Editorial Analysis: How to Get an Editor's-Eye Report on Your Manuscript

There's a moment every writer reaches after the last chapter is done. You've read the book so many times that you can't see it any more. The sentences slide past. You know the twist is coming, so you can't tell whether it surprises. You know why the mentor leaves in chapter nine, so you can't tell whether a reader would.

That's the moment an editorial analysis is for. It's someone, or something, reading the book cold and telling you what's actually on the page, as opposed to what's in your head.

This article covers what an editorial analysis is, what a good one looks at, what it costs when a person does it, and how StoryWith AI runs one on your whole manuscript. (StoryWith AI is our product. We'll be clear about what it can't do as well as what it can.)

---

What an editorial analysis is, and what it isn't

An editorial analysis is a diagnosis of the book as a whole. It answers big questions. Does the structure hold? Do the setups pay off? Does the middle sag? Does each character change, and does each scene earn its place?

It isn't a proofread, and it isn't a rewrite. It's easiest to see where it sits by looking at the three levels of editing a book usually goes through, in order:

An editorial analysis lives at the top level. That order matters: there's no point polishing the sentences of a chapter you're about to cut.

Human editors often sell the top level two ways. A developmental edit comes with notes in the manuscript itself and a long editorial letter. A manuscript assessment is a reader's report on the whole book without the in-text notes. Both answer the same questions.

---

The eight things a good editorial analysis checks

Whoever does it, a useful editorial analysis covers the same ground. If a report skips one of these, ask why.

1. Structure and beats. Do the setups pay off? Do the turning points land where the story needs them?
2. Pacing. Where does tension rise, and where does it go flat for too long?
3. Character arcs. Where does change start, where does it stall, and does it ever arrive?
4. Scene purpose. Which scenes are nicely written but change nothing?
5. Continuity and plot holes. Names spelled two ways, ages that shift, timelines that contradict themselves, a twist with no setup.
6. Dialogue. Do the characters sound different from each other? Is exposition dressed up as conversation?
7. Readability. Is the prose in a sensible range for the genre, or are there walls of text that readers will skim?
8. Theme. What is the book about underneath the plot, and does that run all the way through?

There's a ninth thing, which software can't do and a good editor can: judgment. Is the voice working? Is the ending earned? Will this find its market? Keep that in mind; we'll come back to it.

---

What it costs when a person does it

Good human editors are worth paying for. They're also expensive, and they're usually booked weeks ahead.

Using the Editorial Freelancers Association's 2026 rate chart:

That's a lot of money to spend on a draft with continuity errors and a sagging middle you could have found yourself. The smart move is to fix what can be found mechanically first, so the editor spends their hours (and your money) on the questions only a person can answer.

That first pass is what StoryWith AI is built for. (For the full breakdown of human rates, see our guide to developmental editor cost.)

---

Running an editorial analysis in StoryWith AI, step by step

StoryWith AI reads your whole manuscript and ranks what to fix first, with every finding linked to the passage that caused it. It's a manuscript editor, not a line editor and not a text generator. It doesn't rewrite your book. Analysis gives you findings, not new text.

Here's the workflow from start to finish.

Step 1: Bring the manuscript in

Import a .docx, .md, .txt or .pdf file, or paste the text. StoryWith splits it into chapters and scenes, then pulls out the structure, the characters and the arc. (A scanned PDF without selectable text has to be saved as .docx or .md first.)

You don't need a finished book. The scans work on three chapters as well as thirty.

At the same time, StoryWith builds a Story Bible from what you wrote: characters, relationships, places, themes and a story timeline. That's the memory the analysis checks your manuscript against.

Step 2: Run the whole-book analysis

The analyzers read every chapter, so problems that only show up across the whole book become visible. Between them they cover the eight areas above:

• Structure and beats: setups that never pay off, and turns that don't land.
• Pacing and emotional arc: tension chapter by chapter, shown on a chart, so flat stretches are easy to see.
• Character arcs: where change starts, stalls or never arrives.
• Scene purpose: scenes that are well written but do no work.
• Continuity and plot holes: timeline contradictions, names spelled two ways, empty chapters, payoffs with no setup.
• Dialogue: voices that blur together, and exposition disguised as dialogue.
• Readability: prose outside the typical range for the genre.
• Theme: what the book is about, and where.

For non-fiction, the same pass adds argument mapping and soundness checks, flags unsupported claims, and supports fact-checking against your sources.

Step 3: Read one ranked list, not eight reports

This is where StoryWith differs most from a stack of separate reports. Every analyzer feeds into Diagnostics: one list, with duplicates removed, sorted by severity into Critical, High, Medium and Low. Continuity, dependencies, readability, dialogue, argument soundness and readiness blockers all appear in the same place.

Here's what that looks like on a sample manuscript:

Forty-seven items sounds like a lot, but they're in order. Fix the three critical ones first. Most authors never get to the bottom of the Low pile, and they don't need to.

Step 4: Work it like an editorial letter

A human editor's letter explains why something is a problem. StoryWith's Next Steps does the same. Each finding becomes a card that says why it matters and suggests a direction, without rewriting text you didn't ask it to touch.

From each card you can:

• open the scene it's about and fix it in your own words,
• mark it resolved (you can undo that while the confirmation is showing),
• pin it to keep it in view, or
• defer it, say for 7 days, if it's a job for the next draft.

And if you want help with a stubborn passage, you can ask for it. Select the text in the editor and choose Tighten, Rewrite, Describe, Analyze or Instruct. What comes back is a suggestion. Nothing changes in your manuscript until you accept it.

Step 5: Test the fixes on readers

An editorial analysis tells you what's on the page. Readers tell you how it lands. StoryWith keeps both in the same project.

• AI beta readers. Test a chapter on 3 to 5 simulated reader personas. Each rates engagement, clarity, emotional impact and genre fit, and shows where they'd stop reading. StoryWith is open that these are simulated and directional, a first pass rather than a prediction.
• Real readers. Share chapters with up to 20 beta readers through a private link. You can ask your own questions and see their reactions mapped chapter by chapter, and the draft is never sent out as a file.

Step 6: Rescore, then hand it to a human

The readiness score checks things like whether every chapter has scenes, whether the book is internally consistent, whether setups come before payoffs, and whether readability stays in range. Run it before you revise and again after. Because the findings list stays current as you revise, you can watch the score move.

When the critical and high items are gone, you have a draft that's ready for a human editor, and their time goes on judgment instead of on catching a character whose eyes changed colour.

---

What the analysis won't tell you

An honest editorial analysis tool should say where it stops. StoryWith AI says it plainly:

• It can't judge voice, market fit, whether an ending is earned, or whether a choice reads as deliberate.
• It doesn't know your reader or your market, and no software should pretend to.
• It doesn't rewrite your manuscript. Nothing goes into the text unless you choose to put it there.
• It's not a replacement for a human editor. It's the first pass that gets the draft ready for one.

That's the ninth item from the checklist earlier. Software handles the first eight well and fast. The ninth still belongs to a person, whether that's an editor, a trusted reader or you on a fresh read.

---

Where StoryWith fits among editorial analysis tools

Most AI manuscript tools fall into one of two camps:

• Report tools give you a snapshot of the whole book and leave you to act on it yourself. The report is out of date the moment you start revising.
• Scene-by-scene tools ask you to enter your story elements as you go before they can help.

StoryWith links each finding to its passage, keeps the list current as you revise, and keeps reader feedback in the same project as the analysis. The editorial letter, the manuscript and the readers stay in one place.

---

Your manuscript stays private

Your book is visible only to you and the collaborators and readers you invite. It's not used to train AI models and it's not sold. Feedback on your own book counts as AI-assisted, not AI-generated, under Amazon KDP's guidelines. Still, check KDP's current policies before you publish.

---

What it costs

You can start for free, with no card needed. The Free plan includes every tool and unlimited books, with a small one-time AI allowance (50 text actions, 15 images) and basic continuity and coverage scans.

To give you a sense of scale: a book-level analysis of a 90,000-word novel uses about 18 text actions. Running the scene-level analyzers on every scene adds about 375 more. Rescans draw on your monthly allowance, not on credits for each manuscript. Compare that with the $2,700–$3,150 a full human developmental edit costs. That's not a reason to skip the editor. It's a reason to send them a better draft.

Check storywith.ai/pricing for current prices.

---

Common questions

Can AI do an editorial analysis? It can do most of the first pass: structure, pacing, continuity, character arcs, scene purpose, dialogue and readability. It can't make the editorial judgment calls about voice, market or whether the book moves readers.

Will StoryWith rewrite my book? No. Analysis produces findings, not new text. AI help in the editor is only available when you ask for it, and nothing changes until you accept.

Do I need a finished draft? No. The scans work on three chapters as well as thirty, though whole-book problems show up best on a complete draft.

Does it work for non-fiction and memoir? Yes. Non-fiction adds argument mapping, soundness checks, fact-check support, citations in APA, MLA or Chicago, and a book proposal generator. There's also a memoir editor for arc, timeline and reader feedback.

Which file formats can I import? .docx, .md, .txt and .pdf (with selectable text).

Can I use it on my phone? Yes, in any modern mobile browser or with the Android app on Google Play.

---

The bottom line

An editorial analysis is the step between "I finished it" and "it's ready". A human editor will always bring judgment that software can't. But the structure, the pacing, the continuity and the forty-seven small things you stopped being able to see can be found in minutes, ranked by how much they matter, and fixed in your own words.

Run your first editorial analysis free at storywith.ai.