Soothsay · LLMEO audit for writers
ChatGPT, Perplexity, and Gemini are quietly rewriting how readers find your work. Soothsay grades your draft on LLMEO — whether AI engines will cite it — and gives you specific findings and one-click rewrites. Built for the writer, not the agency.
The main event
Large Language Model Engine Optimization. The reading we’re built around. We score whether this specific draft is the kind of writing ChatGPT, Perplexity, and Gemini reach for when they answer someone’s question.
We also check
SEO
ATitle, meta, headings, schema, links. The classics — for the readers who still arrive via Google.
Readability
AFlesch-Kincaid, sentence variance, passive voice. Plain English, measured. The Hemingway part.
Originality & Voice
CInternal repetition, burstiness, cliché density. We tell you when it sounds like a robot wrote it.
On the engine
About seventy percent of the audit is deterministic — DOM parsing, schema validation, lexical statistics. The LLM only touches the rewrites and the human-language finding explanations. That means consistent scores, fast audits, and a path to running the whole thing on weights you control.
A sample LLMEO finding
Structured data is the single most reliable way to tell AI engines what your page is and who wrote it. Without it, LLMs will fall back to body-text heuristics — and often guess wrong.
Suggestion: Add an Article JSON-LD block with author, datePublished, dateModified.
LLMs lean heavily on the first 100–200 words to decide what a piece is about. Without a TL;DR or 'in short' passage, they may miss your thesis.
Suggestion: Add a 2–3 sentence summary before the first H2.
AI engines pull from FAQPage blocks roughly 3× more often than equivalent prose.
Suggestion: Wrap your Q&A passages in a FAQPage JSON-LD block.
For writers, not agencies
No dashboards over teams over crawls over projects. Paste a draft. Get a reading. Fix what matters.
Open weights, your call
Soothsay runs on Llama 3.3 70B via managed inference. Open weights, swap-friendly, with a clean path to fully self-hosted when it makes sense.
LLMEO-first, SEO-aware
Google still matters. But the next reader of your work is increasingly a model. We score for both, and tell you when they pull in different directions.