{
  "study": "Najdou vas AI vyhledavace, i kdyz je pustite dovnitr?",
  "publisher": "CIAD - Cesky institut pro AI a data",
  "url": "https://ciad.cz/studie/najdou-vas-ai-vyhledavace/",
  "license": "CC BY 4.0",
  "generated": "2026-08-03",
  "scan_date": "2026-08-03",
  "note_on_data": "Agregatni data bez domenovych jmen a bez jmen firem. Vzorek je konvenicni (vyzkumny korpus CIAD pouzivany napric paralelnimi studiemi), ne nahodny nebo reprezentativni vzorek ceske ekonomiky. Studie navazuje na CIAD studii o robots.txt policy (viz https://ciad.cz/studie/ai-crawlery-ceske-weby/) a mereni denominatoru se mezi studiemi lisi.",
  "sample": {
    "total_domains_targeted": 299,
    "n_unreachable_structured_data_scan": 46,
    "n_reachable_structured_data_denominator": 253,
    "n_unreachable_sitemap_scan": 41,
    "n_reachable_sitemap_denominator": 258,
    "n_domains_for_robots_sitemap_declaration": 299,
    "sampling_method": "convenience sample: CIAD research corpus used across parallel studies, not random or representative of the Czech economy"
  },
  "structured_data": {
    "denominator": 253,
    "methodology_note": "Static HTML fetch only, no JS execution: measures what crawlable HTML actually carries, matching how search/AI crawlers consume JSON-LD.",
    "counts": {
      "zero_structured_data": 155,
      "any_structured_data": 98,
      "jsonld": 92,
      "microdata": 17,
      "rdfa": 1
    },
    "pct": {
      "zero_structured_data": 61.3,
      "any_structured_data": 38.7,
      "jsonld": 36.4,
      "microdata": 6.7,
      "rdfa": 0.4
    },
    "note": "A domain may carry more than one format at once (e.g. JSON-LD and microdata together), so jsonld + microdata + rdfa counts exceed any_structured_data. RDFa n=1 is at the noise floor of this sample size and is not treated as a standalone finding."
  },
  "schema_org_type_prevalence": {
    "denominator": 253,
    "note": "Percentage and count of all 253 reachable domains carrying at least one instance of this schema.org type, regardless of format. A domain may carry more than one type. Types with n<10 (LocalBusiness, FAQPage) are direct, non-crossed counts from the live scan, flagged here as too small to treat as a precise measurement.",
    "counts": {
      "WebSite": 73,
      "Organization": 67,
      "WebPage": 55,
      "BreadcrumbList": 38,
      "ImageObject": 23,
      "Person": 15,
      "Article": 14,
      "LocalBusiness": 6,
      "FAQPage": 3
    },
    "pct": {
      "WebSite": 28.9,
      "Organization": 26.5,
      "WebPage": 21.7,
      "BreadcrumbList": 15.0,
      "ImageObject": 9.1,
      "Person": 5.9,
      "Article": 5.5,
      "LocalBusiness": 2.4,
      "FAQPage": 1.2
    }
  },
  "sitemap_conventional_path": {
    "denominator": 258,
    "present": 181,
    "present_pct": 70.2,
    "absent": 77,
    "absent_pct": 29.8,
    "valid_xml_among_present": {
      "denominator": 181,
      "valid": 136,
      "valid_pct": 75.1,
      "not_valid": 45,
      "not_valid_pct": 24.9
    },
    "empty_but_valid_xml_among_present": {
      "denominator": 181,
      "n": 3,
      "note": "Valid XML, zero <loc> entries: formally functional, gives a crawler no page to find. n=3 is too small to publish as a percentage."
    }
  },
  "sitemap_declared_in_robots": {
    "denominator": 299,
    "counts": {
      "declared_matches_conventional_path": 76,
      "declared_at_nonconventional_path_not_fetched": 77,
      "no_sitemap_declared": 83,
      "no_robots_or_no_declaration": 63
    },
    "pct": {
      "declared_matches_conventional_path": 25.4,
      "declared_at_nonconventional_path_not_fetched": 25.8,
      "no_sitemap_declared": 27.8,
      "no_robots_or_no_declaration": 21.1
    }
  },
  "related_study": {
    "title": "AI crawlery na ceskych firemnich webech (robots.txt policy)",
    "url": "https://ciad.cz/studie/ai-crawlery-ceske-weby/",
    "relationship": "That study measures whether AI crawlers are permitted to access a site at all (robots.txt policy, n=260 reachable). This study measures a separate question on a separately-scanned reachability set: once access is permitted, is there anything machine-readable to find. Denominators are not interchangeable between the two studies."
  },
  "publication_threshold": "No category with fewer than 10 domains is published as a crossed statistic. Direct, non-crossed counts below 10 are published with an explicit small-sample caveat rather than omitted, where they are directly reported by the live-scan aggregate."
}