About willaicite
willaicite is a free audit that scores how ready a web page is to be retrieved and cited by AI answer engines: ChatGPT, Perplexity, Google AI Overviews and Claude.
How does the audit work?
An audit makes at most ten polite requests to the target site, honoring robots.txt throughout:
- the page itself, fetched twice: once as a normal browser, once as an AI crawler, to catch firewall rules that block bots
- robots.txt, checked against the full roster of AI retrieval and training crawlers
- the sitemap, an about page, the favicon and llms.txt
The scoring is deterministic. There are no LLM calls and no sampling; every point is tied to observable evidence (an HTTP status, a robots.txt rule, a byte count of extractable text). When a check cannot run, the report marks it “could not verify” and leaves it out of the weighted score.
Even the verdict line is deterministic, drawn from fixed score bands: a site in the fifties reads “Needs work: measurable gaps are limiting how AI engines retrieve and cite this site” while a clean run reads “Excellent: well positioned to be retrieved and cited by AI answer engines”.
What do the nine dimensions measure?
| dimension | weight | question it answers |
|---|---|---|
| AI crawler access | 3 | Can AI crawlers fetch the page, and does the firewall treat them differently from a normal browser |
| Renderability | 3 | Is the content readable without running JavaScript |
| Answer-readiness | 3 | Is there a definition, heading or answer block an engine can quote whole |
| Topical focus & metadata | 3 | Do title, headings, description and body present the same one topic |
| Evidence density | 2 | Does the page carry statistics, quotations and cited sources for engines to quote once it is retrieved |
| Structured data | 2 | Does JSON-LD tell engines who wrote what, and when |
| Freshness | 3 | Is there a machine-readable date from the last 90 days |
| Entity & E-E-A-T | 2 | Can the author and organization behind the page be identified |
| SEO foundation | 1 | Does the canonical URL point at the page itself, and are titles and descriptions unique across pages |
What evidence backs the scoring?
The weights and recommendations follow the published GEO research, leaning on controlled and replicated results over one-off studies. Aggarwal et al., “GEO: Generative Engine Optimization”, KDD 2024 measured that citing sources lifted generative visibility by 24.9%, adding statistics by 25.9%, and quotations by 27.8%, with the page already retrieved.
The 2025–2026 replications narrowed that claim:
- Vishwakarma et al., SIGIR 2026 ran 252,000 controlled trials across six LLMs and found topical relevance and retrieval position decide most citations.
- C-SEO Bench, NeurIPS D&B 2025 found only 3 of 54 content-rewrite combinations made a significant difference.
- The GEO-16 field audit (2025) found metadata, freshness and structured data the signals most strongly tied to 1,702 observed citations.
That is why topical focus and freshness carry a high weight (3) while evidence density carries a medium one (2).
The crawler roster tracks the engines’ own documentation (OpenAI, Anthropic, Perplexity). Blocking rates come from measurement: Fletcher, Reuters Institute, 2024 found “48% of the most widely used news websites … were blocking OpenAI’s crawlers” by the end of 2023, while 24% blocked Google’s AI crawler. Findings from single studies are presented as directional, never as guarantees.
What happens to your data?
An audit touches only the site you ask it to audit, honoring robots.txt. There are no trackers, no analytics scripts and no external resources on these pages. Questions and corrections are welcome: hello@willaicite.com.
About FAQ
Is willaicite free?
Yes. Audits are free, require no account and set no cookies. Rate limiting keeps the load light: a visitor gets a batch of audits every ten minutes rather than an unlimited firehose.
Why does the audit stop at ten requests?
Courtesy to the site being audited. Ten requests are enough to score all nine dimensions (the page twice, robots.txt, the sitemap, an about page, the favicon and llms.txt), and a small fixed cap means an audit can never hammer the site it is inspecting. The audit honors robots.txt while it works.