aeorank 1.6.0 → 2.1.0

package/dist/index.cjs

@@ -303,7 +303,7 @@ var init_full_site_crawler = __esm({
     RESOURCE_EXTENSIONS = /\.(js|css|png|jpg|jpeg|gif|svg|ico|pdf|xml|txt|woff|woff2|ttf|eot|mp4|mp3|webp|avif|zip|gz|tar|json)$/i;
     SKIP_PATH_PATTERNS = /^\/(api|wp-admin|wp-json|static|assets|_next|auth|login|signup|cart|checkout|admin|feed|xmlrpc)\b/i;
     CATEGORY_PATTERNS = [
-      [/\/(blog|articles?|posts?|news|insights|guides)\b/i, "blog"],
+      [/\/([^/]*-?)?(blog|articles?|posts?|news|insights|guides)\b/i, "blog"],
       [/\/(about|about-us|company|who-we-are)\b/i, "about"],
       [/\/(pricing|plans|packages)\b/i, "pricing"],
       [/\/(services?|features?|solutions?|products?|what-we-do|offerings?)\b/i, "services"],
@@ -552,7 +552,7 @@ async function prefetchSiteData(domain) {
         sitemapForBlog = subSitemap.text;
       }
     }
-    const blogUrls = extractBlogUrlsFromSitemap(sitemapForBlog, domain, 10);
+    const blogUrls = extractBlogUrlsFromSitemap(sitemapForBlog, domain, 50);
     if (blogUrls.length > 0) {
       const fetched = await Promise.all(blogUrls.map((url) => fetchText(url)));
       blogSample = fetched.filter(
@@ -909,15 +909,17 @@ function checkOriginalData(data) {
     findings.push({ severity: "critical", detail: "Could not fetch homepage" });
     return { criterion: "original_data", criterion_label: "Original Data & Expert Content", score: 0, status: "not_found", findings, fix_priority: "P2" };
   }
+  const allPages = [data.homepage, ...data.blogSample || []].filter(Boolean);
   const html = data.homepage.text;
-  const text = html.replace(/<[^>]*>/g, " ").replace(/\s+/g, " ");
+  const allText = allPages.map((p) => p.text.replace(/<[^>]*>/g, " ").replace(/\s+/g, " ")).join(" ");
+  const text = data.homepage.text.replace(/<[^>]*>/g, " ").replace(/\s+/g, " ");
   let score = 0;
   const statPatterns = /\d+%|\d+\s*(patients|clients|customers|cases|years|professionals|specialists|companies|users|businesses|domains|audits)/i;
-  if (statPatterns.test(text)) {
+  if (statPatterns.test(allText)) {
     const researchContext = /\b(our\s+(?:study|analysis|research|data|survey|findings|report)|we\s+(?:surveyed|analyzed|studied|measured|tracked)|proprietary|methodology|original\s+research)\b/i;
-    if (researchContext.test(text)) {
+    if (researchContext.test(allText)) {
       score += 3;
-      findings.push({ severity: "info", detail: "Proprietary statistics with research context found on homepage" });
+      findings.push({ severity: "info", detail: "Proprietary statistics with research context found" });
     } else {
       score += 1;
       findings.push({ severity: "low", detail: 'Statistics found but without research context (e.g., "500+ clients")', fix: 'Add context about your methodology: "Our analysis of X found..." or "We surveyed Y..."' });
@@ -1432,20 +1434,24 @@ function checkFactDensity(data) {
     findings.push({ severity: "critical", detail: "Could not fetch homepage" });
     return { criterion: "fact_density", criterion_label: "Fact & Data Density", score: 0, status: "not_found", findings, fix_priority: "P2" };
   }
-  const text = data.homepage.text.replace(/<[^>]*>/g, " ").replace(/\s+/g, " ");
+  const allPages = [data.homepage, ...data.blogSample || []].filter(Boolean);
+  const allText = allPages.map((p) => p.text.replace(/<[^>]*>/g, " ").replace(/\s+/g, " ")).join(" ");
+  const text = allText;
+  const pageCount = allPages.length;
   let score = 0;
   const dataPoints = text.match(/\d+(?:\.\d+)?(?:\s*%|\s*\$|\s*USD|\s*EUR)/g) || [];
   const countPhrases = text.match(/\d+(?:,\d{3})*\+?\s+(?:users?|clients?|customers?|companies|businesses|patients?|members?|employees?|projects?|downloads?)/gi) || [];
   const totalDataPoints = dataPoints.length + countPhrases.length;
-  if (totalDataPoints >= 6) {
+  const avgPerPage = pageCount > 0 ? totalDataPoints / pageCount : 0;
+  if (avgPerPage >= 4) {
     score += 5;
-    findings.push({ severity: "info", detail: `${totalDataPoints} quantitative data points found on homepage` });
-  } else if (totalDataPoints >= 3) {
+    findings.push({ severity: "info", detail: `${totalDataPoints} quantitative data points found across ${pageCount} pages (avg ${avgPerPage.toFixed(1)}/page)` });
+  } else if (avgPerPage >= 2) {
     score += 3;
-    findings.push({ severity: "info", detail: `${totalDataPoints} quantitative data points found` });
+    findings.push({ severity: "info", detail: `${totalDataPoints} quantitative data points found across ${pageCount} pages` });
   } else if (totalDataPoints >= 1) {
     score += 1;
-    findings.push({ severity: "low", detail: `Only ${totalDataPoints} quantitative data point(s) found`, fix: "Add more specific numbers, percentages, and metrics to strengthen credibility" });
+    findings.push({ severity: "low", detail: `Only ${totalDataPoints} quantitative data point(s) found across ${pageCount} pages`, fix: "Add more specific numbers, percentages, and metrics to strengthen credibility" });
   } else {
     findings.push({ severity: "high", detail: "No quantitative data points found", fix: "Add specific statistics (percentages, counts, comparisons) that AI engines can cite" });
   }
@@ -1551,9 +1557,9 @@ function countRecentSitemapDates(sitemapText) {
     distinctRecentDays: recentDays.size
   };
 }
-var BLOG_PATH_PATTERNS = /\/(?:blog|articles?|insights?|guides?|resources?|news|posts?|learn|help|how-?to|tutorials?|case-stud|whitepapers?)\b/i;
+var BLOG_PATH_PATTERNS = /\/(?:[^/]*-?)?(?:blog|articles?|insights?|guides?|resources?|news|posts?|learn|help|how-?to|tutorials?|case-stud|whitepapers?)\b/i;
 var EXCLUDE_PATH_PATTERNS = /\/(?:tag|category|author|page|feed|wp-content|wp-admin|wp-json|cart|checkout|login|search|api|static|assets|_next)\b/i;
-function extractBlogUrlsFromSitemap(sitemapText, domain, limit = 5) {
+function extractBlogUrlsFromSitemap(sitemapText, domain, limit = 50) {
   const urlBlocks = sitemapText.match(/<url>([\s\S]*?)<\/url>/gi) || [];
   const candidates = [];
   const cleanDomain = domain.replace(/^www\./, "").toLowerCase();
@@ -1849,7 +1855,7 @@ function jaccardSimilarity(a, b) {
   const union = a.size + b.size - intersection;
   return union === 0 ? 0 : intersection / union;
 }
-function checkContentCannibalization(data) {
+function checkContentCannibalization(data, topicCoherenceScore) {
   const findings = [];
   if (!data.homepage) {
     findings.push({ severity: "critical", detail: "No homepage available for cannibalization analysis" });
@@ -1859,7 +1865,7 @@ function checkContentCannibalization(data) {
     { html: data.homepage.text, url: data.homepage.finalUrl || `https://${data.domain}/` }
   ];
   if (data.blogSample) {
-    for (const page of data.blogSample.slice(0, 5)) {
+    for (const page of data.blogSample) {
       pages.push({ html: page.text, url: page.finalUrl || "" });
     }
   }
@@ -1869,10 +1875,29 @@ function checkContentCannibalization(data) {
   }
   const pageTitles = pages.map((p) => ({ title: extractPageTitle(p.html), url: p.url }));
   const wordSets = pageTitles.map((p) => titleToWordSet(p.title));
+  const termPageCount = /* @__PURE__ */ new Map();
+  for (const ws of wordSets) {
+    for (const w of ws) {
+      termPageCount.set(w, (termPageCount.get(w) || 0) + 1);
+    }
+  }
+  const commonTermThreshold = Math.max(3, pages.length * 0.4);
+  const siteThemeTerms = /* @__PURE__ */ new Set();
+  for (const [term, count] of termPageCount) {
+    if (count >= commonTermThreshold) siteThemeTerms.add(term);
+  }
+  const filteredSets = wordSets.map((ws) => {
+    const filtered = /* @__PURE__ */ new Set();
+    for (const w of ws) {
+      if (!siteThemeTerms.has(w)) filtered.add(w);
+    }
+    return filtered;
+  });
   const cannibalPairs = [];
   for (let i = 0; i < pages.length; i++) {
     for (let j = i + 1; j < pages.length; j++) {
-      const sim = jaccardSimilarity(wordSets[i], wordSets[j]);
+      if (filteredSets[i].size === 0 && filteredSets[j].size === 0) continue;
+      const sim = jaccardSimilarity(filteredSets[i], filteredSets[j]);
       if (sim > 0.6) {
         cannibalPairs.push({
           urlA: pageTitles[i].url.slice(0, 60),
@@ -1882,23 +1907,39 @@ function checkContentCannibalization(data) {
       }
     }
   }
+  const cannibalUrls = /* @__PURE__ */ new Set();
+  for (const pair of cannibalPairs) {
+    cannibalUrls.add(pair.urlA);
+    cannibalUrls.add(pair.urlB);
+  }
+  const cannibalRatio = pages.length > 0 ? cannibalUrls.size / pages.length : 0;
   let score;
   if (cannibalPairs.length === 0) {
     score = 10;
     findings.push({ severity: "info", detail: `${pages.length} pages analyzed - no content cannibalization detected` });
-  } else if (cannibalPairs.length === 1) {
-    score = 8;
-    findings.push({ severity: "low", detail: `1 pair of pages with overlapping topics (${cannibalPairs[0].similarity}% similarity)`, fix: "Differentiate titles and H1 headings to reduce topic overlap" });
-  } else if (cannibalPairs.length === 2) {
+  } else if (cannibalRatio <= 0.05) {
+    score = 9;
+    findings.push({ severity: "info", detail: `${cannibalPairs.length} pair(s) of pages with minor topic overlap (${cannibalUrls.size}/${pages.length} pages affected)` });
+  } else if (cannibalRatio <= 0.1) {
+    score = 7;
+    findings.push({ severity: "low", detail: `${cannibalUrls.size} pages (${Math.round(cannibalRatio * 100)}%) have overlapping topics`, fix: "Differentiate titles and H1 headings to reduce topic overlap" });
+  } else if (cannibalRatio <= 0.2) {
     score = 5;
-    findings.push({ severity: "medium", detail: `${cannibalPairs.length} pairs of pages with overlapping topics`, fix: "Consolidate overlapping pages or differentiate their titles and content focus" });
+    findings.push({ severity: "medium", detail: `${cannibalUrls.size} pages (${Math.round(cannibalRatio * 100)}%) competing for overlapping topics`, fix: "Consolidate overlapping pages or differentiate their titles and content focus" });
+  } else if (cannibalRatio <= 0.4) {
+    score = 3;
+    findings.push({ severity: "medium", detail: `${cannibalUrls.size} pages (${Math.round(cannibalRatio * 100)}%) have significant content overlap`, fix: "Many pages compete for the same topics - consolidate or clearly differentiate them" });
   } else {
     score = 0;
-    findings.push({ severity: "high", detail: `${cannibalPairs.length} pairs of pages competing for the same topics`, fix: "Significant content overlap detected - consolidate or clearly differentiate competing pages" });
+    findings.push({ severity: "high", detail: `${cannibalUrls.size} pages (${Math.round(cannibalRatio * 100)}%) competing for the same topics`, fix: "Severe content cannibalization - consolidate overlapping pages or create clear topic differentiation" });
   }
   for (const pair of cannibalPairs.slice(0, 3)) {
     findings.push({ severity: "low", detail: `Overlap (${pair.similarity}%): ${pair.urlA} vs ${pair.urlB}` });
   }
+  if (topicCoherenceScore !== void 0 && topicCoherenceScore <= 4 && score >= 8) {
+    score = 6;
+    findings.push({ severity: "low", detail: "Low topic overlap but content lacks coherent focus - not a strong signal for AI authority", fix: "Focus content on fewer core topics to build topical authority that AI engines can identify" });
+  }
   return { criterion: "content_cannibalization", criterion_label: "Content Cannibalization", score, status: score >= 7 ? "pass" : score >= 4 ? "partial" : "fail", findings, fix_priority: score >= 7 ? "P3" : "P1" };
 }
 function checkVisibleDateSignal(data) {
@@ -2124,7 +2165,233 @@ function extractRawDataSummary(data) {
     crawl_skipped: data.crawlStats?.skipped ?? 0
   };
 }
+function getPageTopicText(html) {
+  const titleMatch = html.match(/<title[^>]*>([^<]+)<\/title>/i);
+  const h1Match = html.match(/<h1[^>]*>([\s\S]*?)<\/h1>/i);
+  return [
+    titleMatch?.[1] || "",
+    h1Match?.[1]?.replace(/<[^>]*>/g, "") || ""
+  ].join(" ").toLowerCase().trim();
+}
+function extractBigrams(text) {
+  const words = text.split(/[\s,.!?;:()\[\]{}"'\/&]+/).filter((w) => w.length > 2 && !STOP_WORDS.has(w) && !/^\d+$/.test(w));
+  const bigrams = [];
+  for (let i = 0; i < words.length - 1; i++) {
+    bigrams.push(words[i] + " " + words[i + 1]);
+  }
+  return bigrams;
+}
+function checkTopicCoherence(data) {
+  const findings = [];
+  if (!data.homepage) {
+    findings.push({ severity: "critical", detail: "Could not fetch homepage" });
+    return { criterion: "topic_coherence", criterion_label: "Topic Coherence", score: 0, status: "not_found", findings, fix_priority: "P0" };
+  }
+  if (!data.blogSample || data.blogSample.length < 3) {
+    findings.push({ severity: "info", detail: `Only ${data.blogSample?.length || 0} blog pages found - insufficient for topic coherence analysis` });
+    return { criterion: "topic_coherence", criterion_label: "Topic Coherence", score: 5, status: "partial", findings, fix_priority: "P2" };
+  }
+  const blogPages = data.blogSample;
+  const domainBase = data.domain.replace(/^www\./, "").replace(/\.(com|org|net|io|co|ai)$/i, "").toLowerCase();
+  const brandWords = /* @__PURE__ */ new Set();
+  brandWords.add(domainBase);
+  for (const part of domainBase.split(/[-_]/)) {
+    if (part.length > 2) brandWords.add(part);
+  }
+  const rawTermFreq = /* @__PURE__ */ new Map();
+  const pageTitleTexts = [];
+  for (const page of blogPages) {
+    const topicText = getPageTopicText(page.text);
+    pageTitleTexts.push(topicText);
+    const words = topicText.split(/[\s,.!?;:()\[\]{}"'\/&]+/).filter((w) => w.length > 2 && !STOP_WORDS.has(w) && !/^\d+$/.test(w));
+    const uniqueWords = new Set(words);
+    for (const w of uniqueWords) {
+      rawTermFreq.set(w, (rawTermFreq.get(w) || 0) + 1);
+    }
+  }
+  for (const [term, count] of rawTermFreq) {
+    if (count / blogPages.length >= 0.8 && domainBase.includes(term)) {
+      brandWords.add(term);
+    }
+  }
+  const termFreq = /* @__PURE__ */ new Map();
+  for (const page of blogPages) {
+    const topicText = getPageTopicText(page.text);
+    const words = topicText.split(/[\s,.!?;:()\[\]{}"'\/&]+/).filter((w) => w.length > 2 && !STOP_WORDS.has(w) && !/^\d+$/.test(w) && !brandWords.has(w));
+    const uniqueWords = new Set(words);
+    for (const w of uniqueWords) {
+      termFreq.set(w, (termFreq.get(w) || 0) + 1);
+    }
+  }
+  const sortedTerms = [...termFreq.entries()].sort((a, b) => b[1] - a[1]);
+  const topTerm = sortedTerms[0];
+  const bigramFreq = /* @__PURE__ */ new Map();
+  const pageBigrams = [];
+  for (const topicText of pageTitleTexts) {
+    const bigrams = extractBigrams(topicText).filter((bg) => !bg.split(" ").some((w) => brandWords.has(w)));
+    pageBigrams.push(bigrams);
+    const uniqueBigrams = new Set(bigrams);
+    for (const bg of uniqueBigrams) {
+      bigramFreq.set(bg, (bigramFreq.get(bg) || 0) + 1);
+    }
+  }
+  const sortedBigrams = [...bigramFreq.entries()].sort((a, b) => b[1] - a[1]);
+  const topBigram = sortedBigrams[0];
+  const significantBigrams = sortedBigrams.filter(([, count]) => count >= 2);
+  const clusterRoots = [];
+  const assigned = /* @__PURE__ */ new Set();
+  for (const [bg] of significantBigrams) {
+    if (assigned.has(bg)) continue;
+    clusterRoots.push(bg);
+    assigned.add(bg);
+    const [w1, w2] = bg.split(" ");
+    for (const [otherBg] of significantBigrams) {
+      if (assigned.has(otherBg)) continue;
+      if (otherBg.includes(w1) || otherBg.includes(w2)) {
+        assigned.add(otherBg);
+      }
+    }
+  }
+  const topicClusterCount = clusterRoots.length;
+  const dominantTerm = topTerm?.[0] || "";
+  const dominantTermCount = topTerm?.[1] || 0;
+  const focusRatio = blogPages.length > 0 ? dominantTermCount / blogPages.length : 0;
+  const dominantBigram = topBigram?.[0] || "";
+  const dominantBigramCount = topBigram?.[1] || 0;
+  const bigramFocusRatio = blogPages.length > 0 ? dominantBigramCount / blogPages.length : 0;
+  let score = 0;
+  const bestFocusRatio = Math.max(focusRatio, bigramFocusRatio);
+  if (bestFocusRatio >= 0.8) {
+    score += 7;
+  } else if (bestFocusRatio >= 0.6) {
+    score += 6;
+  } else if (bestFocusRatio >= 0.45) {
+    score += 5;
+  } else if (bestFocusRatio >= 0.3) {
+    score += 3;
+  } else if (bestFocusRatio >= 0.15) {
+    score += 2;
+  } else {
+    score += 1;
+  }
+  const clusterPenaltyReduced = focusRatio >= 0.7;
+  if (topicClusterCount <= 3) {
+    score += 3;
+    findings.push({ severity: "info", detail: `${topicClusterCount} topic cluster(s) - tightly focused content` });
+  } else if (topicClusterCount <= 6) {
+    score += clusterPenaltyReduced ? 2 : 1;
+    findings.push({ severity: "info", detail: `${topicClusterCount} topic clusters${clusterPenaltyReduced ? " within a focused niche" : " - moderately focused"}` });
+  } else if (topicClusterCount <= 10) {
+    score += clusterPenaltyReduced ? 1 : 0;
+    if (!clusterPenaltyReduced) {
+      findings.push({ severity: "low", detail: `${topicClusterCount} topic clusters - scattered content`, fix: "Reduce the number of distinct topics. Focus blog content on 2-3 core expertise areas." });
+    } else {
+      findings.push({ severity: "info", detail: `${topicClusterCount} topic clusters but strong core topic focus (${Math.round(focusRatio * 100)}%)` });
+    }
+  } else {
+    score += clusterPenaltyReduced ? 0 : -2;
+    if (!clusterPenaltyReduced) {
+      findings.push({ severity: "medium", detail: `${topicClusterCount} topic clusters - highly scattered content`, fix: "Content covers too many unrelated topics. AI engines cannot identify your expertise. Focus on your core niche." });
+    } else {
+      findings.push({ severity: "low", detail: `${topicClusterCount} topic clusters despite strong core topic focus`, fix: "Consider narrowing subtopics within your niche for even stronger AI visibility." });
+    }
+  }
+  score = Math.max(0, Math.min(10, score));
+  if (dominantTerm) {
+    const focusPct = Math.round(focusRatio * 100);
+    findings.push({ severity: "info", detail: `Dominant topic term: "${dominantTerm}" (${focusPct}% of ${blogPages.length} pages)` });
+  }
+  if (dominantBigram && dominantBigramCount >= 2) {
+    findings.push({ severity: "info", detail: `Dominant topic phrase: "${dominantBigram}" (${dominantBigramCount}/${blogPages.length} pages)` });
+  }
+  const offTopicExamples = [];
+  for (let i = 0; i < pageTitleTexts.length && offTopicExamples.length < 3; i++) {
+    if (dominantTerm && !pageTitleTexts[i].includes(dominantTerm)) {
+      const title = blogPages[i].text.match(/<title[^>]*>([^<]+)<\/title>/i)?.[1]?.trim();
+      if (title && title.length > 3) offTopicExamples.push(title.slice(0, 60));
+    }
+  }
+  if (offTopicExamples.length > 0 && score < 8) {
+    findings.push({ severity: "low", detail: `Off-topic examples: ${offTopicExamples.join("; ")}` });
+  }
+  return { criterion: "topic_coherence", criterion_label: "Topic Coherence", score, status: score >= 7 ? "pass" : score >= 4 ? "partial" : "fail", findings, fix_priority: score >= 7 ? "P3" : "P0" };
+}
+function countWords(html) {
+  const text = html.replace(/<script[^>]*>[\s\S]*?<\/script>/gi, "").replace(/<style[^>]*>[\s\S]*?<\/style>/gi, "").replace(/<[^>]*>/g, " ").replace(/\s+/g, " ").trim();
+  return text.split(/\s+/).filter((w) => w.length > 0).length;
+}
+function countHeadings(html) {
+  const headings = html.match(/<h[2-6][^>]*>/gi) || [];
+  return headings.length;
+}
+function checkContentDepth(data, topicCoherenceScore) {
+  const findings = [];
+  if (!data.blogSample || data.blogSample.length < 2) {
+    findings.push({ severity: "info", detail: `Only ${data.blogSample?.length || 0} blog pages found - insufficient for depth analysis` });
+    return { criterion: "content_depth", criterion_label: "Content Depth", score: 3, status: "partial", findings, fix_priority: "P2" };
+  }
+  const blogPages = data.blogSample;
+  const wordCounts = blogPages.map((p) => countWords(p.text));
+  const headingCounts = blogPages.map((p) => countHeadings(p.text));
+  const avgWords = wordCounts.reduce((a, b) => a + b, 0) / wordCounts.length;
+  const avgHeadings = headingCounts.reduce((a, b) => a + b, 0) / headingCounts.length;
+  const deepPages = wordCounts.filter((w) => w >= 1e3).length;
+  const thinPages = wordCounts.filter((w) => w < 300).length;
+  const deepRatio = deepPages / blogPages.length;
+  const thinRatio = thinPages / blogPages.length;
+  let score = 0;
+  if (avgWords >= 2e3) {
+    score += 5;
+    findings.push({ severity: "info", detail: `Average ${Math.round(avgWords)} words per page across ${blogPages.length} pages - excellent depth` });
+  } else if (avgWords >= 1200) {
+    score += 4;
+    findings.push({ severity: "info", detail: `Average ${Math.round(avgWords)} words per page across ${blogPages.length} pages - good depth` });
+  } else if (avgWords >= 800) {
+    score += 3;
+    findings.push({ severity: "info", detail: `Average ${Math.round(avgWords)} words per page - moderate depth` });
+  } else if (avgWords >= 400) {
+    score += 2;
+    findings.push({ severity: "low", detail: `Average ${Math.round(avgWords)} words per page - shallow content`, fix: "Expand articles with more detail, examples, and expert analysis to build AI citation authority" });
+  } else {
+    score += 1;
+    findings.push({ severity: "medium", detail: `Average ${Math.round(avgWords)} words per page - very thin content`, fix: "Content is too thin for AI engines to cite. Aim for 1000+ words per article with structured sections." });
+  }
+  if (avgHeadings >= 8) {
+    score += 3;
+    findings.push({ severity: "info", detail: `Average ${avgHeadings.toFixed(1)} subheadings per page - well-structured` });
+  } else if (avgHeadings >= 5) {
+    score += 2;
+    findings.push({ severity: "info", detail: `Average ${avgHeadings.toFixed(1)} subheadings per page - decent structure` });
+  } else if (avgHeadings >= 2) {
+    score += 1;
+    findings.push({ severity: "low", detail: `Average ${avgHeadings.toFixed(1)} subheadings per page`, fix: "Add more H2/H3 headings to break content into extractable sections" });
+  } else {
+    findings.push({ severity: "medium", detail: `Average ${avgHeadings.toFixed(1)} subheadings per page - minimal structure`, fix: "Add question-format H2/H3 headings so AI engines can extract specific answers" });
+  }
+  if (deepRatio >= 0.5) {
+    score += 2;
+    findings.push({ severity: "info", detail: `${deepPages}/${blogPages.length} pages (${Math.round(deepRatio * 100)}%) have 1000+ words` });
+  } else if (deepRatio >= 0.25) {
+    score += 1;
+    findings.push({ severity: "info", detail: `${deepPages}/${blogPages.length} pages have 1000+ words` });
+  }
+  if (thinRatio >= 0.5) {
+    score = Math.max(0, score - 2);
+    findings.push({ severity: "medium", detail: `${thinPages}/${blogPages.length} pages (${Math.round(thinRatio * 100)}%) have under 300 words - high thin content ratio`, fix: "Remove or expand thin pages. Thin content dilutes site quality for AI engines." });
+  } else if (thinRatio >= 0.25) {
+    score = Math.max(0, score - 1);
+    findings.push({ severity: "low", detail: `${thinPages}/${blogPages.length} pages have under 300 words` });
+  }
+  let finalScore = Math.min(10, score);
+  if (topicCoherenceScore !== void 0 && topicCoherenceScore <= 4 && finalScore >= 8) {
+    finalScore = 7;
+    findings.push({ severity: "low", detail: "Deep content but low topic coherence - depth on scattered topics has reduced AI citation value", fix: "Focus content depth on your core expertise area for maximum AI visibility" });
+  }
+  return { criterion: "content_depth", criterion_label: "Content Depth", score: finalScore, status: finalScore >= 7 ? "pass" : finalScore >= 4 ? "partial" : "fail", findings, fix_priority: finalScore >= 7 ? "P3" : "P1" };
+}
 function auditSiteFromData(data) {
+  const topicCoherence = checkTopicCoherence(data);
+  const cannibalization = checkContentCannibalization(data, topicCoherence.score);
   return [
     checkLlmsTxt(data),
     checkSchemaMarkup(data),
@@ -2150,52 +2417,84 @@ function auditSiteFromData(data) {
     checkSchemaCoverage(data),
     checkSpeakableSchema(data),
     checkQueryAnswerAlignment(data),
-    checkContentCannibalization(data),
-    checkVisibleDateSignal(data)
+    cannibalization,
+    checkVisibleDateSignal(data),
+    topicCoherence,
+    checkContentDepth(data, topicCoherence.score)
   ];
 }
 
 // src/scoring.ts
 var WEIGHTS = {
-  // Original 10
-  llms_txt: 0.1,
-  schema_markup: 0.15,
-  qa_content_format: 0.15,
-  clean_html: 0.1,
-  entity_consistency: 0.1,
-  robots_txt: 0.05,
-  faq_section: 0.1,
+  // ─── Content Substance (~55%) ─────────────────────────────────────────────
+  // WHY an AI engine would cite you. These drive citation quality directly.
+  topic_coherence: 0.14,
+  // Topical authority - THE gating signal
   original_data: 0.1,
-  internal_linking: 0.1,
-  semantic_html: 0.05,
-  // New 12
-  content_freshness: 0.07,
-  sitemap_completeness: 0.05,
-  rss_feed: 0.03,
-  table_list_extractability: 0.07,
-  definition_patterns: 0.04,
-  direct_answer_density: 0.07,
-  content_licensing: 0.04,
-  author_schema_depth: 0.04,
-  fact_density: 0.05,
-  canonical_url: 0.04,
-  content_velocity: 0.03,
-  schema_coverage: 0.03,
-  speakable_schema: 0.03,
-  query_answer_alignment: 0.08,
-  content_cannibalization: 0.05,
-  visible_date_signal: 0.04
+  // Unique value AI can't find elsewhere
+  content_depth: 0.07,
+  // Comprehensive vs thin coverage
+  fact_density: 0.06,
+  // Information density per page
+  direct_answer_density: 0.05,
+  // Direct answers to queries
+  qa_content_format: 0.05,
+  // Answer-shaped content structure
+  query_answer_alignment: 0.05,
+  // Relevance to actual AI queries
+  faq_section: 0.04,
+  // Structured Q&A pairs
+  // ─── Content Organization (~30%) ──────────────────────────────────────────
+  // HOW easily AI engines can extract and trust your content.
+  entity_consistency: 0.05,
+  // Brand authority and E-E-A-T
+  internal_linking: 0.04,
+  // Site structure and topic clusters
+  content_freshness: 0.04,
+  // Recency signals
+  schema_markup: 0.03,
+  // Structured data for discovery
+  author_schema_depth: 0.03,
+  // Expert attribution
+  table_list_extractability: 0.03,
+  // Extractable structured data
+  definition_patterns: 0.02,
+  // Clear definitions
+  visible_date_signal: 0.02,
+  // Publication date trust
+  semantic_html: 0.02,
+  // Clean semantic structure
+  clean_html: 0.02,
+  // Parseable markup
+  // ─── Technical Plumbing (~15%) ────────────────────────────────────────────
+  // WHETHER AI crawlers can find you. Table stakes with diminishing returns.
+  content_cannibalization: 0.02,
+  llms_txt: 0.02,
+  robots_txt: 0.02,
+  content_velocity: 0.02,
+  content_licensing: 0.02,
+  sitemap_completeness: 0.01,
+  canonical_url: 0.01,
+  rss_feed: 0.01,
+  schema_coverage: 0.01,
+  speakable_schema: 0.01
 };
 function calculateOverallScore(criteria) {
   let totalWeight = 0;
   let weightedSum = 0;
   for (const c of criteria) {
-    const weight = WEIGHTS[c.criterion] ?? 0.1;
+    const weight = WEIGHTS[c.criterion] ?? 0.05;
     weightedSum += c.score / 10 * weight * 100;
     totalWeight += weight;
   }
   if (totalWeight === 0) return 0;
-  return Math.round(weightedSum / totalWeight);
+  let score = Math.round(weightedSum / totalWeight);
+  const coherence = criteria.find((c) => c.criterion === "topic_coherence");
+  if (coherence && coherence.score < 6) {
+    const cap2 = 35 + coherence.score * 5;
+    score = Math.min(score, cap2);
+  }
+  return score;
 }
 
 // src/headless-fetch.ts
@@ -2326,7 +2625,9 @@ var CRITERION_LABELS = {
   "Speakable Schema": "Speakable Schema",
   "Query-Answer Alignment": "Query-Answer Alignment",
   "Content Cannibalization": "Content Cannibalization",
-  "Visible Date Signal": "Visible Date Signal"
+  "Visible Date Signal": "Visible Date Signal",
+  "Topic Coherence": "Topic Coherence",
+  "Content Depth": "Content Depth"
 };
 function scoreToStatus(score) {
   if (score === 0) return "MISSING";
@@ -2412,32 +2713,37 @@ function buildDetailedFindings(results) {
 
 // src/narrative-generator.ts
 var CRITERION_WEIGHTS = {
-  llms_txt: 0.1,
-  schema_markup: 0.15,
-  qa_content_format: 0.15,
-  clean_html: 0.1,
-  entity_consistency: 0.1,
-  robots_txt: 0.05,
-  faq_section: 0.1,
+  // Content Substance (~55%)
+  topic_coherence: 0.14,
   original_data: 0.1,
-  internal_linking: 0.1,
-  semantic_html: 0.05,
-  content_freshness: 0.07,
-  sitemap_completeness: 0.05,
-  rss_feed: 0.03,
-  table_list_extractability: 0.07,
-  definition_patterns: 0.04,
-  direct_answer_density: 0.07,
-  content_licensing: 0.04,
-  author_schema_depth: 0.04,
-  fact_density: 0.05,
-  canonical_url: 0.04,
-  content_velocity: 0.03,
-  schema_coverage: 0.03,
-  speakable_schema: 0.03,
-  query_answer_alignment: 0.08,
-  content_cannibalization: 0.05,
-  visible_date_signal: 0.04
+  content_depth: 0.07,
+  fact_density: 0.06,
+  direct_answer_density: 0.05,
+  qa_content_format: 0.05,
+  query_answer_alignment: 0.05,
+  faq_section: 0.04,
+  // Content Organization (~30%)
+  entity_consistency: 0.05,
+  internal_linking: 0.04,
+  content_freshness: 0.04,
+  schema_markup: 0.03,
+  author_schema_depth: 0.03,
+  table_list_extractability: 0.03,
+  definition_patterns: 0.02,
+  visible_date_signal: 0.02,
+  semantic_html: 0.02,
+  clean_html: 0.02,
+  // Technical Plumbing (~15%)
+  content_cannibalization: 0.02,
+  llms_txt: 0.02,
+  robots_txt: 0.02,
+  content_velocity: 0.02,
+  content_licensing: 0.02,
+  sitemap_completeness: 0.01,
+  canonical_url: 0.01,
+  rss_feed: 0.01,
+  schema_coverage: 0.01,
+  speakable_schema: 0.01
 };
 var OPPORTUNITY_TEMPLATES = {
   llms_txt: {
@@ -2569,6 +2875,16 @@ var OPPORTUNITY_TEMPLATES = {
     name: "Add Visible Date Signals",
     effort: "Low",
     description: "Display publication/modification dates visibly using <time> elements and add datePublished/dateModified to JSON-LD schema."
+  },
+  topic_coherence: {
+    name: "Focus Content on Core Topics",
+    effort: "High",
+    description: 'Ensure blog content consistently covers your core expertise areas rather than scattering across unrelated topics. AI engines build authority models - a site about "Medicare coverage" that also publishes about humidifiers and groceries dilutes its topical authority.'
+  },
+  content_depth: {
+    name: "Increase Content Depth",
+    effort: "Medium",
+    description: "Expand articles to 1000+ words with structured H2/H3 sections, comparison tables, and expert analysis. Thin content (under 300 words) is rarely cited by AI engines. Deep, well-structured articles demonstrate expertise."
   }
 };
 function calculateImpact(score, weight, effort) {
@@ -2690,7 +3006,7 @@ function generatePitchNumbers(score, rawData, scorecard) {
   const passing = scorecard.filter((s) => s.score >= 7).length;
   metrics.push({
     metric: "Criteria Passing",
-    value: `${passing}/26`,
+    value: `${passing}/28`,
     significance: passing >= 18 ? "Excellent coverage across AEO dimensions" : passing >= 12 ? "Good foundation with room to improve remaining criteria" : `${26 - passing} criteria need attention for full AI visibility`
   });
   return metrics;
@@ -2882,20 +3198,23 @@ async function fetchMultiPageData(siteData, options) {
 
 // src/page-scorer.ts
 var PAGE_CRITERIA = {
-  schema_markup: { weight: 0.15, label: "Schema.org Structured Data" },
-  qa_content_format: { weight: 0.15, label: "Q&A Content Format" },
-  clean_html: { weight: 0.1, label: "Clean, Crawlable HTML" },
-  faq_section: { weight: 0.1, label: "FAQ Section Content" },
+  // Content Substance
   original_data: { weight: 0.1, label: "Original Data & Expert Content" },
-  query_answer_alignment: { weight: 0.08, label: "Query-Answer Alignment" },
-  content_freshness: { weight: 0.07, label: "Content Freshness Signals" },
-  table_list_extractability: { weight: 0.07, label: "Table & List Extractability" },
-  direct_answer_density: { weight: 0.07, label: "Direct Answer Paragraphs" },
-  semantic_html: { weight: 0.05, label: "Semantic HTML5 & Accessibility" },
-  fact_density: { weight: 0.05, label: "Fact & Data Density" },
-  definition_patterns: { weight: 0.04, label: "Definition Patterns" },
-  canonical_url: { weight: 0.04, label: "Canonical URL Strategy" },
-  visible_date_signal: { weight: 0.04, label: "Visible Date Signal" }
+  fact_density: { weight: 0.06, label: "Fact & Data Density" },
+  direct_answer_density: { weight: 0.05, label: "Direct Answer Paragraphs" },
+  qa_content_format: { weight: 0.05, label: "Q&A Content Format" },
+  query_answer_alignment: { weight: 0.05, label: "Query-Answer Alignment" },
+  faq_section: { weight: 0.04, label: "FAQ Section Content" },
+  // Content Organization
+  content_freshness: { weight: 0.04, label: "Content Freshness Signals" },
+  schema_markup: { weight: 0.03, label: "Schema.org Structured Data" },
+  table_list_extractability: { weight: 0.03, label: "Table & List Extractability" },
+  definition_patterns: { weight: 0.02, label: "Definition Patterns" },
+  visible_date_signal: { weight: 0.02, label: "Visible Date Signal" },
+  semantic_html: { weight: 0.02, label: "Semantic HTML5 & Accessibility" },
+  clean_html: { weight: 0.02, label: "Clean, Crawlable HTML" },
+  // Technical Plumbing
+  canonical_url: { weight: 0.01, label: "Canonical URL Strategy" }
 };
 function extractJsonLdBlocks(html) {
   const blocks = [];
@@ -3235,7 +3554,7 @@ function extractTitle(html) {
 function getTextContent2(html) {
   return html.replace(/<script[\s\S]*?<\/script>/gi, "").replace(/<style[\s\S]*?<\/style>/gi, "").replace(/<[^>]*>/g, " ").replace(/\s+/g, " ").trim();
 }
-function countWords(text) {
+function countWords2(text) {
   if (!text) return 0;
   return text.split(/\s+/).filter((w) => w.length > 0).length;
 }
@@ -3386,7 +3705,7 @@ function checkHasQuestionHeadings(html) {
 function analyzePage(html, url, category) {
   const title = extractTitle(html);
   const textContent = getTextContent2(html);
-  const wordCount = countWords(textContent);
+  const wordCount = countWords2(textContent);
   const issues = [];
   const strengths = [];
   const issueChecks = [
@@ -3539,7 +3858,7 @@ function extractTitle2(html) {
 function getTextContent3(html) {
   return html.replace(/<script[\s\S]*?<\/script>/gi, "").replace(/<style[\s\S]*?<\/style>/gi, "").replace(/<[^>]*>/g, " ").replace(/\s+/g, " ").trim();
 }
-function countWords2(text) {
+function countWords3(text) {
   if (!text) return 0;
   return text.split(/\s+/).filter((w) => w.length > 0).length;
 }
@@ -3687,7 +4006,7 @@ function buildLinkGraph(pages, domain, homepageUrl) {
     if (nodes.has(norm)) continue;
     const title = extractTitle2(page.text);
     const text = getTextContent3(page.text);
-    const wordCount = countWords2(text);
+    const wordCount = countWords3(text);
     nodes.set(norm, {
       url: norm,
       title,
@@ -3750,32 +4069,37 @@ function buildLinkGraph(pages, domain, homepageUrl) {
 
 // src/fix-engine.ts
 var CRITERION_WEIGHTS2 = {
-  llms_txt: 0.1,
-  schema_markup: 0.15,
-  qa_content_format: 0.15,
-  clean_html: 0.1,
-  entity_consistency: 0.1,
-  robots_txt: 0.05,
-  faq_section: 0.1,
+  // Content Substance (~55%)
+  topic_coherence: 0.14,
   original_data: 0.1,
-  internal_linking: 0.1,
-  semantic_html: 0.05,
-  content_freshness: 0.07,
-  sitemap_completeness: 0.05,
-  rss_feed: 0.03,
-  table_list_extractability: 0.07,
-  definition_patterns: 0.04,
-  direct_answer_density: 0.07,
-  content_licensing: 0.04,
-  author_schema_depth: 0.04,
-  fact_density: 0.05,
-  canonical_url: 0.04,
-  content_velocity: 0.03,
-  schema_coverage: 0.03,
-  speakable_schema: 0.03,
-  query_answer_alignment: 0.08,
-  content_cannibalization: 0.05,
-  visible_date_signal: 0.04
+  content_depth: 0.07,
+  fact_density: 0.06,
+  direct_answer_density: 0.05,
+  qa_content_format: 0.05,
+  query_answer_alignment: 0.05,
+  faq_section: 0.04,
+  // Content Organization (~30%)
+  entity_consistency: 0.05,
+  internal_linking: 0.04,
+  content_freshness: 0.04,
+  schema_markup: 0.03,
+  author_schema_depth: 0.03,
+  table_list_extractability: 0.03,
+  definition_patterns: 0.02,
+  visible_date_signal: 0.02,
+  semantic_html: 0.02,
+  clean_html: 0.02,
+  // Technical Plumbing (~15%)
+  content_cannibalization: 0.02,
+  llms_txt: 0.02,
+  robots_txt: 0.02,
+  content_velocity: 0.02,
+  content_licensing: 0.02,
+  sitemap_completeness: 0.01,
+  canonical_url: 0.01,
+  rss_feed: 0.01,
+  schema_coverage: 0.01,
+  speakable_schema: 0.01
 };
 var PHASE_CONFIG = [
   {
@@ -3798,7 +4122,9 @@ var PHASE_CONFIG = [
       "content_freshness",
       "table_list_extractability",
       "query_answer_alignment",
-      "visible_date_signal"
+      "visible_date_signal",
+      "topic_coherence",
+      "content_depth"
     ]
   },
   {
@@ -4702,6 +5028,55 @@ Summarization: yes`,
       affectedPages: affected,
       pageCount: affected?.length
     }];
+  },
+  topic_coherence: (c) => {
+    if (c.score >= 10) return [];
+    const impact = impactFromScore(c.score);
+    const effort = effortForCriterion("topic_coherence", c.score);
+    return [{
+      id: "fix-topic-coherence",
+      criterion: c.criterion_label,
+      criterionId: c.criterion,
+      title: "Focus blog content on core expertise",
+      description: "Ensure blog content consistently covers your core topic areas. Scattered content across unrelated topics weakens AI engine authority signals.",
+      impact,
+      effort: effort === "trivial" ? "low" : effort,
+      impactScore: 0,
+      category: "content",
+      steps: [
+        "Identify 2-3 core expertise areas your brand is known for",
+        "Audit existing blog posts and remove or consolidate off-topic content",
+        "Create a content calendar focused on core topics",
+        "Use topic clusters: pillar pages linking to supporting articles within the same niche"
+      ],
+      successCriteria: "80%+ of blog content covers core expertise areas with consistent topic focus"
+    }];
+  },
+  content_depth: (c, pages) => {
+    if (c.score >= 10) return [];
+    const impact = impactFromScore(c.score);
+    const effort = effortForCriterion("content_depth", c.score);
+    const affected = getAffectedPages("content_depth", pages);
+    return [{
+      id: "fix-content-depth",
+      criterion: c.criterion_label,
+      criterionId: c.criterion,
+      title: "Increase content depth and structure",
+      description: "Expand thin content with more detail, examples, and structured sections. AI engines prefer comprehensive articles with clear heading hierarchies.",
+      impact,
+      effort: effort === "trivial" ? "low" : effort,
+      impactScore: 0,
+      category: "content",
+      steps: [
+        "Aim for 1000+ words per article with expert analysis and examples",
+        "Use H2/H3 subheadings every 200-300 words for clear structure",
+        "Add comparison tables, numbered steps, and data points",
+        "Remove or expand thin pages (under 300 words) that dilute site quality"
+      ],
+      successCriteria: "Average article length exceeds 1000 words with 5+ subheadings per page",
+      affectedPages: affected,
+      pageCount: affected?.length
+    }];
   }
 };
 function generateFixPlan(domain, overallScore, criteria, pagesReviewed, linkGraph) {