The Scene Checklist
What every scene needs to do — and the questions to ask when one isn't working.
A scene is the basic unit of story. Not a chapter, not a section — a scene: a continuous piece of action occurring in a single time and place, involving at least one character who wants something and encounters some form of resistance. Everything in a story is built from scenes, the way everything in a building is built from individual components. A story with weak scenes is a weak story, regardless of how good the overall concept is.
The questions below are the ones worth asking about every scene you write — both before you write it, to know what you're building, and after, to diagnose what isn't working. Not every scene will answer all of them in the same way. But a scene that can't answer any of them probably shouldn't be in the story.
The essentials — every scene needs these
What does the point-of-view character want in this scene?
Not in the story overall — in this specific scene, right now. A scene goal can be small: to get information, to leave a room, to make someone believe something. But the character must want something, or there is no engine. A scene without a want is a summary pretending to be drama.
If you can't answer this in one sentence, the scene doesn't have a center yet.
What is the obstacle?
Something must push back against what the character wants. The obstacle can be another person, a physical barrier, the character's own fear or false belief, or simple circumstance. Without resistance, want becomes wish, and the scene becomes inert. The quality of the obstacle determines the quality of the scene.
The obstacle should make the want harder to achieve, not just slower.
What changes by the end of the scene?
Something must be different at the end of the scene from how it was at the beginning — a situation, a relationship, a piece of knowledge, a character's internal state. A scene that ends in the same place it began has no reason to exist. The change doesn't need to be large, but it must be real.
The change can be positive (the character gets what they want) or negative (they don't) or a complication (they get something they didn't expect). All three are valid. Stasis is not.
What does this scene do for the story?
A scene should accomplish at least one — ideally two or three — of the following: advance the plot, develop character, establish or deepen theme, provide necessary information, create or release tension. A scene that only does one of these things is less efficient than one that does three. If a scene does none of them, it should be cut.
Ask: if I removed this scene entirely, what would the reader miss? If the answer is "nothing important," the scene is decorative.
Craft questions — the difference between a scene that works and one that sings
Where does the scene begin?
Most scenes begin too early. The reader doesn't need to watch the characters arrive, sit down, order coffee, and exchange pleasantries before anything happens. Enter the scene as late as possible — at the moment something is already in motion. Everything before that is setup the reader will read through impatiently.
Find where the scene actually starts. That's usually several paragraphs in from where you began writing it.
Where does the scene end?
Leave the scene as early as you can after the essential thing has happened. The last line of a scene is its most important line — it's what the reader carries into whatever comes next. A strong scene ending either creates a new question, lands an emotional note, or reframes what just occurred. A weak scene ending ties everything off too neatly and kills momentum.
The scene should end when the reader wants to know what happens next — not after you've answered it.
What does the POV character notice, and why?
A character in a specific emotional state notices specific things. What a frightened person sees is different from what an angry person sees, even in the same room. The details a scene includes reveal who the character is and what they're feeling — without requiring the author to announce it. If the details feel generic, the character isn't present in the scene yet.
Replace neutral description with perception. Not "the room was cluttered" but what about the clutter this person specifically notices, and what it means to them.
Is there subtext in the dialogue?
People rarely say exactly what they mean, especially when something important is at stake. Dialogue that states everything directly is both unrealistic and dramatically weak. Subtext — the gap between what characters say and what they mean — is where tension lives. The most powerful dialogue scenes are about one thing on the surface and something else entirely underneath.
If the dialogue in a scene could be summarized as "Character A wanted X, Character B said yes/no," the subtext probably isn't there yet.
Diagnostic questions — for scenes that feel broken
Is the scene summarizing instead of dramatizing?
Summarizing tells the reader what happened. Dramatizing shows them. "They argued about money for an hour" is summary. The actual argument, with its specific words and silences and gestures, is drama. Scenes that summarize instead of dramatize are usually placeholders — they know what needs to happen but haven't yet found the specific moment to show it.
Is the conflict externalized?
Internal conflict — a character thinking about how conflicted they feel — is hard to sustain as scene-level drama. The internal conflict needs to manifest in action, dialogue, or physical behavior. The character who is terrified to tell the truth should not spend three pages telling the reader they are terrified. They should do something that shows it while they avoid saying it.
Does the scene feel inevitable in retrospect?
A well-constructed scene, once you've read it, feels like it could only have gone the way it did — even if it surprised you while you were reading it. If the ending of the scene feels arbitrary, it's usually because the setup didn't contain the seeds of that outcome. Work backward: what would need to be true at the start for this ending to feel earned?
These questions synthesize ideas from a range of craft books — among them John Gardner's The Art of Fiction, Ursula K. Le Guin's Steering the Craft, and Robert McKee's Story. The framings and emphases here are my own. A scene that fails to answer most of these questions isn't necessarily bad — it may be deliberately impressionistic, or structurally experimental. But knowing the rules is the prerequisite for breaking them usefully.