Unlike traditional search index rankings, AI engine generators extract and synthesize content. Read this detailed breakdown of what our grader measures, why AI bots prioritize it, and how to update your code.
This measures whether your site allows AI indexing user-agents to crawl your file structures. Traditional search bots use `Googlebot` or `Bingbot`. AI chatbots and search engines utilize specialized scrapers, such as `GPTBot` (OpenAI), `ClaudeBot` (Anthropic), `PerplexityBot` (Perplexity), and `Googlebot-Extended` (Gemini overview layers).
If your `robots.txt` denies crawler access to these agents, LLM indices cannot read your updates. Consequently, your brand name will not be represented, cited, or recommended in natural conversation responses.
robots.txt does not use Disallow: / under user-agents *, GPTBot, ClaudeBot, PerplexityBot, or Googlebot-Extended.
Allow: /docs/ or Allow: /blog/.
Structured metadata represents machine-readable scripts embedded in HTML, usually formatted in **Schema.org JSON-LD**. It explicitly classifies the entity types on your page (e.g. telling the parser "this page represents a Company, this item is a Service, and these fields are FAQs").
AI retrievers use semantic graphs to link concepts together. By embedding explicit schemas, you help bots resolve context ambiguities without having to guess, increasing their confidence score when querying your page metadata.
Organization and WebSite JSON-LD script block containing address parameters, corporate logo links, and parent company indicators.
Product, Service, FAQPage, or TechArticle.
This audits how easily your content can be summarized into a single sentence or bullet point. It reviews whether you provide concise, definition-heavy text blocks at the beginning of layout blocks.
Generative answers are highly condensed. Bots prioritize pages that contain clear, brief summaries of topics. If an LLM finds an easy-to-quote summary paragraph, it is much more likely to pull it as a direct quote and place a citation footnote referencing your site.
This measures the density of domain-specific terminology (vocabulary and jargon) on your page. It analyzes if the content uses the precise technical words and nomenclature associated with your industry sector.
LLMs represent meanings in a vector semantic space. By writing with precise domain terminology, your text aligns closely with the search vectors AI models construct when processing user questions, proving that your site is a highly authoritative and relevant resource on the topic.
OpenGraph metadata tags represent XML/HTML headings (like `og:title`, `og:description`, and `og:image`) placed in your page's `
`. They define the preview title, description, and thumbnail image shown when a link is shared.When AI systems cite a source, they often render a visual preview card. Standardizing your OpenGraph metadata guarantees that Perplexity or ChatGPT can generate a sleek, visually complete card link, increasing click-through rates back to your website.
<meta property="og:title" content="..."> and <meta property="og:description" content="..."> to the head tags of all pages.
og:image (ideally 1200x630 pixels) to prevent empty graphic grids.
This assesses the density of objective statistics, technical specs, numbers, list blocks, and data tables on your page. AI parsers prioritize organized databases over unstructured prose.
Generative models are trained to extract facts. Numerical specifications and tabular layouts represent highly condensed packets of knowledge. By structuring information in HTML tables (`