus pantser debate is one of the most reliably unproductive arguments in fiction writing. Pantsers -- "I write by the seat of my pants, discovering the story as I go" -- often produce first drafts that need structural surgery. Outliners often produce drafts that are technically sound but feel flat because every surprise was spent in the planning. Both failures are common. Neither approach is the approach.
What working novelists actually do is somewhere in the middle, calibrated to their temperament and their kind of story. King is famously against outlines; he starts with a situation and follows his characters. Brandon Sanderson outlines in detail because his plot structures require the reader to know magic rules that must be seeded early to pay off late. Atwood has described planning the "weather" and "landscape" of a book -- its emotional geography -- without plotting scenes beat-by-beat. These are not incompatible. They are calibrated.
The case for some structure: knowing your ending before you start changes what you plant in the first act. Not the ending scene necessarily, but the emotional truth your protagonist will have arrived at. If you know your character will end the book having accepted the loss of the person she loved, you seed that arc differently than if you are still deciding. The ending does not have to be fixed -- many novelists find their endings in the writing -- but an intention, however provisional, gives the first draft direction.
A useful minimal structure: know your beginning, know a rough midpoint reversal, know the emotional register of your ending. Everything else can be discovered. For a more detailed treatment of the available structural methods -- the three-act model, Save the Cat, the snowflake method, pantsing with purpose -- there is a full comparison in a post that weighs the tradeoffs of each for novelists working in different modes.
The Setup: Three Folders and a Clear Head
The novel needs a home before you start writing it. Not an elaborate home -- not a color-coded system with tags and status fields and a master spreadsheet. A simple home. Three top-level folders: one for the manuscript, one for the notes and thinking, one for cut material that is not ready to be gone forever. Within the manuscript folder, a separate document for each chapter. That is the whole system.
This sounds like housekeeping. It is not. The container shapes the work. A novelist who keeps everything in one long document thinks differently than one whose chapters live separately -- they navigate differently, revise differently, cut differently. The physical fact of having to open "Chapter 11" instead of scrolling to it changes the psychological relationship to that chapter. It can stand alone. It can be improved without touching the rest of the manuscript. It can be skipped for a day while you work elsewhere.
The full architecture -- three folders, chapter-level documents, the logic behind each layer, and why the Scraps folder makes it possible to cut without fear -- is worked through in a separate post on organizing a novel project. Set it up in fifteen minutes. Then close the sidebar and open chapter one.
The First Draft: Getting It Wrong on Purpose
The first draft is not the book. This is the thing most first-time novelists need to hear and cannot quite believe: the first draft is a very long, very rough sketch of what the book will eventually be. It is wrong in dozens of ways, many of which you will not see until you are in the third chapter of the second draft and you realize that the character in chapter two said something that makes no sense given what you now know about her. The first draft is allowed to be wrong. It is required to be wrong. The wrongness is part of how it teaches you the truth of what the book is trying to do.
King's daily quota is 2,000 words. Others write 500, or 1,000, or whatever fits their schedule and their pace. The number matters less than the consistency. Writing a novel is not a sprint, and it is not exactly a marathon -- it is more like farming. You show up every day and do the small daily work, and after enough days the harvest is there. Missing a day occasionally does not ruin the crop. Missing thirty days in a row does. The daily practice is what keeps the world of the novel alive in your head between sessions, and a world that goes cold is much harder to re-enter.
The 30-50% slump is real. At around the one-third mark of most first drafts, the initial momentum of beginning has worn off and the end is nowhere in sight. The middle is the longest part of the novel to write and the least structurally clear. The excitement of the opening has burned through, and the energy that comes from approaching the climax has not yet arrived. Many first novels die here. Not because the writer lacks the capacity to finish, but because the middle is genuinely difficult and nobody warned them it would feel like this.
What to do when you stall: do not go back and reread what you have. Not in the first draft. Rereading activates the editor's brain, which is not the brain you need for drafting. The editor's brain looks at what is on the page and sees problems. The drafter's brain asks what comes next. If you have lost the thread, write the next scene you can picture clearly -- not necessarily the next scene in order, just the next scene that feels alive. You can sort out sequence in revision. Momentum is what matters in the draft. Forward, even if the direction is slightly wrong.
One technique that helps many writers through the slump: write the scene you most want to write. The scene you have been looking forward to since before you started. This is not "skipping ahead" in any harmful sense -- it is investing in the draft's energy. A scene written with genuine engagement is better than a scene written dutifully, and finding your way back to engagement is more important than maintaining strict linear order.
The Middle: Where Most Novels Actually Die
The middle of a novel -- roughly the long stretch between the inciting incident and the climactic confrontation -- is where most first novels fail. Not for lack of talent. Not for lack of effort. Because the middle is structurally the hardest part to write, and most beginning novelists do not realize this until they are already inside it.
In the opening, you have the energy of beginning. New characters, new situation, the reader is orienting, and so are you. In the closing, you have the momentum of approaching the end. The climax pulls you forward. In the middle, you have neither. You have the ongoing problem -- the protagonist's want versus the obstacle -- and you have to sustain it for thirty, forty, fifty thousand words without losing either the reader's or your own sense that it is going somewhere meaningful.
The two most common failures: the "rubber band" problem, where the protagonist makes progress and then loses it and then makes progress again without escalation -- every scene feels like the same scene, slightly rearranged -- and the "empty middle," where subplots have not been developed enough to carry the reader through the rough patches in the main plot.
Subplots are not decoration. They are the structural load-bearing walls of the middle. They create secondary lines of tension that engage the reader when the main plot is in a transitional phase. They reveal character from angles the main plot cannot reach. They complicate the thematic question in ways the protagonist's story alone cannot sustain. A well-developed subplot arrives at its own resolution at the same time the main plot does, creating the sense of convergence and inevitability that makes a climax feel earned. The mechanics of this -- how to weave multiple story threads without losing the reader, how to calibrate their timing, how to make subplots mean something beyond their own resolution -- are worked through in a post on subplot architecture that is worth reading before you begin the structural revision.
The End: Getting Across the Finish Line
The last third of a novel has its own difficulties. The momentum that builds toward the climax is real -- I have felt it, the sense that the story is finally pulling me rather than me pushing it -- but the closing is also the place where everything you have set up has to pay off, and the gap between what you promised in the opening and what you are actually able to deliver in the closing can be painfully visible. You cannot fix this in the first draft.
Donna Tartt spent eleven years writing The Goldfinch. Not because the ending was hard -- because all of it was hard, and the ending needed the rest of the novel to already be right before it could be what it needed to be. The lesson from Tartt is not that you should spend eleven years on a novel. It is that the ending cannot be separated from the novel it ends, and writing it quickly out of exhaustion, or out of relief that you are almost done, will show.
Practically: in the first draft, just finish. Get across the line. A bad ending that exists is infinitely more useful than a perfect ending that does not. The ending will be revised. It may be entirely rewritten. The first draft's ending is a placeholder -- "this is where the story ends and this is roughly what it is saying" -- and that is enough for now.
The Revision: Three Passes That Actually Work
Revision is not proofreading. This distinction matters enormously, and most beginning writers do not learn it until after the first novel. Revision is structural and narrative work. Proofreading is surface work. They require different brains, different timings, and different tools -- and doing them in the wrong order is like painting before you have finished the carpentry.
The system that works is three passes, in order.
Pass one: structural. Print the draft or create a separate read-only copy. Read it straight through, quickly, taking notes but not line-editing. You are looking at the largest shapes: does the protagonist want something? Does the resistance escalate? Is the midpoint a genuine reversal or just a pause? Does the ending feel earned given what the first act set up? Write down every scene that is doing too much work (covering a gap that should not exist) or not enough (filler that could be cut). This pass usually takes a week or two and produces a list of structural problems. Resist the urge to start fixing until you have finished reading.
Pass two: scene-level. For each scene: what does the POV character want in this scene? What opposes that want? What changes between the beginning and the end? If the answer to the last question is "nothing," the scene is probably not earning its place. This is the most labor-intensive pass. A chapter might change by 40%, a scene might be cut entirely, a new scene might be added to cover a beat the structural pass identified as missing. This is also the pass where most of the real writing happens -- the writing-as-discovery work that you could not do in the first draft because you did not know enough yet about the story.
Pass three: line-level. Only after the structure is right and the scenes are doing their jobs. Read aloud -- not silently, aloud. Your ear will catch what your eye misses: repetition of words and sentence rhythms within a paragraph, the dead metaphor you have used so many times it no longer registers, the sentence that is syntactically correct but rhythmically wrong. This pass is slower than you think it will be and more pleasurable than the first two, and it carries a real risk: it can seduce you into line-editing scenes that are still structurally broken. Polish last. Always last.
The Feedback Loop: When to Ask and Who to Ask
Beta readers can save a novel. Beta readers can also damage one -- not through malice, but through timing. Feedback received too early, before the draft is structurally stable, activates the worst instinct in beginning writers: the tendency to revise in the direction of the last comment, regardless of whether that comment is right. A beta reader who reads a rough first draft is reading a different book than the one you are trying to write, and the notes they give you are notes on that draft, not on the novel you are moving toward. Early feedback creates uncertainty where you need conviction.
The right time for beta readers is after the structural pass. After you have fixed the large shapes and know what the book is doing. Then outside eyes can tell you what you cannot see -- where readers lose the thread, where a character's motivation is unclear to someone who has not read the notes and intentions inside your head, where the pacing sags in ways you have stopped noticing because you have read those pages forty times.
Who to ask: not your mother, not your best friend, not your partner, unless they are also serious readers who will tell you the truth and not protect your feelings. The most useful beta readers are writers who are roughly as experienced as you and who read widely in your genre. They can distinguish between "I personally would have written it differently" and "the scene is not working" -- a distinction that matters enormously, because the first note is noise and the second is signal.
The single most useful question to ask a beta reader: where did your attention drift? Not "what did you think?" Not "what should I fix?" Just: where did your attention go somewhere else? The honest answer to that question tells you more about structural and pacing problems than any other feedback you will receive.
The Polish: What Proofreading Actually Means
After the structural passes and the line-editing, there is the polish. Spelling, punctuation, grammar, consistency -- the character whose eyes are described as blue in chapter two and brown in chapter fourteen. It is the smallest work and the last work, and it matters: a manuscript full of easily-avoidable errors signals to agents and editors that the writer does not take the work seriously. But it is, importantly, the last thing you do, not the first.
Manuscript formatting is a separate subject from proofreading, with specific conventions for traditional submission: standard manuscript format, Times New Roman 12-point, double-spaced, header conventions for page numbers and your name and title. These are covered in the manuscript formatting guide, which is worth reading before you prepare any submission draft. The conventions are not arbitrary -- they exist to make your manuscript easy to read and compare with others, and following them costs nothing.
Three Paths: Submission, Self-Publishing, or the Drawer
When the novel is done -- actually done, not "done enough for now" -- you face a choice that the writing itself does not make for you.
Traditional publishing means querying literary agents. A query is a one-page letter describing your novel: hook, brief plot summary, comparable titles, a line about yourself. Agents who represent your genre receive hundreds of queries a week; the acceptance rate is very low, and the process from first query to signed contract averages eighteen months to three years, if it happens at all. After that, editorial work and production add another year or two before the book is in stores. The case for it: distribution infrastructure, the visibility that comes with a major publisher, editorial guidance, and the prestige that opens doors for subsequent books. Also: it costs nothing to try.
Self-publishing has changed dramatically since 2010. Amazon KDP and IngramSpark let you publish an e-book or print-on-demand paperback in days. You keep a larger royalty percentage per sale. You control everything: cover, price, timing, marketing. The case for it: speed, control, and the ability to reach readers in genres where self-publishing has built enormous audiences. The work: you are also the editor, the cover designer, the marketer, and the publicist -- or you hire all of those things, which costs money before you have made any.
The drawer is a third path that does not get talked about honestly enough. Some first novels are learning novels. They taught you everything about the craft that you needed to learn to write the next one. They deserve to rest in a folder on your desktop, finished, uncharged, intact. Toni Morrison did not publish everything she wrote. No novelist does. "If there's a book that you want to read, but it hasn't been written yet, then you must write it," she said -- and the complement of that is true too: not every book you had to write for yourself is a book that needs to reach the world. There is no shame in a novel that served its purpose as a teacher. The next one is waiting.
Starting the Next One: The Part Nobody Explains
Nobody tells you about the grief of finishing a novel. The characters have been in your head for a year or more. The world they inhabit has been, in some ways, more real to you than your daily life. Then you type the last sentence and it is over, and they leave. The silence that follows is specific and strange, and it takes a few weeks to pass.
Then the next idea arrives. It always does.
The knowledge you bring to the second novel is not the knowledge you wished you had had on the first one. It is something different and more useful: you know, in your body and not just in your head, what it is to start with uncertainty and finish with something real. You know how the 30-50% slump feels and that it ends. You know that the worst chapter in the first draft is not a sign that the book is failing. You know that revision is the work, not the afterthought. You know that the first draft is only the first draft. This is knowledge that cannot be described to anyone who has not experienced it.
Joan Didion, who wrote some of the most precise and honest sentences about the way the mind processes experience, said: "We tell ourselves stories in order to live." I think about this differently now than I did before I finished a novel. The story you tell yourself during the writing -- the one about who your characters are and what they want and what it will cost them -- is also a story about who you are and what you are willing to do and how long you will stay in the chair when the work is hard. Writing a novel is not just narrative craft. It is, among other things, a sustained exercise in not giving up.
Chapter eleven, rewritten seven times, became the best chapter in the book. That is not inspiration. That is the work -- and if you stay in it long enough, the work becomes something you trust.
The messy, sprawling, non-linear work of writing a novel -- three folders for the manuscript, workshop notes alongside each chapter, version history so nothing is ever lost, flowcharts for structure and planning -- is what Plotiar is built for. Free to start.