GEO Content Auditor — GEO Audit & AI Answerability Score avatar

GEO Content Auditor — GEO Audit & AI Answerability Score

Pricing

from $40.00 / 1,000 page auditeds

Go to Apify Store
GEO Content Auditor — GEO Audit & AI Answerability Score

GEO Content Auditor — GEO Audit & AI Answerability Score

Run a GEO audit on any web page: AI answerability score 0-100, five sub-scores, quick wins, missing schema and Q&A ideas. USD 50 per 1,000 pages.

Pricing

from $40.00 / 1,000 page auditeds

Rating

0.0

(0)

Developer

Muhammad Afzal

Muhammad Afzal

Maintained by Community

Actor stats

0

Bookmarked

2

Total users

1

Monthly active users

20 days ago

Last modified

Categories

Features

Limited permissions

Share

GEO Content Auditor runs a GEO audit on the web pages you list and returns, for each URL, an AI answerability score from 0 to 100, five sub-scores (structure, facts, entities, readability, authority), a letter grade, strengths, prioritized issues, three quick wins, missing schema types and suggested question-and-answer pairs. It costs USD 50 per 1,000 audited pages on the FREE tier (USD 0.050001 per page), and pages that cannot be fetched or audited are not charged.

Free users: maximum 20 results per run. Paid Apify plans use your maxPages setting, up to 50 pages per run.

It is built for content teams, SEO and GEO agencies and AI agents that need page-level fixes to make content easier for AI models to extract, summarize and cite. An AI model reads each page's text and markup and scores it against a fixed rubric; the Actor does not query AI assistants or measure whether they cite you.

What does a GEO audit return for each page?

Every audited URL becomes one dataset row with 21 fields.

FieldWhat it tells you
url, page_titleThe audited page and its HTML title
ai_answerability_score0–100, the sum of the five sub-scores; null if the audit failed
structure_score0–30: question-style headings, a clear answer near the top, FAQ blocks, lists and tables, heading hierarchy
facts_score0–25: specific numbers, dates, named entities, definitions and examples
entity_score0–20: brand clearly defined, author byline, JSON-LD, product or service named
readability_score0–15: clean semantic structure, chunkable content, llms.txt presence
authority_score0–10: cited sources, publication date, internal links, breadcrumbs
gradeA (90+), B (75–89), C (60–74), D (40–59) or F
strengths, issuesWhat works and what to fix, most impactful issue first
quick_winsThe three highest-impact fixes, each about 15 words or fewer
missing_schema_typesJSON-LD types that would help, such as FAQPage or Organization
suggested_qa_pairsBuyer questions the page should answer, each with a one-sentence answer hint
prompt_fit_score0–100 fit with your target prompts; null when none are given
llms_txt_present, has_json_ld, word_countMeasured page and site signals
model, error, audited_atNeutral label of the built-in AI model configuration (for example built-in-ai-2026-10), failure reason and timestamp

A SUMMARY record in the run's key-value store adds the site-level score and grade, the ten most frequent issues, quick wins, the five lowest-scoring priority pages, a count of missing schema types and the URLs that could not be fetched.

How is the AI answerability score calculated?

The Actor fetches each page once, extracts the title, meta description, up to 30 H1–H3 headings, JSON-LD types, word count, the first 3,000 characters of body text and whether /llms.txt exists on the domain, then asks an AI model to score five dimensions with fixed maximums (30, 25, 20, 15 and 10 points). The overall score is their sum, so you can see which dimension costs you the most points. Scores are model judgments on a rubric, not a measurement of real AI citations, and they can move a few points between runs.

How do I audit pages against target prompts?

Add the buyer questions each page should answer to targetPrompts. The model then returns prompt_fit_score (0–100) for how well the page answers them.

{
"startUrls": [
{ "url": "https://apify.com/about" },
{ "url": "https://www.rust-lang.org/" }
],
"brandName": "Apify",
"targetPrompts": ["What is the best web scraping platform?"],
"maxPages": 10
}
  • startUrls (required): direct HTTP(S) page URLs. Each is fetched once; duplicates are skipped. The Actor does not follow links or crawl a site.
  • brandName: used to judge entity clarity. Leave it empty to let the model infer the brand from each page.
  • targetPrompts: optional buyer questions for prompt-fit scoring.
  • maxPages: 1–50, default 10. URLs beyond the limit are ignored and not charged.

When should I use a GEO audit tool like this, and when not?

Use it to prioritize content rewrites: run your key landing pages, guides and product pages, sort by ai_answerability_score, and work through quick_wins and missing_schema_types.

Do not use it for:

Output example

Abridged rows from owner runs on 2026-10-20 and 2026-10-26.

[
{
"url": "https://apify.com/about",
"page_title": "About · Apify",
"ai_answerability_score": 69,
"structure_score": 16,
"facts_score": 21,
"entity_score": 18,
"readability_score": 9,
"authority_score": 5,
"grade": "C",
"quick_wins": ["Add a concise 'What is Apify?' summary near the top of the page.", "Label every headline metric and include its measurement date and definition."],
"missing_schema_types": ["FAQPage", "BreadcrumbList"],
"llms_txt_present": true,
"has_json_ld": true,
"word_count": 2756,
"prompt_fit_score": null,
"error": null
},
{
"url": "https://www.rust-lang.org/",
"page_title": "Rust Programming Language",
"ai_answerability_score": 57,
"grade": "D",
"issues": ["No JSON-LD or other structured data is present, limiting explicit machine-readable identification of the Rust project and its resources."],
"quick_wins": ["Add a short \"What is Rust?\" definition and direct summary near the top."],
"missing_schema_types": ["Organization", "WebSite", "FAQPage"],
"suggested_qa_pairs": [{ "question": "What is the Rust programming language?", "answer_hint": "Rust is a systems programming language designed to help developers build reliable, memory-safe, and efficient software." }],
"error": null
}
]

Pricing: how much does a GEO audit cost?

Pay per event; Apify platform usage and AI model costs are included. The page-audited event is charged only after a successful audit row is saved.

Free users: maximum 20 results per run.

EventFREEBRONZESILVERGOLD / PLATINUM / DIAMOND
Actor start (per started GB of allocated memory)USD 0.006251USD 0.006094USD 0.005938USD 0.005
Page audited (primary)USD 0.050001USD 0.04875USD 0.0475USD 0.04

The default memory is 512 MB, so a normal run has one start event.

  • Default input, 1 page (FREE): 0.006251 + 0.050001 = about USD 0.06.
  • 50 pages (GOLD): 0.005 + 50 × 0.04 = about USD 2.01.

Pages that return HTTP 400 or higher, fail to load, or fail the AI step are not charged. A maximum cost per run stops the Actor before the next page would exceed it.

How long does an audit take, and what can fail?

A one-page default run took about 10 seconds on 2026-10-26; slow or very large pages took up to about a minute in owner tests. Pages are fetched with up to two retries and audited one at a time, with a 90-second limit for each AI request. Unreachable URLs are listed in SUMMARY.failed_urls; an AI failure saves a row with an error and no score. If the AI service rejects the request for account or spending reasons, the run stops and keeps the rows already saved.

Use it from an AI assistant

GEO Content Auditor works from Claude, ChatGPT, Cursor or any MCP client through Apify's MCP server. Add https://mcp.apify.com/?tools=muhammadafzal/geo-content-auditor to your client.

Example prompt: "Run a GEO audit on our pricing page and our three most-read blog posts, then list the quick wins and missing schema types for the lowest-scoring page."

Run it from the API

JavaScript (apify-client):

import { ApifyClient } from 'apify-client';
const client = new ApifyClient({ token: process.env.APIFY_TOKEN });
const run = await client.actor('muhammadafzal/geo-content-auditor').call({
startUrls: [{ url: 'https://apify.com/about' }],
brandName: 'Apify',
maxPages: 10,
});
const { items } = await client.dataset(run.defaultDatasetId).listItems();
const summary = await client.keyValueStore(run.defaultKeyValueStoreId).getRecord('SUMMARY');
console.log(items, summary?.value);

Python (apify-client 3.x):

from apify_client import ApifyClient
client = ApifyClient(token='YOUR_APIFY_TOKEN')
run = client.actor('muhammadafzal/geo-content-auditor').call(run_input={
'startUrls': [{'url': 'https://apify.com/about'}],
'targetPrompts': ['What is the best web scraping platform?'],
})
items = client.dataset(run.default_dataset_id).list_items().items

FAQ

What is a GEO audit?

A GEO (generative engine optimization) audit reviews whether a page is easy for AI models to read, understand and quote. This Actor scores structure, facts, entities, readability and authority for each page and lists concrete fixes.

Does a high score guarantee that AI assistants will cite my page?

No. The score measures page qualities that make extraction and citation easier. Whether an assistant cites you also depends on authority, competition and the question asked; track that separately with the AI Visibility Tracker.

Do I need an API key for an AI provider?

No. Model costs are included in the per-page price.

Can it audit a whole website?

It audits only the URLs you list, up to 50 per run. List the pages that matter, or use the AI Search Visibility Audit to crawl a site with rule-based checks.

What is llms.txt and does the score need it?

llms.txt is a plain-text file at the site root that summarizes the site for language models. Its presence adds to the readability dimension but is a small part of the total.

How often should I re-audit?

After each meaningful content change, or monthly for key pages, so you can confirm the score moved.

Responsible use

Audit only public pages you own or are allowed to analyze, and respect each site's terms and reasonable request rates. Recommendations are AI-generated; review them before publishing changes.

Last verified: 2026-10-27