About & Methodology

Outline Analyzer was soft-launched in March 2026 and officially launched in October 2026. It is owned and operated by VHS Video Services, LLC, a U.S. limited liability company.

What Outline Analyzer is for

Outline Analyzer is a developmental tool for writers working before the draft. It reads an outline, screenplay, episode brief, stage-play brief, short-fiction plan, or commercial treatment and identifies structural opportunities: where pressure weakens, turns arrive too late, character choices do not yet drive the plot, or an intended payoff lacks sufficient setup.

It is not a verdict on whether a story is good, marketable, or worth writing. The goal is to give writers a more concrete starting point for revision than generic encouragement or a page of disconnected notes.

How scoring works

Each Complete Analysis evaluates the supplied material through a consistent developmental rubric. The assessment considers the declared format, primary genre, and any genre qualifiers you provide, then reviews the story as a whole and at the chapter, scene, or beat level where that detail is available.

The core rubric looks at:

• Structure and pacing: whether the opening creates a usable dramatic engine, major turns change the story, the middle escalates, and the ending earns its landing.
• Plot and payoff: causality, clarity of goals and obstacles, setup and payoff, mystery, reveals, and whether events produce consequential change.
• Character movement: protagonist agency, internal or external change, relationships, opposition, and whether meaningful choices carry a cost.
• Reader momentum: conflict, tension, intensity, excitement, genre delivery, and how effectively individual sections sustain forward pull.
• Continuity and logic: optional checks that surface possible contradictions, missing links between events, knowledge-timing problems, and unresolved cause-and-effect questions.

The rubric is genre-aware rather than one-size-fits-all. A thriller is assessed more heavily for suspense, momentum, reversals, and danger payoff; a romance for relationship progression, chemistry, barriers, and emotional payoff; a mystery for clues, curiosity, fair-play reveals, and solution satisfaction; and literary fiction for emotional truth, thematic cohesion, interior movement, and scene-level tension. The declared form also changes the expected shape: a short story is allowed compression, while a novel, feature, television episode, stage play, or commercial treatment is judged against the structural demands of that form.

Genre qualifiers refine that emphasis. They help signal meaningful promises within a broad genre—such as cozy, detective, hard science fiction, coming-of-age, slow burn, enemies-to-lovers, military, political, supernatural, or comic—so the analysis can give the relevant elements more attention. Subgenre, tone, or structural intent is the free-text field for the context a dropdown cannot capture: a heist, gothic atmosphere, quiet character study, nonlinear mosaic, deliberately unresolved anti-mystery, or fragmented grief story.

If your story is intentionally unconventional, hybrid, subversive, nonlinear, or built outside a familiar three-act model, say so in that field. The analysis uses that context when judging the outline's promises and flags, while still examining whether the reader has enough clarity, pressure, progression, and payoff for the experience you intend. It may identify risks such as reader disorientation or an unearned ending, but those are questions to evaluate against the work's stated promise.

Scores summarize that evidence on a 100-point scale. They are developmental indicators—not objective measurements of artistic quality, predictions of commercial success, or a substitute for a human editor. The accompanying report matters as much as the number: it explains what is working, what is weakening the current version, and what to test next.

How we make scores more stable

A Complete Analysis does not rely on a single first impression. Outline Analyzer uses multiple structured assessment passes, checks for meaningful disagreement, and applies a separate stabilization step to the core scoring metrics. This reduces the influence of an unusually generous or unusually harsh individual response.

We test and tune the process through repeated trials against a curated benchmark set spanning incomplete, failing, middling, competent, and strong outlines. We examine score spread, score bands, metric stability, and whether known weaknesses are reflected in the report. The benchmark set and testing process evolve as the product does.

Consistency, variation, and revisions

AI-assisted judgment is not perfectly deterministic. A fresh assessment can still vary slightly, particularly where an outline leaves room for more than one reasonable interpretation. The structured rubric, multiple assessment passes, stabilization step, and benchmark testing are intended to reduce avoidable variance—not to present scores as laboratory measurements.

Revision Checks are designed to be more useful than unrelated reruns. When you analyze a revised version in the same project, the new pass is given the prior analyzed version and a map of the meaningful changes. Unchanged sections remain anchored to the earlier assessment; changed material receives a fresh judgment. The comparison then explains what improved, weakened, or simply changed, and why.

We publish the categories and intended interpretation of the rubric here, while keeping the detailed prompt design, weighting implementation, and model configuration proprietary. This protects the system from being optimized for a score instead of a stronger story.

Language support

Outline Analyzer is currently designed and evaluated for outlines written in English. It may be able to process material in other languages, but we do not yet represent those results as reliably supported. If a story contains short non-English names, terms, or dialogue within an otherwise English outline, that is generally workable; the overall analysis should still be based on English material.

Your work and AI

Your outline remains yours. Outline Analyzer sends your content to AI providers only when you deliberately use a feature that needs it, and does not use your creative content to train or fine-tune AI models. For the complete explanation of data handling, provider retention, and deletion options, read our Privacy Policy.

Questions or feedback

If a result seems inconsistent, overly certain, or unhelpful, we want to know. Send feedback to [email protected] with the project context you are comfortable sharing.