Glossary / Prose Parity
Prose Parity
Also: prose parity score, AI crawler text parity
Prose Parity is Lantad's metric for the share of a page's rendered, human-visible text that is present in the raw HTML an AI crawler receives, where 100% means a crawler sees everything a human visitor sees.
Most AI crawlers fetch a page's initial HTML and do not run JavaScript. If a page builds its text in the browser after load, that text never reaches the crawler. Prose Parity measures that gap directly, as a single percentage.
A high score means the words a visitor reads are the same words a crawler receives. A low score means much of your visible content is invisible to the systems that read the web to answer questions. Being readable is a prerequisite for being cited, though it never guarantees it.
What Prose Parity measures
Prose Parity is the fraction of a page's rendered, human-visible text that is also present in the raw HTML an AI crawler receives. It is expressed as a percentage from 0 to 100. At 100%, every word a visitor reads in the browser is already in the HTML a crawler downloads, so the crawler and the human see the same content. The comparison is one-directional: the question is not whether the two versions look similar, but whether what the human sees actually shipped in the HTML. Prose Parity is one of five dimensions in Lantad's AI Visibility Score, alongside Access, Retrievability, Structure, and Schema, because text a crawler never receives cannot be read, quoted, or attributed.
Why a low Prose Parity score hides your content from AI
Most AI crawlers fetch a page's initial HTML and do not execute JavaScript. When a site assembles its text in the browser after the page loads, through client-side rendering or hydration, that text is built on the visitor's machine and never reaches the crawler. What the crawler receives is closer to an empty shell: navigation, a script bundle, and little readable prose. A low Prose Parity score is the measurable size of that gap. The page looks complete to a person and is effectively absent for the systems that read the web. Being fetched and read is a prerequisite for being cited by an AI system. It never guarantees citation, but without readable text it cannot happen at all.
How Lantad measures Prose Parity
Lantad scans a URL two ways and diffs the results. The first is a raw HTTP fetch that identifies honestly as LantadBot and runs no JavaScript, which is how most AI crawlers retrieve a page. The second is a full browser render, which is what a human visitor sees. To score the two, Lantad extracts the main content text from the rendered page, splits it into overlapping token sequences called shingles, hashes each shingle, and measures what fraction of those shingles also exist in the raw HTML. This is containment, not similarity. Two honesty mechanisms apply: pages without semantic containers are measured on all visible text, and pages that serve different content on each request are judged against their own self-similarity. Both conditions are stated in the report.
How to improve Prose Parity
The reliable way to raise Prose Parity is to deliver your content in the initial HTML rather than assembling it in the browser. Server-side rendering and static generation both ship finished text that a non-JavaScript crawler can read on the first fetch. Content loaded only after a user interaction, injected by client-only scripts, or held in hydration data with no readable HTML equivalent will not count toward parity. Rescan after a change to confirm the gap closed. Higher Prose Parity makes your text available to AI crawlers, which is a precondition for being read and cited. It does not by itself determine ranking or selection, which no public system fully discloses.
Common questions
- What is a good Prose Parity score?
- Aim as close to 100% as possible, where the crawler receives essentially all of the text a visitor sees. The further below 100% a page falls, the more of its visible content never reaches AI crawlers. Lantad also folds the score into a letter grade from A to F so you can see where a page stands at a glance.
- Does Prose Parity affect whether AI tools cite my page?
- Indirectly. A crawler cannot quote or attribute text it never received, so readable content is a prerequisite for citation, and Prose Parity measures that prerequisite. It does not guarantee a citation. To be cited you must first be readable, allowed by robots rules, and clearly structured, and even then selection depends on factors that no public AI system fully discloses.
- Why is my Prose Parity low when the page looks complete in a browser?
- Because your browser runs JavaScript and most AI crawlers do not. If the page renders its text client-side or through hydration after the initial HTML loads, a human sees a finished page while a non-JavaScript crawler sees an empty shell. Lantad's raw fetch reproduces exactly what that crawler receives, which is why the two views can differ.
See if AI can read your site
Paste a URL and see exactly what AI crawlers can and cannot read, with a graded report and ranked fixes. Free, no signup.