Author FAQ
RevisionGrade is built to separate manuscript readiness from the publishing black box. This FAQ explains what the system evaluates, what it does not support, how revision works, and where author control remains absolute.
Current scope
RevisionGrade evaluates manuscript evidence. It is not a generic document-scoring tool for letters, resumes, academic papers, legal documents, or marketing copy.
Evaluation basics
RevisionGrade evaluates full manuscripts, partial manuscripts, individual chapters, and serious narrative excerpts across thirteen story criteria: concept, narrative drive, character, voice, scene construction, dialogue, theme, worldbuilding, pacing, prose control, tone, narrative closure, and marketability. The goal is diagnosis, not generic encouragement or a single opaque score.
RevisionGrade evaluates manuscripts and serious narrative excerpts: novels, novellas, book-length memoirs, narrative nonfiction manuscripts, partial manuscripts, individual chapters, and serious fiction or nonfiction excerpts. Depending on length and complexity, submissions may receive short-form, long-form, or long-form multi-layer evaluation.
No. RevisionGrade may help authors create or prepare query letters, synopses, author biographies, and agent submission materials through Agent Readiness™, but those materials are not evaluated through RevisionGrade’s manuscript-evaluation engine.
RevisionGrade is not designed for general document critique, including personal letters, business letters, professional correspondence, employment cover letters, resumes or CVs, academic papers, research papers, legal documents, contracts, marketing copy, or sales materials. These document types require different standards than manuscript diagnosis.
RevisionGrade’s evaluation engine is built for literary and narrative analysis: story structure, character development, point of view, pacing, narrative cohesion, reader promise, and manuscript readiness. Letters, synopses, query letters, and author bios serve different purposes. RevisionGrade may help create or improve publishing materials, but it does not score them as manuscripts.
If RevisionGrade determines that a submission is not a manuscript or serious narrative excerpt, the evaluation will not proceed. The system will explain why the document type is unsupported and invite you to submit an eligible manuscript instead. Our goal is reliable manuscript diagnostics — not generic document scoring.
No. The score is useful only when paired with evidence, severity, confidence, reader effect, and revision priority. RevisionGrade is designed to explain why a manuscript is or is not ready, not merely label it good or bad.
No. RevisionGrade diagnoses manuscript readiness. It cannot guarantee agent interest, representation, publication, sales, market timing, or commercial demand.
Yes. A manuscript can be craft-ready and still be rejected because of list fit, market timing, category preference, agent bandwidth, comparable-title concerns, or simple subjectivity. RevisionGrade helps separate readiness problems from market-door problems.
Document eligibility
RevisionGrade currently evaluates novels, novellas, book-length memoirs, narrative nonfiction manuscripts, and serious fiction/nonfiction excerpts. Depending on length and complexity, eligible manuscripts route to short-form, long-form, or long-form multi-layer evaluation.
No. Letters, synopses, query letters, and author bios are not evaluated by the manuscript-evaluation engine. RevisionGrade may help prepare these through Agent Readiness Package™, but they are packaging materials, not manuscript diagnostics.
RevisionGrade is not designed for general document critique such as personal/business correspondence, employment cover letters, resumes/CVs, academic or research papers, legal documents, contracts, marketing copy, or sales materials.
If a submission is detected as non-manuscript content (for example a letter, synopsis, or bio), evaluation will not proceed. The system returns a clear explanation that the document type is unsupported and invites you to submit an eligible manuscript or narrative excerpt instead.
Evaluation modes
Short-form evaluation is for submissions under 25,000 words. It evaluates the submitted pages against the 13 story criteria only. It should not claim full-manuscript continuity, Golden Spine, or WAVE-level repair governance.
Long-form evaluation begins at 25,000+ words. At that length, RevisionGrade can evaluate manuscript-scale behavior such as continuity, recurrence, setup and payoff, pacing over distance, character development, structural readiness, and cumulative reader experience.
Long-form multi-layer evaluation is the deeper architecture path for complex manuscripts. It may use layered story evidence, long-form continuity, Golden Spine/WAVE governance where appropriate, and deeper structural repair priorities.
No. WAVE belongs to eligible long-form or multi-layer contexts. Short excerpts should not be marketed or interpreted as if they received full-manuscript repair-governance treatment.
Reports and downloads
A completed evaluation should give you a readable report with scores, evidence-backed findings, strengths, risks, criteria-level diagnosis, and revision priorities appropriate to the selected evaluation depth.
Report downloads are intended to be author-facing artifacts, not machine logs. The PDF should feel like the prestige document; DOCX and TXT exports remain useful utility formats for review, records, or later editing.
Confidence and severity help prevent overreach. A low-confidence observation should not be treated like a proven structural defect, and a small local issue should not be handled like a manuscript-level failure.
The dialogue/narrative ratio is a pacing and texture signal. It helps show whether the manuscript relies mostly on scene dialogue, mostly on narration, or a balanced mix. A very low dialogue percentage may suggest a dense, interior, descriptive, or summary-heavy reading experience; a very high dialogue percentage may suggest a fast, theatrical, or thinly narrated reading experience. The ratio is not a score by itself. It is measured so authors can understand how the manuscript may feel to readers and whether the balance supports the genre, voice, and story strategy.
Yes. A RevisionGrade report can help clarify whether the manuscript needs structure, scene repair, line polish, market positioning, voice protection, or another specific intervention before you hire or brief a human editor.
Revise and TrustedPath™
Revise turns evaluation findings into controlled repair opportunities. Each opportunity should show evidence, diagnosis, repair options, voice-risk considerations, and an explicit author decision path.
A is the recommended repair, the best default fix with the least unnecessary disruption. B is a rhythm or conservative variant. C is a bolder rendering shift that may be more interpretive while still remaining within the repair goal.
Yes. The author should be able to accept a proposed option, keep the original, reject all choices, defer the issue, or write a custom revision. Author control is part of the product doctrine.
TrustedPath™ is the governed convenience path for authors who do not want to review every repair opportunity manually. It should apply eligible recommended repairs to a protected duplicate draft, preserve the original, and produce a change log.
English Variant
RevisionGrade supports American English by default, plus British, Canadian, Australian, South African, and New Zealand English. The selected variant governs all RevisionGrade-generated output: reports, recommendations, revision guidance, Revise Queue content, TrustedPath rewrites, and other editorial text. Manuscript quotations, evidence excerpts, and author-written content are preserved exactly as submitted regardless of the selected variant.
The selected English variant becomes part of the evaluation record and is used throughout all associated outputs for that job. To generate output using a different English convention, create a new evaluation using the desired variant.
Privacy, security, and author control
You do. RevisionGrade treats submitted manuscripts as author-owned creative work. The system may diagnose, organize, and recommend, but it does not claim authorship.
No. The product posture is private by default and shared only by author action, such as downloading a report or preparing a controlled package surface.
RevisionGrade’s public trust posture is no: submitted manuscripts are author-owned creative work and are not positioned as model-training material. They are used to generate the requested evaluation, report, revision guidance, and author-controlled outputs.
External research should be bounded and purposeful. It may support genre context, public-domain comparison, comparables, market shelf, or factual plausibility, but it should not override the submitted manuscript evidence.
Agent Readiness and Storygate Studio™
Agent Readiness Package™ helps prepare publishing-facing submission materials such as query letters, synopses, author bios, comparables, and manuscript positioning. It does not evaluate those materials as manuscripts and does not guarantee agent interest.
Storygate Studio™ is a controlled manuscript-access layer for readiness-vetted book projects and verified publishing professionals. Current public scope is manuscript-first and publishing-facing.
Not in the current public scope. Storygate should stay focused on books, manuscripts, query packages, synopses, author bios, comparables, and controlled publishing-facing access until additional workflows exist.
No. Storygate is intended as controlled access, not open slush or public indexing. Creator approval, access control, and logging should govern visibility.
Next step
Read the methodology for the evaluation model, or begin an evaluation when you are ready to see where the manuscript stands.