A trade-pedigreed developmental letter reads nothing like a copyedit query list, and neither one reads like a marked-up proof. Different artifacts, different crafts, different skills. Pricing them against one another is how authors end up paying line-edit money for a polish and copyedit money for structural notes that won’t survive revision.

1. Developmental edit — the conversation about the book’s bones. A structural read. Does the story work? Do the characters carry the arcs they’re given? Does suspension of disbelief hold under the manuscript’s own logic? The editor is naming the questions the book is asking and isn’t yet answering. She is not fixing sentences.

2. Line edit — the conversation about how the prose performs. Sentence by sentence. Rhythm, voice consistency, word choice, the three sentences in a row that start the same way, the verbal tic the author stopped noticing at draft four. Not a grammar pass. A craft pass where polished gets sharpened into memorable.

3. Copyedit — the conversation about usage, consistency, and the style sheet that survives everything. Chicago Manual, a preferred dictionary, a numbered query list, and — this is the artifact the uninitiated don’t expect — a ten-to-forty-page style sheet documenting every name, place, timeline fact, and house-style choice the book contains. The document the proofreader and typesetter will ask for by name.

4. Proofread — the conversation on the typeset pages. Final pass. On the laid-out interior, not on the manuscript. The proofreader is looking for what the copyeditor missed, what the typesetter introduced, widows and orphans, bad hyphenation breaks, running heads that lie. Production-stage, not writing-stage.

Four passes. Four deliverable sets. When those get collapsed, what disappears is the artifact that makes the next pass possible.

Pass 1 — the developmental edit

Developmental is the pass authors most often skip, because the price tag lands in the first email and the gut reaction is to hope the book is already in better shape than it is. A working consensus across editorial veterans: for a literary debut, this is the pass with the highest return on every dollar, and the one where a poorly chosen editor can spend three months’ budget leaving the book no better than she found it.

What a developmental edit produces. A good engagement delivers four artifacts, and the absence of any is a red flag:

Rate ranges typically paid and quoted. The EFA 2024 guide lists developmental at $0.03–$0.10 per word, or $2,700–$9,000 on a 90,000-word novel. Real-world engagements and sample-edit quotes land squarely inside that band: packages for a trade-length manuscript quoted from $3,200 at the lower-middle end (trade-crossover freelancer, letter plus call) to $8,500 at the upper end (ex-imprint editor, letter plus full manuscript markup plus two calls). A standalone manuscript evaluation — the shorter-form cousin that some Reedsy profiles offer at $500–$1,500 — priced as a standalone deliverable without a full developmental letter is generally not worth the money for a book the author is serious about; what you get back is a reading report, not a structural plan.

When to commission it. After a complete draft and at least one round of self-revision. Developmental on a first draft is too early — the book doesn’t yet know what it is, and the notes won’t survive the version it becomes. Jane Friedman’s guide on timing a developmental edit lays the case out clearly.

How to hire. A working approach starts with a paid sample of 10–20 pages, priced at $150–$400. What to read for: engagement with what the book is trying to do, not a polished performance of what’s on the page. The signal that gets an editor hired is a tracked-changes sample plus a short written memo in which the editor names a structural question she can’t yet answer and asks for authorial intent before she guesses. The signal that gets an editor passed over is a polished sample with margin notes that describe what’s on the page without engaging with it — the under-insightful sample, which is the most common failure mode at this tier. In a typical developmental search, the majority of paid samples land in that second category; the hire goes to the one editor who diagnoses a structural question the letter will later anchor.

Pass 2 — the line edit

Line editing is where the prose stops sounding competent and starts sounding inevitable. The editor reads sentence by sentence for rhythm, voice, compression, and the places where narration slips into authorial register. Not grammar, not consistency. The work is subtle enough that it’s the hardest pass to evaluate from a résumé, and the one where a sample edit is most load-bearing.

What a line edit produces.

Rate ranges. The EFA guide puts line editing at $0.04–$0.08 per word: $3,600–$7,200 on a 90,000-word novel. Trade-pedigreed line editors — the ones who come up in acknowledgments pages of books you admire — price at the top of that range and occasionally above. A typical engagement with an editor whose sample is so sentence-level specific the margins reward a second read prices around $4,900 all-in, manuscript plus follow-up memo. A different pass from a different editor, whose sample reads as an overwrite of the author’s voice, gets declined at the sample stage — a useful $300 spent against a $5,000 decision.

When to commission it. After developmental revisions, once the structure holds. Line-editing a manuscript with an unresolved structural problem is, as a friend in the trade puts it, polishing furniture on a sinking ship — and you’ll pay for a second pass after the fix, because the sentences that got sharpened live inside scenes the revision will displace.

How to hire. The sample protocol: 2,000 words, paid at $100–$250. The editor’s fingerprints should feel like an improvement, not an overwrite. Read the before and after out loud. If the edited version sounds like the book, better — hire. If it sounds like the editor’s book — pass. Different line editors work at different scales (rhythm, compression, voice), and the only way to tell which scale a given editor works at is to see fifteen minutes of her actual pen on actual pages. A typical sample archive ranges from the voicefully-aligned (an editor who tightens a paragraph by six words without touching character voice) to the voicefully-obliterated (an editor who flattens dialogue rhythm in the name of “cleaning up”). The sample is the interview.

Pass 3 — the copyedit

Copyediting is the pass the world means when it says edited. It’s also the pass authors routinely underestimate, because it looks cheapest on the rate card and therefore feels most affordable — which is not how editorial value actually works. A strong copyedit produces the single document every subsequent pass on the book relies on: the style sheet.

What a copyedit produces. Three artifacts, all arriving together:

Rate ranges. The EFA guide puts copyediting at $0.02–$0.04/word: $1,800–$3,600 on a 90,000-word novel. Copyeditors with a heavy-fiction list price at or above the top of that range; generalists price below it. The difference is rarely the number of errors caught — it’s the quality of the style sheet that arrives when the pass closes.

When to commission it. After the line edit. Copyediting first and then line-editing undoes most of the copyedit. Line-editing first and then copyediting is a short hop — the style sheet survives it.

How to hire. Look for a copyeditor who names her style manual and preferred dictionary in the first reply — almost always The Chicago Manual of Style for book-length fiction, and either Merriam-Webster unabridged for American English or New Oxford for UK. A copyeditor who works “by feel” isn’t a copyeditor. Then ask for a redacted style sheet from a past fiction project. The quality of someone’s style sheet is the tightest signal available about how carefully they work — more predictive than years of experience, more predictive than imprint résumés. A strong recent engagement produces a style sheet past 20 pages; a weaker engagement from a different editor produces a four-page document that is really a glossary. One is a working instrument; the other an afterthought. The sample style sheet is the interview.

Pass 4 — the proofread

Proofreading is the final pass and the one most commonly misunderstood. It happens on the typeset pages, not on the manuscript. The actual laid-out interior, in the same PDF a printer would send to press. A proofreader is looking for what the copyeditor missed, what the typesetter introduced, bad line breaks, widows and orphans, bad hyphenation decisions, a chapter running head that says Chapter 14 on page 217. This is a production-stage pass, not a writing-stage pass, and it is not the same thing as a copyedit.

What a proofread produces.

Rate ranges. The EFA guide puts proofreading at $0.01–$0.03/word, or often billed by page at $3.50–$5.50/page. A 90,000-word novel typesets to roughly 320–400 pages depending on trim size and interior design. On a page-count basis, expect $1,100–$2,200. If the interior is produced by a turnkey service like 1106 Design, proofreading is sometimes bundled into the layout package — the word proofreading gets used loosely in those contracts, so confirm in writing what the deliverable actually is. A real proofread happens on the typeset pages.

When to commission it. After typesetting. Never before. A proofreader working on an un-typeset Word file is doing a duplicative copyedit at a proofreading rate, which is a waste of everyone’s money.

How to hire. Hire a different human from the copyeditor. Two different sets of eyes across the copyedit and proofread passes has a meaningfully lower miss rate than the same person doing both — which is why trade publishers use separate copyeditors and proofreaders, and it’s worth the modest cost increase on any project meant to last. The EFA directory filters by specialty, including proofreading specifically; the Chartered Institute of Editing and Proofreading maintains a UK directory; Editors Canada runs the Canadian equivalent. A working practice: hand the proofreader the copyeditor’s style sheet, the tracked-changes source file, and the typeset PDF. She works from all three. The style sheet is what makes this pass efficient and what makes it catch errors a proofreader without it would miss.

Why the four happen in that order

The sequencing matters because each pass produces artifacts the next pass depends on. Developmental revision is the pre-condition for a line edit; there’s no point sharpening sentences that sit inside a scene the structural revision will cut. The line edit is the pre-condition for the copyedit; copyediting prose that’s about to be rewritten for rhythm means paying twice. The copyedit is the pre-condition for the proofread; the style sheet the copyedit produces is what the proofreader checks against. Running the passes out of order doesn’t just waste money — it actively destroys work product, because each subsequent pass’s artifacts (revised scenes, sharpened paragraphs, the style sheet) only survive if the passes before them closed cleanly.

Expect the full sequence to take eight to twelve months of real time on a trade-length manuscript, including the author revision windows between passes. Developmental engagement: 6–10 weeks of editor time plus a revision window of 2–4 months. Line edit: 3–5 weeks of editor time plus a revision window of 3–6 weeks. Copyedit: 3–5 weeks of editor time plus author review of 2–3 weeks. Typesetting: 2–4 weeks depending on interior complexity. Proofread: 2–3 weeks plus a correction cycle of 1–2 weeks. The compression temptation is real. So is the bill for what compression produces.

The full editorial landscape, named

Not every author needs to start from scratch building a shortlist. The working landscape of where author-facing editorial services live:

The marketplace’s low end is worth naming for what it is. Fiverr and similar marketplaces list “editing” at rates a fraction of the EFA floor; the work product is almost always proofreading-grade at best, frequently grammar-check-grade, sold as though it were developmental or line-edit work. FirstEditing and Scribendi run large-scale editing-service operations — capable of competent copyedits, rarely the right shop for a literary novel’s developmental or line-edit pass.

What a credible sample-edit looks like

The sample is where the hiring decision actually gets made. Across every pass, the working archive of sample edits clusters into three categories, and the patterns repeat.

The credible sample. Specific engagement with what the pages are trying to do. Tracked changes that read as improvements the author would have chosen if she’d seen them. Margin notes that explain the why of the non-obvious moves. At the developmental stage, a written memo naming one or two structural questions the editor can’t yet answer and asking for authorial intent. At the line-edit stage, a short note at the end of the sample calling out pattern-level observations. At the copyedit stage, a preliminary style sheet — even a one-page version — naming the choices she’d formalize across the full manuscript. At the proofread stage, a marked-up 10–20-page sample on typeset pages with layout flags alongside textual ones.

The polished-but-empty sample. Clean-looking markup, professional-seeming margins, but the edits are surface-level — commas, tightening obvious redundancies, swapping a word for a synonym. The editor is performing care rather than engaging with the manuscript. The tell: she hasn’t named a single structural or craft question. She’s giving you what she thinks you’ll accept without pushback, which is exactly the wrong posture for an editor you want to hire.

The overwrite sample. Edits that replace the author’s voice with the editor’s. Flattened dialogue rhythm. Sentences restructured to fit the editor’s preferred cadence. Margin notes that read as corrections rather than suggestions. This is the sample that, if accepted, produces a manuscript where the prose no longer sounds like the author. The tell: read the before and after out loud; if the after doesn’t sound like the book, it isn’t the right edit.

Across a dozen-plus sample edits in a typical editorial archive, the proportions track a rough pattern: a minority of samples are credible, a plurality are polished-but-empty, and the rest are overwrites. The polished-but-empty sample is the most common failure mode and the hardest to catch from a cursory read, which is exactly why the sample gets evaluated slowly, read aloud, and compared against the original across at least an hour of focused attention.

Common hiring traps

A few patterns recur across editorial searches, worth naming because the failure modes generalize:

The one-editor-does-everything temptation. Hiring one editor at a blended rate to do “a full edit” usually means three of the four passes didn’t happen. The editor — who knows she can’t do four passes well for the price of one — chooses which one she actually performs: a line edit at a developmental rate, or a copyedit at a blended rate. Either way, the structural problems or prose problems or style-sheet artifacts are still waiting.

The sample-to-contract gap. An editor’s sample reads one way; her actual engagement product reads another. The sample is polished because the editor spent two hours on fifteen pages; the full pass gets the editor’s 20-minutes-per-chapter pace, and the craft ceiling drops. Two mitigations: ask the editor how long the sample took, and write a clause into the contract that says the first 10% of the pass will be reviewed before the remainder is paid. A credible editor accepts that language without friction.

The flat-fee misunderstanding. Unlimited revisions or includes two passes language rarely survives a real engagement. “Two passes” is a marketing term unless the contract specifies the deliverable of each.

The reference check that doesn’t happen. Editors whose references dodge are editors whose reputations are unverifiable. A working practice: email two named past clients before wiring any deposit. Ask two questions: what surprised you about the engagement? and what would you do differently if you hired her again? Ten minutes of phone time is the cheapest due diligence available. Across editorial searches, the reference conversation has reshaped the hiring decision more than once.

Budgets and the decision matrix

A 90,000-word literary novel hitting the full editorial sequence at a competent mid-to-upper mid-market floor runs, in practice, roughly:

Adjust upward for an ex-imprint developmental editor, a heavy-fiction line editor whose acknowledgments pages read like a literary syllabus, or an interior-complex book that drives the proofread page count. Adjust downward for shorter manuscripts and for commercial-fiction categories where the genre’s conventions are doing structural work the developmental pass would otherwise address. Book-length nonfiction operates on different margins; this is the literary-fiction bookmark.

The passes most worth paying a premium for, on a budget-constrained search: developmental (the structural decision shapes everything downstream and the editor’s taste is doing load-bearing work), and copyedit (the style sheet is an asset the book will live inside for the next several production stages). The passes where mid-market competence closes the gap: line edit (with a credible sample test to confirm voice alignment), and proofread (where the craft is close to commodity across the competent tier, so a careful hire at the lower-middle of the range is usually the right call).

The pass to skip, if any gets skipped: none of them. If the budget doesn’t cover four, skip forward in time, not across the sequence — bank the budget, finish the developmental revisions, come back for the line edit in six months, copyedit and proofread when the manuscript is ready. The sequence will outwait the calendar. What it won’t survive is being collapsed.

Four passes, four conversations with the book, four different craftspeople engaged at different moments for different reasons. Spend what the sequence costs, in the right order, and the manuscript arrives at the end more itself than when it started — which is the whole point. You have the patience this requires, and the book will thank you for it.

e.

Cross-links