Questions to Ask When Writing a Story
Craft questions to put to a story you are drafting or revising: whose it is, what the character wants, why it starts here, which scene it is built to reach, and where a reader is likely to stop. For short fiction and novels alike.
The questions
Open any question for the note
Whose story is this, and how would a reader know by the end of the first page?
Why ask it
Stories with two candidates for protagonist tend to feel unfocused rather than rich. If the answer changes depending on which chapter you look at, that is the structural problem, not a stylistic one.
What does your main character want, and what do they want more than that without admitting it?
Why ask it
The stated want drives the plot; the unadmitted one produces the moments readers remember. When a character only has the first, they behave consistently and read as flat.
What is in the way, and is it a person, a situation, or the character themselves?
Why ask it
Purely external obstacles turn the character into a problem solver and the story into a sequence of tasks. Purely internal ones stall the plot. Most working stories have one of each, pushing in different directions.
Why does the story start on this day and not a month earlier or later?
Why ask it
The answer should be that something has just become impossible to ignore. If the opening could sit anywhere in the character's year, you are probably starting with context the reader does not need yet.
What does your character actually lose if they fail, and would a reader care about that loss?
Why ask it
Stakes named in the abstract, like a city or a kingdom, rarely land. Readers care about the loss they can picture: a person, a place, a version of the self the character was hoping to become.
What is the worst defensible decision this character could make, and does the story let them make it?
Why ask it
Protagonists who always choose sensibly leave the author manufacturing external accidents. A bad decision the reader understands is the cheapest way to generate consequence that feels inevitable.
Who is telling this, and what can the teller not see?
Why ask it
A narrator's blind spot is what gives the reader something to do. If your point of view knows everything and hides nothing, most of the story's tension has to be carried by plot alone.
What is your character wrong about at the start, and when do they find out?
Why ask it
This is the spine of a change arc, and its position dictates pacing: too early and the rest is aftermath, too late and the ending feels rushed. If they are wrong about nothing, they will end the story as they began it.
Which single scene is the whole story built to arrive at?
Why ask it
Naming it tells you what every earlier scene has to set up and what can be cut. Writers who cannot name it usually have two stories overlapping, each with its own climax.
Which scenes end with the situation exactly as it started?
Why ask it
A scene should change a relationship, a plan, a piece of knowledge, or a balance of power. Scenes that only convey information are the ones readers skim, and they are almost always the ones a writer is fond of.
Where does the reader know more than the character, and where less?
Why ask it
That gap is where dread and surprise come from, and it is controllable. Stories that keep the reader exactly level with the protagonist for two hundred pages feel flat even when the events are dramatic.
What is the setting doing besides sitting there?
Why ask it
Useful settings limit what the character can do, or make a private thing public, or supply weather that costs something. If the story could happen in any room, the description is decoration and can be halved.
What does each secondary character want on their own account?
Why ask it
Supporting characters who exist to advise, oppose or comfort the protagonist read as furniture. Giving each one an errand of their own is what makes a scene feel like it might go sideways.
If you covered the dialogue tags, could you tell who is speaking?
Why ask it
This is the fastest test for voice, and most drafts fail it. Differences in sentence length, vocabulary, what a person avoids saying and how they change the subject do more work than dialect spelling.
What are you explaining that the reader could work out instead?
Why ask it
Explanations follow the beats a writer does not yet trust. Look for the sentence after a good line of dialogue that tells the reader how to take it, and cut it to see whether anything is lost.
Where is the first place a reader might put this down, and what is happening there?
Why ask it
Usually it is a stretch of backstory in the first twenty pages, or a middle section where the character is waiting for information. Finding it is the whole job of a first revision.
Does the ending answer the question the opening asked?
Why ask it
Drafts drift, and the ending frequently resolves a question that emerged in the middle while the opening promised something else. Readers experience that as an unsatisfying ending even when the last scene is strong.
What does the ending cost the character?
Why ask it
Endings where the protagonist gets what they wanted with nothing given up feel unearned even when the plot is watertight. The cost does not have to be tragic; it has to be real and named.
What is this story about underneath the plot, and does anything in the text argue with that?
Why ask it
The theme is what your choices already imply, not what you intend. If a subplot quietly contradicts it, either the subplot goes or you have found a more interesting story than the one you planned.
Which passage you are proudest of is doing no work?
Why ask it
Nearly every draft has one, and it is usually a set piece written before the story knew its shape. Cutting it, or moving it somewhere it earns its place, is the most reliable single improvement available in revision.
Using these on a draft
Practical guidance for the conversation itself
When you are still planning
Get want, obstacle and cost onto one page
Three lines: what the character wants, what stands in the way, what failing would take from them. If any line is hard to write, the story will be hard to write, and the difficulty will show up at the midpoint.
Find the scene you are writing towards
You do not need a full outline, but knowing the destination lets you plant what it needs. Writers who discover as they go still tend to have one image or confrontation they are steering at.
Start later than feels comfortable
Most openings improve when the first two or three pages come off. Whatever the reader needs from the material you cut can be delivered in half a line once the story is moving.
Scene-level checks
- Name what changes between the first and last line of the scene. If nothing does, either give it a change or fold it into a neighbor.
- Check who wants something in the scene besides the protagonist, and what they would settle for.
- Ask what each character believes at the start of the scene that is untrue.
- Look at the last line: cutting the final explanatory sentence usually strengthens the exit.
- Count how long the reader has been in one character's head, and whether the shift is deliberate.
- Read the dialogue aloud. Anything you stumble over is a rhythm problem, not a typo.
Revising without losing your nerve
Read it once without a pen
Mark only where your attention drifted, with no notes and no fixes. Those marks are the most honest data you will get about your own draft, and they disappear once you start editing lines.
Fix structure before sentences
Polishing prose in a chapter that will be cut is the most common waste of revision time. Move and remove whole scenes first, then rewrite at the paragraph level, then at the line.
Believe where readers stumble, not why
Test readers reliably report the moment they lost interest and unreliably diagnose the cause. When two of them stall in the same place, treat the place as fact and ignore the prescriptions.
Keep the cut material
A separate file for deleted scenes makes bold cuts psychologically cheap, which is the only way most writers manage them. Most of it never returns, and one piece of it usually solves a later problem.