How Many Drafts Should a Story Have?
Quick answer: How many drafts a story needs: why three to five passes is the working norm, what each draft is for, and when to stop revising and submit your story.
Most stories need three to five drafts: a fast first draft to discover the story, a structural second draft to fix what the story is, and one to three focused passes for scenes, sentences and final polish. The number is less important than the sequence, because a 'draft' is really a question you put to the manuscript, and the questions must come in the right order: story first, scenes second, sentences last.
The question matters because writers fail at both ends. Some submit steaming first drafts and burn good markets on unready work; others sand a story through fifteen drafts until the voice is gone and the file names have exclamation marks. Knowing what each draft is for tells you both when to keep going and when to stop.
The longer answer: drafts are passes with different jobs
Counting drafts is fuzzy, one writer's 'second draft' is another's 'seventeen tweaks', so think in passes with distinct jobs. Pass one, the discovery draft: write to the end, forbid backtracking, accept mess; its only deliverable is a complete shape. Pass two, structure: read the whole thing cold and answer big questions only. Does it start in the right place? Is the middle escalation or repetition? Does the ending pay what the opening promised? This pass moves, merges and deletes scenes, and it is where the story most changes.
Pass three, scene and character: within the now-correct structure, make each scene earn its place, sharpen motivation, fix continuity, convert summary to drama where drama is owed. Pass four, the line edit: rhythm, word choice, cutting the ten percent of flab almost every draft carries. Pass five, the proof: typos, formatting, and reading aloud, which catches what the eye forgives.
The honest 'it depends': experienced writers compress passes because they draft cleaner and diagnose faster; some, like Dean Koontz, famously polish page by page and produce one glacial 'first' draft that is really all five passes interleaved. Method varies. The jobs never do.
What real writers' drafts look like
Hemingway rewrote the ending of A Farewell to Arms dozens of times, thirty-nine by his own count in one interview, and when asked what the problem was, answered that it was getting the words right. Raymond Carver spoke of doing twenty or thirty drafts of a story, and his surviving manuscripts confirm heavy, repeated cutting. At the other pole, Anthony Burgess and Katherine Anne Porter both described producing near-final pages in one pass, after long mental composition.
Two lessons hide in the anecdotes. First, published perfection is manufactured, not typed: the effortless-looking Carver story is the residue of a dozen ruthless passes. Second, draft counts are personality, not virtue. The single-pass writers front-load the work into planning and slow sentences; the many-pass writers back-load it into revision. Both pay the same total bill. What no successful writer does is skip the bill: nobody's first attempt, however talented, reliably survives a cold reading unimproved.
How to run your own revision passes
Between draft one and draft two, wait. A week minimum for a story, longer if the deadline allows, because the structural pass requires reading like a stranger, and strangers need distance. Read the whole story in one sitting, no pen for the first read, then write yourself a one-page memo: what this story is about, what works, what is broken. Revise from the memo, not from line-by-line dabbing.
Keep each pass on mission. When you spot a clumsy sentence during the structure pass, mark it and move on; polishing scenes that the next pass may delete is the classic way to spend a month achieving nothing. Save each draft as a new file so cutting feels reversible; writers cut braver when nothing is truly lost.
Know the stopping rule: stop when a pass produces only sideways changes, swaps rather than improvements, or when you catch yourself reversing edits from the previous pass. That oscillation is the signal that the story has reached your current ceiling, and the next improvement will come from readers, editors or judges, not another solo lap. Scores with a breakdown, of the kind Writory returns to every story contest entrant, are useful precisely at this point: they tell you which pass to run next on the next story.
Common mistakes
Polishing too early is the most expensive: beautiful sentences in scenes that pass two will delete. Structural work on lacquered prose also hurts more, so writers unconsciously protect bad scenes because the sentences are pretty. Draft rough on purpose; ugliness keeps you honest.
Second, revising without a diagnosis, opening the file and 'making it better' line by line, which changes texture and fixes nothing. Third, infinite drafting as procrastination with a work ethic: if you are on draft nine and cannot name what draft nine fixes, you are hiding from submission. Fourth, treating feedback rounds as extra drafts to obey wholesale; readers diagnose brilliantly and prescribe badly, so revise to the symptom they report, not the surgery they suggest. And fifth, deleting old drafts; the line you cut in March is sometimes the ending you need in June.
What poets can take from this
- Run passes in order: structure, scenes, sentences, proof; never backwards.
- Cool the story for at least a week before the structural pass.
- Write a one-page memo after the cold read and revise from it.
- Mark line problems during early passes; fix them only in the line pass.
- Save every draft as a new file so cuts feel reversible.
- Stop when edits turn sideways; the next gain comes from outside eyes.
Try it yourself
Take a finished story of yours and write the one-page memo a tough, fair editor would send about it: what it is about, what works, the two biggest problems. Then run one revision pass that addresses only problem one.
Questions
What is the difference between revising and editing?
Revision is re-seeing: changing what the story is, its structure, scenes, characters and ending. Editing is refining how it reads, at sentence and word level, plus correctness. Doing them in that order saves enormous wasted polish.
Can a first draft be good enough to submit?
Almost never, and not because first drafts lack talent: they lack the information that only a complete draft reveals about the story's true shape. Even fast professional writers give work a cold read and a cleanup pass. The cost of waiting a week is tiny; the cost of a burned market is not.
How do you know when a story is finished?
When successive passes produce trades rather than gains, when the story matches the effect you named for it, and when reading aloud snags nowhere. Finished does not mean perfect; it means this story has taught you all it can, and the improvement curve now runs through readers.
By the Writory editorial team, reviewed by working poets. Updated July 2026.
Enter a Writory poetry contest
Enter the August contest and have your own poem read and scored.