How The K Search Visibility Score™ Works

What is The K Search Visibility Score™?

The K Search Visibility Score™ is a 0–100 measure of how easily customers can find, trust, and choose your business online. It looks beyond traditional SEO scores by combining technical health, content relevance, local presence, authority signals, AI search readiness, and conversion readiness into a single number with a letter grade and status band.

The score is generated from evidence collected on your live website — not from estimates or third-party rank trackers. Every category score links back to the specific pages and signals we found.

How websites are crawled

The K Search Visibility Score™ evaluates the website using sitemap and internal-link discovery to assess the pages that contribute to overall search visibility — it is never a homepage-only audit. When you enter a URL, the checker reads your robots.txt, locates and expands your XML sitemaps recursively (following nested sitemap indexes up to 10 levels deep), and follows internal links from the homepage to build a full list of URLs before analysis begins.

Pages are fetched respectfully, with concurrency limits (8 parallel fetches, 4 for sitemaps) and per-request timeouts, and only publicly accessible pages are read. For each page we capture titles, meta descriptions, heading structure, body content, structured data, internal and external links, contact details, and conversion elements. Performance signals come from Google PageSpeed Insights, pulled for representative templates rather than every URL.

Pages discovered vs pages analyzed

Pages discovered is every unique URL we found through navigation, internal links, and your sitemap. Pages crawled is the subset we selected to fetch. Pages successfully analyzed is the number that returned usable content and were used in scoring.

The gap between these numbers is normal. Redirects, blocked paths, timeouts, login-protected pages, and non-HTML files are discovered but not analyzed.

Smart Sampling: how pages are chosen

Discovery is deliberately wide, but analysis is sampled intelligently. If your site has 50 or fewer discoverable URLs, every page is analyzed in full. Larger sites are analyzed with stratified, template-aware Smart Sampling: URLs are classified into page types (homepage, service, product, job detail, article, article archive, author, location, conversion, and more), and the sample includes multiple pages from every significant page type.

The sample size grows gradually with website size — 50 pages for small sites (up to 500 URLs), 75 for medium sites (up to 5,000 URLs), and up to 100 for large and enterprise sites — with a hard ceiling of 100 pages to keep runs fast and polite. When a known platform is detected, its typical template structure informs the classification so the sample matches how your site is actually built.

The report shows the full Smart Sampling Coverage panel: URLs discovered, page types detected, pages sampled per type, URL coverage, and page-type representation. Page-type representation — not the percentage of URLs crawled — is the more meaningful measure for large websites, and site size never lowers your score.

Crawl coverage and page-type representation

Crawl coverage is pages successfully analyzed divided by pages discovered, shown as a percentage. It is reported for transparency, but it is not the primary driver of confidence on large sites: a 20,000-page site will always show a small URL percentage even with a perfectly representative sample.

That is why the report also shows page-type representation — the share of significant page types covered at an adequate depth. Large sites reuse templates, so template coverage is what tells you how well the sample represents the whole site.

Confidence levels

Every report includes a confidence level of High, Medium, or Low. Confidence is computed from how well the analyzed sample represents your site: page-type coverage, template representation, sample diversity (pages from different sections), sample adequacy (pages analyzed vs the size-aware target), and crawl success rate. URL percentage is informational only.

Confidence never changes the score itself. It describes how much evidence supported the score — a large website is judged on the same scale as a small one, with uncertainty expressed through confidence rather than a penalty.

How we discover your website pages

Before anything is scored, we build the most complete list of your pages we can. The K Search Visibility Score™ first attempts to locate your website's XML sitemap — the file your platform publishes to tell search engines what exists.

The system supports Rank Math, Yoast SEO, AIOSEO, Shopify, Wix, Squarespace, and custom sitemap formats. If a sitemap index exists, the crawler automatically follows every child sitemap it links to, and keeps going recursively through nested sitemaps (up to 10 levels deep) until all available URLs are discovered. Duplicate URLs, images, and media files are removed before analysis, so the result is a clean, deduplicated inventory of your real pages — a far more complete picture than a homepage-only audit.

sitemap_index.xml
├─ page-sitemap.xml
├─ post-sitemap.xml
├─ category-sitemap.xml
└─ product-sitemap.xml

What happens if no sitemap exists?

Plenty of websites have no sitemap, or publish one that cannot be read. When that happens the platform automatically switches to internal-link discovery instead of returning a thin result.

• ✓Crawling begins from your homepage.
• ✓Internal links are followed outward, discovering up to 500 internal URLs.
• ✓This also reveals orphaned content and crawlability issues — pages that exist but are hard to reach.
• ✓The fallback system prevents incomplete audits, so you always get a usable score.

Your report always states which method was used. If it says link discovery was used, publishing a valid sitemap is usually the fastest single improvement you can make.

Why URL discovery matters

Discovery is not a technical detail — it is the ceiling on your visibility. Search engines can only rank pages they can discover, and AI search engines can only reference content they can access. Missing, blocked, or orphaned pages quietly reduce your visibility no matter how good the content is.

A complete crawl also produces a more accurate visibility score. The more page types we can see, the more evidence sits behind every category, and the higher your page-type representation and confidence level.

Assessment categories

Each of the eight categories is scored 0–100 from verified evidence. Five of them combine into your overall visibility score; the other three are scored and reported as their own dimensions.

How the score is calculated

Underneath the eight reported categories, every assessment examines ten scoring pillars. Individual pages are evaluated against these pillars, and the results are combined into your overall visibility score — so one strong page cannot mask a weak site, and one weak page cannot sink a strong one.

How was my score calculated?

Your overall score is the weighted average of the five scoring components, calculated only from evidence we could verify during the crawl. Sub-signals are graded on presence, quality, completeness, and validation — so a signal earns credit for being implemented well, not just for existing. The report shows the exact breakdown behind each component, your page-type representation, and your confidence level so you can see why the number is what it is.

Overall score weightings

These are the base weightings, tuned for small and mid-sized businesses competing in both classic search and AI-generated answers. Your detected industry profile may adjust them — for example, local businesses are weighted slightly toward local search and enquiry paths. Local SEO, Social Authority, and Entity Authority are scored and reported as their own dimensions, and feed into the Search, Authority, and AI composites.

Transparency & crawl coverage

Every report shows exactly how much of your website was evaluated, so you never have to take a score on faith. These metrics appear in the Assessment Details and Smart Sampling Coverage panels of your results.

Together these numbers make the audit auditable: you can see how many sitemap files we read, how many URLs exist, which pages were sampled and why, how well the sample represents your page types, and exactly when it happened.

Scoring framework

Each dimension is scored on a 0–100 scale from a set of sub-signals. A sub-signal scores on how well it is implemented — presence, quality, completeness, and validation — not merely whether it exists, so a signal never earns full marks just for being present. Sub-signals return verified (full credit), partial (proportional credit), or absent (no credit) states.

The overall score is the weighted average of five components: Search Visibility 40%, AI Search 25%, Authority & Trust 15%, Technical SEO 10%, and Conversion Readiness 10%. Local SEO, Social Authority, and Entity Authority feed into those composites — they are scored and reported as their own dimensions but do not carry separate weight in the overall number.

Scores map to letter grades from A to F. The report also compares your score to an industry benchmark with your percentile rank.

Industry profiles

The checker detects your business type from evidence on the site — signals in your homepage and main sections, structured data, navigation, and platform — and applies an industry profile: Local Business, Professional Service, Marketing/PR Agency, Recruitment Platform, Job Board, SaaS, E-commerce, Publisher/Blog, Healthcare, or Nonprofit. The industry you optionally select in the form also helps.

Each profile adjusts how the five overall components are weighted and which signals are treated as Critical versus Recommended. A job board, for example, is weighted toward crawlable listings and structured data, while LocalBusiness schema and location pages are low priority. This keeps recommendations relevant instead of generic.

Before the report is generated, findings are deduplicated (the same issue appears once, in its strongest form) and re-checked against your detected business type, so advice that does not apply to how your business actually works is filtered out rather than shown as a critical issue.

Platform detection

The checker identifies the platform your site runs on when it can — SmartJobBoard, Shopify, WordPress, Wix, Squarespace, Webflow, or Custom Platform — and reports it with a confidence level. This is informational only: no platform is rewarded or penalized.

Its practical value is context. Known platforms have known structural constraints, so when a structured data recommendation may be limited by the platform, the report says so: "Some structured data opportunities may be limited by the current platform." That way you know whether an improvement is something you can control directly.

Crawl logic

Discovery starts with the submitted URL, then reads /robots.txt and /sitemap.xml (including sitemap indexes, followed recursively). Internal links are followed from the homepage to prioritize navigation targets, service pages, location pages, About and contact pages, and recently updated URLs. If no sitemap can be read, link discovery from the homepage can find up to 500 internal URLs.

Fetches use timeouts and concurrency limits; non-HTML assets, query-parameter duplicates, and off-domain URLs are excluded. Directives in robots.txt are respected, which means disallowed sections are discovered but never fetched.

Pages are then classified by template and sampled per Smart Sampling (see above), and the results are aggregated at template level — so a missing meta description found on most service pages is reported as a likely template-wide issue, not as dozens of separate page findings.

Evidence-based scoring

No sub-signal is scored from assumption. Every point awarded is tied to an artifact captured during the crawl: a tag, a schema block, a heading, a phone number, a link, a form, or an API metric. The report surfaces this evidence per category so findings can be independently verified.

When a signal cannot be observed, it is recorded as absent rather than estimated. Findings also carry their own confidence level (High, Medium, or Low) reflecting detection certainty — separate from the site-wide Site Coverage Confidence.

Confidence scoring

Site Coverage Confidence is computed from page-type coverage, template representation, sample diversity, sample adequacy, and crawl success rate — never primarily from the percentage of total URLs crawled. For small sites, URL share plays a modest role; for large sites, page-type representation dominates, because templates repeat.

Confidence never changes the score. A confidence of Low means the crawl saw little of your site, so findings should be treated as directional; it does not reduce the number. The same principle applies to individual findings, which carry their own detection confidence.

AI Readiness vs AI Visibility

AI Readiness measures technical preparedness: how well your site can be understood, parsed, and cited by AI systems. It is composed of structural checks (60%) — AI crawlability, structured data, E-E-A-T signals, citable content, and entity markup — and your AI Citation Readiness Score™ (40%). Each check scores 100 when verified, 50 when partly in place, and 0 when not detected.

AI Visibility measures something different: your actual off-site presence and citation likelihood. It is based on real-world entity signals — knowledge graph entries (Wikipedia, Wikidata), verified social profiles, third-party citations, brand mentions, and content footprint — not on schema. A site can be technically ready while still having limited recognition, and the report explains when that gap exists: technically well prepared for AI citation, but limited current off-site brand recognition.

The AI Citation Readiness Score™ is built from five pillars: AI Crawlability, Structured Data, E-E-A-T Signals, AI-Citable Content, and Entity Authority. It also produces a per-platform citation probability estimate for ChatGPT, Gemini, Claude, Perplexity, and Google AI Overviews: a likelihood of High at 70 or above, Medium at 45–69, and Low below 45. These are modeled estimates from your site's signals, not measurements of actual citations.

Entity Authority methodology

Entity authority measures how unambiguously your business can be identified and corroborated. Inputs include consistent brand naming, Organization schema with sameAs links, a described founder or author entity, linked and verifiable social profiles, consistent NAP data, external citations, and an About page that states who you are and what you do.

Consistency across sources weighs more than volume: conflicting names, addresses, or profile links reduce the score even when many profiles exist.

Local SEO methodology

Local scoring evaluates NAP presence and consistency across the crawled pages, dedicated location or service-area pages, embedded maps and directions, LocalBusiness or relevant sub-type schema with address and geo fields, opening hours, review markup, and Google Business Profile references discoverable from the site.

Businesses without a physical service area are scored against a service-area model so that online-only businesses are not penalized for lacking an address — and for business types where local search is not relevant at all, local signals are treated as informational.

Limitations of website crawling

A crawl-based assessment sees what is publicly rendered and served. It cannot read gated pages, content behind logins or paywalls, sections blocked in robots.txt, data inside third-party widgets that load after interaction, or off-site signals such as backlink profiles, Google Business Profile dashboards, and analytics.

Heavily client-rendered sites, aggressive bot protection, rate limiting, or slow responses can reduce coverage and therefore confidence. Scores are a practical snapshot of observable signals, not a replacement for a full manual SEO audit.

Frequently asked questions

Ready to see your own numbers? Run your free visibility check.

Built for search visibility in the AI era

Being found online no longer means one thing. Customers still search Google, but they increasingly ask ChatGPT, Gemini, Perplexity, Claude, and Google AI Overviews — and those systems only reference websites they can crawl, parse, and trust. Modern visibility requires optimizing for both at once.

The K Search Visibility Score™ is designed to evaluate both. Complete URL discovery, structured data, entity clarity, authority signals, and conversion readiness are scored side by side with classic technical SEO, and AI Readiness is kept separate from AI Visibility so you can see the difference between being technically prepared and being genuinely recognized.

Run your free visibility check and see your page-type coverage, pages analyzed, and score in about 60 seconds.