Skip to main content

How it works

From a URL to an evidence-backed action plan

RankInAI reads your public website the way a search crawler does, scores what it finds against a fixed rubric, and turns every gap into a specific change you can make.

  1. 1

    Enter the website

    Supply the URL, business name, industry, target audience, primary services, main location and up to five competitor URLs. You confirm you are authorized to audit the primary website.

    • The URL is validated before anything leaves our servers: only http and https, no credentials in the address, no private or reserved network ranges, no cloud metadata endpoints.
    • The extra business context is used to check terminology alignment — whether the words you use to describe your services actually appear on the site.
  2. 2

    RankInAI crawls and analyzes it

    A queued worker fetches robots.txt, your sitemap and llms.txt, then crawls up to 50 public pages depending on your plan, prioritizing the pages that matter most for AI visibility.

    • The crawler identifies itself as RankInAI-Auditor and honors robots.txt rules for that agent.
    • It never submits a form, never signs in, never executes a downloaded file and skips cart, checkout, account and login URLs entirely.
    • Homepage, about, services, FAQ, contact, case studies and editorial pages are prioritized over deep or parameterized URLs.
  3. 3

    The audit engine scores seven categories

    Roughly seventy checks run against the collected evidence. Each check awards a fraction of its available points and records exactly what was found.

    • Scoring is deterministic: the same page content always produces the same score. Nothing is randomized.
    • Checks that cannot apply to a site are excluded from both the numerator and the denominator, so nothing is penalized unfairly.
    • An optional AI enhancement can improve the wording of summaries — it can never change a score or alter observed evidence.
  4. 4

    It produces actionable recommendations

    Every failing or partial check becomes an action, ordered by severity, then impact, then effort, and grouped into three horizons.

    • Do first — high-impact fixes, most completable in a single working session.
    • Next 30 days — meaningful improvements needing coordination or content work.
    • Longer-term improvements — investments that compound, such as original research and deep service pages.
  5. 5

    You track improvements over time

    Re-run the audit after making changes. The report shows the score change against your previous audit of the same domain.

    • Historical score tracking is included on Growth and Agency plans.
    • Because scoring is deterministic, a score change reflects a real change on the site rather than measurement noise.

Methodology

How the overall score is calculated

Each category produces a 0-100 score from its own checks. The overall score is the weighted sum of the categories that could be measured.

Category score

category_score = round(100 × Σ(points_awarded) / Σ(points_possible))

  — summed only over checks that apply to this website
  — a check that cannot apply is excluded from both sums

Overall score

overall = round( Σ category_score × effective_weight )

effective_weight = base_weight / Σ(base_weight of available categories)
Category base weights
CategoryBase weight
Technical AccessibilityWhether AI crawlers and answer engines can reach, load and index your pages at all.15%
Entity ClarityHow unambiguously your site states who the business is, what it does and who it serves.17%
Content AuthorityDepth, originality and expertise signals in the content a model would draw on.18%
Answer ReadinessHow easily individual passages can be extracted as direct answers to real questions.17%
Structured DataMachine-readable schema markup describing your organization, content and offerings.13%
Trust and EvidenceAuthorship, contact details, citations and proof that support what the site claims.15%
Competitive VisibilityHow your observable signals compare with the competitors you supplied.5%

Competitive Visibility rebalancing

If you supply no competitor URL — or none can be reached — Competitive Visibility is marked unavailable and its 5% weight is redistributed across the other six categories in proportion to their base weights. The overall score always sits on the same 0-100 scale.

Worked examples, the full check list and every point allocation are documented in AUDIT_METHODOLOGY.md in the product repository.

Honest limits

What an audit can and cannot tell you

RankInAI measures observable website signals. It is important to be precise about what that does and does not prove.

What RankInAI does

  • Measures signals directly from your website with recorded evidence
  • Scores AI visibility readiness against a fixed, published rubric
  • Tells you which specific changes may improve discoverability
  • Compares your observable signals against competitors you supply
  • Optionally reports publicly visible search results, clearly labeled as such

What no tool can do

  • Guarantee that ChatGPT, Perplexity, Gemini or Copilot will mention your brand
  • See inside any AI system’s training data, index or ranking logic
  • Promise a position, a citation rate or a volume of referrals
  • Measure how often you are currently mentioned inside a private AI product
  • Replace the underlying work of publishing genuinely useful content

Every report distinguishes clearly between signals measured directly from your website, optional public-web search observations, AI-generated interpretation and sample or demo information.

See every check the audit runs

Watch it run before you buy

The instant demo takes five seconds. The free homepage preview runs on your real website without an account.

Try the free demoSee pricing

One-time audits from $49. Monthly plans from $29. All prices in US dollars.