Trust & Evidence

How NiceCite earns trust

Every score links to live crawl evidence. Every recommendation is scoped to a real page. We do not guarantee AI citations — we guarantee honesty about what we measured and why it matters.

Does NiceCite use real crawl data?

Yes. Every audit fetches your live URLs over HTTP, parses the HTML response, extracts headings, meta tags, JSON-LD schema (including nested @graph blocks), Open Graph data, internal and external links, sitemap references, and robots.txt rules. We do not use cached snapshots, third-party estimates, or synthetic scores. The evidence panel in every report shows the exact URLs crawled and the signals found on each one.

What signals are measured?

NiceCite scores six weighted signals that map to how answer engines evaluate sources: Answer Readiness (FAQ presence, direct-answer copy, H1 clarity),Entity Clarity (who you are, what you do, where),Schema & Structured Data (Organization, FAQPage, Article, Product, LocalBusiness),Trust & Citation Signals (about, contact, authors, sources),Content Coverage (word counts, topic depth, crawlable text), andTechnical Accessibility (sitemap, robots, canonical, mobile rendering). See the methodology page for weights and definitions.

Why scores can change

Two reasons. First, your site changed — you shipped fixes, added pages, or removed schema. Second, the audited page set changed (for example, after a plan upgrade that crawls more pages). The Score Integrity panel exposes the exact page list, a page-set signature, and the score version so changes are explainable, not mysterious. We never silently re-weight signals or special-case a domain.

What free users can see

Free audits include the full six-signal NiceCite Score, per-signal breakdowns, the audited page list, schema detection, FAQ and heading analysis, and a shareable read-only report link. You can run a limited number of free audits per month. No credit card required.

What premium and credits unlock

Paid plans increase the number of pages crawled per audit, unlock deeper schema analysis, add per-domain audit history, enable PDF exports, and provide generation credits for FAQs, schema blocks, on-page rewrites, and content briefs. Credits are consumed only when you generate an asset — you pay for what you ship.

What NiceCite does not promise

We do not — and cannot — guarantee that you will be cited by ChatGPT, Perplexity, Gemini, Claude, or Google AI Overviews. No vendor can. We measure how ready your site is to be understood and trusted by answer engines, and we give you a prioritized fix list built from what your site actually contains today. Any vendor selling “guaranteed AI citations” is not telling you the truth. Read more on our about page.

How score versioning protects against bad recommendations

Each report carries a score version (NiceCite Score v2 today) and a page-set signature. When we ship a new version, historical reports remain tagged with the version they were scored under, so you can compare apples to apples. If a signal definition changes, we note it in the changelog and update the methodology page. We do not retroactively change old scores to make the product look better.

Privacy & security basics

  • We only crawl pages your site already serves publicly. We do not attempt to access admin panels, hidden routes, or password-protected content.
  • Audit data is stored against your account and is not sold to third parties or used to train external models.
  • Shareable report links are read-only and do not expose your account or billing information.
  • Account deletion and data-export requests can be initiated from Settings. We process them within 30 days.

Reviewed by the NiceCite methodology team.