Screenplay Notes: How to Read, Sort, and Use Them

Published by StoryNotes, which makes an AI coverage tool, one of the five sources of notes covered below.

You finish a draft, send it to four people, and get back four sets of notes that don't agree. One reader loved the opening. Another says it takes too long to get going. A contest reader writes "the second act sags" and stops there. A friend suggests the protagonist should be a cop.

Almost everything written about screenplay notes is about manners: say what works first, don't argue in the room, thank the reader. That advice is fine, and it doesn't help with the actual problem on your desk, which is deciding what the notes mean and which ones deserve a rewrite.

This guide treats notes as information to be decoded and sorted. Each section goes deep enough to act on today, then points to a longer article if you need more.

TL;DR: A note has three layers. Trust the reaction, test the diagnosis, and mostly set the prescription aside. Most vague notes trace back to a short list of causes you can check on the page yourself. Act on notes a second reader could verify, and log the rest until three readers agree. Then fix in order of blast radius: story first, scenes second, lines last.

Screenplay notes are feedback on a draft, written to help the writer revise it. They range from a peer's margin comments to a manager's email to a producer's formal memo, and they can be about anything from the premise to a single line.

Notes are not the same thing as coverage, although the words get used loosely. Coverage is a structured document, usually a synopsis, comments, and a verdict or grades. It began largely as a way for an executive to decide about a script without reading it. Notes are addressed to the writer and are about the next draft. Coverage sold directly to writers sits in between: it is formatted like coverage, but its job is the same as notes.

The distinction matters because you read the two differently. A verdict tells you whether one reader would pass the script along. A note tells you where a reader got lost, bored, or unconvinced. Only the second kind tells you what to rewrite. If you have a full coverage report in front of you, How to Read a Coverage Report walks through a real one and shows you how to turn it into a rewrite list.

Each source is good at a different question. None is good at all of them.

• Peer writers notice craft mechanics such as scene construction, setups and payoffs, and dialogue rhythm, because they solve the same problems on their own pages.
• Beta readers who don't write tell you what an audience experiences: where they got bored, confused, or moved. They usually can't tell you why, and that's fine.
• Paid readers and contest readers read a high volume of scripts and know what reads as competent against the rest of the pile.
• Managers and producers read for whether the script can be sold or made. Their notes carry market and budget considerations alongside story ones.
• AI coverage reads the whole script with the same attention on page 95 as on page 5, returns notes in minutes, and in the better tools cites the page behind each note. It doesn't know how a room full of people will react.

Credentials matter less than writers assume. In a randomized study of student writers, those who got feedback from several peers improved their next draft more than those who got feedback from one expert (Cho & Schunn, 2007). The follow-up found that the multiple-peer group received more feedback, including more feedback that described problems rather than prescribing fixes (Cho & MacArthur, 2010). That research is about academic writing, not screenplays, and Screenplay Beta Readers vs Peer Swaps vs AI Coverage discusses how far it transfers. The practical lesson holds either way: three readers with different vantage points beat one reader with a better résumé.

Every note has three layers, and they deserve different amounts of trust.

1. The reaction is what the reader felt: bored, lost, unconvinced, ahead of the story.
2. The diagnosis is why the reader thinks it happened: "the stakes are too low."
3. The prescription is what the reader would do about it: "make the brother a cop."

The reaction is almost always real. The reader did lose interest around page 50, and arguing with that is like arguing with a thermometer. The diagnosis is sometimes right, since readers are good at noticing problems and less consistent at explaining them. The prescription is usually a description of the movie the reader would have written.

Neil Gaiman made a version of this point in his rules for writing fiction: when people say something doesn't work for them they are "almost always right," and when they say exactly what's wrong and how to fix it they are almost always wrong. He groups the diagnosis with the prescription. In practice the diagnosis is worth testing, because a reader who names a cause is sometimes right about it. The research on revision points the same way. In the Cho and MacArthur study, the feedback that predicted the deepest revisions was the kind that described a problem without prescribing a fix.

Here is a note pulled apart. A reader writes: "Act two is slow. Cut the subplot with the sister." The reaction is that act two felt slow. The diagnosis, which is implied, is that the sister subplot is the cause. The prescription is to cut it. The sister subplot might be the cause. It's just as likely that the main plot stops escalating at page 45 and the subplot is simply where the reader noticed it.

The fix: before you decide anything, rewrite every note you receive as "Around page X, the reader felt Y." Keep the reader's diagnosis and prescription as hypotheses underneath. You are now working from evidence instead of instructions.

Vague screenplay notes are not random. The same dozen phrasings come up again and again, and each one usually points at a small number of causes you can check without a second opinion. This table is a starting point for that check, not a verdict.

The rows fall into four families. Three of them have a full article.

"I didn't know what this wanted to be." This note reads like a request to pick one genre, and it almost never is. Plenty of great scripts mix genres. The note usually means the opening made a promise the script stopped keeping. Screenplay Genre and Tone: Why Yours Reads as Confused works the problem on two public sample reports, including the difference between breaking a convention on purpose and drifting away from it by accident.

"What's it about?" and "It felt preachy." These look like opposite complaints, too little meaning and too much, and they share a cause. The script hasn't turned its meaning into a question the events have to answer. Theme in Screenplay: Why Yours Feels Absent or Preachy covers the single revision move that fixes both.

"Flat," "on the nose," and "everyone sounds the same." Dialogue draws a lot of notes, and fewer of them are usable, because everyone feels qualified to judge how people talk. Screenplay Dialogue Feedback: Which Notes to Act On gives four tests you can run alone, and each returns a pass or a fail.

Character and structure notes don't have their own article yet, so here is the short version. "I didn't care about the lead" is rarely about likability. Readers will follow a terrible person who wants something specific and pays for pursuing it. If you can't state the want in one sentence that could fail, that's the note. "The villain is thin" usually means the antagonist has no life of their own: they want nothing except to stop the hero. Give them a goal that would make sense if the hero never existed.

On structure, "it drags" and "the ending didn't land" are two of the most common structure notes, and they're usually symptoms of the same two patterns. One is an act two that is a sequence of events instead of an escalation. The other is a payoff with no setup, or a setup with no payoff. Why Your Script Keeps Getting Passes covers both, alongside three other patterns behind most passes.

Act on a note when a second reader could check it on the page without sharing the first reader's taste.

"Around page 50 I stopped caring" reports a feeling. It's real, and on its own you can't do anything with it. "Between pages 45 and 60 nothing makes the protagonist's situation worse" names a mechanism. Anyone can open the script and agree or disagree. Mechanism notes transfer between readers. Reaction notes don't.

Reaction notes still count, because they're the evidence, not the conclusion. Log every one against its page number and wait. When three readers react to the same stretch of pages, even in different words, the pattern is the note. It doesn't matter that one said "slow," one said "confusing," and one said "I lost track of what she wanted." Three readers stumbling at the same place means something is wrong there, and now you go and find the mechanism yourself.

Three situations come up constantly:

• Two readers contradict each other. Look at the page they disagree about rather than picking a side. Opposite notes on the same scene usually mean the scene is ambiguous about what it's for. Clarify its job, and both notes often go away.
• One reader, one strong note. Log it. If it names a mechanism you can see on the page, fix it. If it's a reaction nobody else shares, keep it on the list and watch for it in the next round.
• A note that's right but isn't your script. Some notes would improve a different movie. You're allowed to decline them. For every note you decline, write one sentence explaining why. If you can't write the sentence, give the note another look. Step five of How to Read a Coverage Report goes further on deciding what you're not going to fix.

Fix screenplay notes in order of blast radius, which means starting with the notes whose fixes change the most pages.

Notes on a script work like a dependency chain. A note about what the protagonist wants can change every scene she's in. A note about why act two drags can cut or merge whole sequences. A dialogue note changes a line. If you polish the dialogue first and then restructure act two, you've polished lines in scenes that no longer exist.

Sort your notes into three tiers:

1. Story notes change what happens or who wants what: the protagonist's goal, the antagonist's role, the shape of the middle, what the ending resolves.
2. Scene notes change how a scene plays: where it starts and ends, what each character is pursuing, whether it raises the pressure.
3. Line notes change how a moment reads: dialogue, description, formatting, typos.

Work top to bottom, and finish each tier before starting the next. That's the hard part, because line notes are the easiest to address and the most satisfying to tick off. A draft where every line note has been fixed and no story note has been touched reads better than the last one, and it fails for the same reasons.

The fix: tag every note on your list Story, Scene, or Line before you open the script. Don't touch a Line note until every Story note is resolved or deliberately declined.

When you're the one giving screenplay notes in a peer swap, the three layers tell you what to give.

Ask the writer what they're worried about before you read. They usually know where the soft spots are, and a reader who confirms or eases a specific worry is more useful than one who starts from scratch.

While you read, mark your reactions with page numbers. "I got ahead of the story here, p. 34" is the most useful sentence a reader can write, because the writer can go to page 34 and see it. Offer a diagnosis if you have one, and label it as a guess: "I think it might be because we already know she'll say yes." Hold prescriptions back unless the writer asks for them. The Cho and MacArthur finding applies here too. Describing the problem led to deeper revisions than prescribing the fix.

Finally, say what's working, and be specific about that as well. The point isn't to soften the criticism. Writers in revision cut good material by accident, and a note that says "the diner scene on page 12 is doing the most work in the script" protects it.

Show the new draft to the same kinds of readers and check whether the note is gone.

That's the only real test. A rewrite that addresses a note in your own head can still produce the same reaction in a reader's. If two of three readers stumbled at page 50 last round and nobody does this round, the fix worked. If they now stumble at page 60, you moved the problem rather than solving it, which is still information.

Reruns are where the cost of each source starts to matter. A peer swap costs you a read of their script. A paid read has to be booked and waited on each round. AI coverage comes back in minutes, so it's practical to rerun after every draft. A note that stops appearing across reruns, and that your human readers also stop raising, is a strong signal that the page changed rather than the reader's mood. Our coverage workflow guide lays out one way to sequence these sources across a revision. Our sample report shows what a page-cited AI note looks like, so you can judge whether that format suits how you work. If you want the wider picture first, The Honest Guide to AI Script Coverage covers what it does well, and AI vs Human Script Coverage covers when each kind of reader earns its place. Pricing for StoryNotes is on the homepage.

Whatever mix of readers you use, the method for screenplay notes stays the same: decode each note, sort it, fix big before small, and check the result with a fresh read.