[{"data":1,"prerenderedAt":1812},["ShallowReactive",2],{"article-alternates":3,"article-\u002Fru\u002Fai\u002Fllm-citation-measurement":13},{"i18nKey":4,"paths":5},"ai-002-2026-07",{"de":6,"en":7,"es":8,"fr":9,"it":10,"ru":11,"tr":12},"\u002Fde\u002Fai\u002Fllm-zitierungsmetriken","\u002Fen\u002Fai\u002Fllm-citation-measurement-new-seo-metrics","\u002Fes\u002Fai\u002Fmedicion-de-citacion-llm","\u002Ffr\u002Fai\u002Fmesure-des-citations-llm","\u002Fit\u002Fai\u002Fmisurazione-citazioni-llm","\u002Fru\u002Fai\u002Fllm-citation-measurement","\u002Ftr\u002Fai\u002Fllm-citation-olcumu-yeni-seo-metrik-setiniz",{"_path":11,"_dir":14,"_draft":15,"_partial":15,"_locale":16,"title":17,"description":18,"publishedAt":19,"modifiedAt":19,"category":14,"i18nKey":4,"tags":20,"readingTime":26,"author":27,"body":28,"_type":1806,"_id":1807,"_source":1808,"_file":1809,"_stem":1810,"_extension":1811},"ai",false,"","LLM Citation Measurement — Your New SEO Metric Set","How to track your brand's citation rate on Perplexity, ChatGPT, Gemini. Generative engine visibility metrics and measurement architecture.","2026-07-06",[21,22,23,24,25],"llm-citation","geo-metrics","ai-search","generative-seo","brand-visibility",8,"Roibase",{"type":29,"children":30,"toc":1795},"root",[31,45,52,57,67,77,87,93,98,117,127,137,147,154,159,165,177,187,192,218,680,685,695,851,856,1120,1136,1142,1154,1162,1192,1197,1526,1531,1537,1542,1547,1552,1676,1681,1687,1692,1700,1724,1732,1750,1764,1770,1775,1780,1784,1789],{"type":32,"tag":33,"props":34,"children":35},"element","p",{},[36,39],{"type":37,"value":38},"text","Traffic from Google's SERP dropped 40%, but your brand received 3 citations in ChatGPT's response. Gain or loss? Traditional SEO metrics — impressions, CTR, position — are no longer sufficient. Users are asking questions to LLMs, and Google Analytics won't show you whether your brand got cited. In 2026, the new reality for performance marketing teams is this: ",{"type":32,"tag":40,"props":41,"children":42},"strong",{},[43],{"type":37,"value":44},"if you don't measure citation rate, inference share, and source attribution, you're invisible.",{"type":32,"tag":46,"props":47,"children":49},"h2",{"id":48},"serp-metric-blindness",[50],{"type":37,"value":51},"SERP Metric Blindness",{"type":32,"tag":33,"props":53,"children":54},{},[55],{"type":37,"value":56},"Google Search Console tells you you're in 10th position with 5,000 impressions. But the same user querying Perplexity saw your content cited in the answer and went straight to your site — GSC logs this traffic as \"direct.\" An email summary generated with Claude API showed your brand as a source — Search Console can't see this interaction. This blindness operates on three levels:",{"type":32,"tag":33,"props":58,"children":59},{},[60,65],{"type":32,"tag":40,"props":61,"children":62},{},[63],{"type":37,"value":64},"Traffic attribution:",{"type":37,"value":66}," LLMs don't send referrer headers or use utm parameters. A visitor from a citation gets logged as \"organic search\" or \"direct.\" True source information is lost — you can't A\u002FB test, you can't calculate ROI.",{"type":32,"tag":33,"props":68,"children":69},{},[70,75],{"type":32,"tag":40,"props":71,"children":72},{},[73],{"type":37,"value":74},"Brand awareness:",{"type":37,"value":76}," Even if the user doesn't visit your site, they're learning about your brand. If ChatGPT lists your site as a \"trusted source\" in a 500-word answer, that creates brand lift. Traditional SEO tools can't capture this effect.",{"type":32,"tag":33,"props":78,"children":79},{},[80,85],{"type":32,"tag":40,"props":81,"children":82},{},[83],{"type":37,"value":84},"Competitive positioning:",{"type":37,"value":86}," Your competitor gets cited 5 times in the same prompt, you get 0 — but Search Console shows you both in 3rd place. Citation frequency is the new \"featured snippet win rate,\" but it's not on your dashboard yet.",{"type":32,"tag":46,"props":88,"children":90},{"id":89},"defining-citation-metrics",[91],{"type":37,"value":92},"Defining Citation Metrics",{"type":32,"tag":33,"props":94,"children":95},{},[96],{"type":37,"value":97},"To measure LLM visibility, you need 4 core metrics:",{"type":32,"tag":33,"props":99,"children":100},{},[101,106,108,115],{"type":32,"tag":40,"props":102,"children":103},{},[104],{"type":37,"value":105},"Citation rate:",{"type":37,"value":107}," How often your brand\u002Fcontent is referenced in LLM responses. Formula: ",{"type":32,"tag":109,"props":110,"children":112},"code",{"className":111},[],[113],{"type":37,"value":114},"(number of responses citing your brand) \u002F (total number of relevant queries)",{"type":37,"value":116},". Example: In \"headless commerce\" queries, ChatGPT cited you in 120 out of 1,000 responses = 12% citation rate. This metric directly indicates brand authority.",{"type":32,"tag":33,"props":118,"children":119},{},[120,125],{"type":32,"tag":40,"props":121,"children":122},{},[123],{"type":37,"value":124},"Source position:",{"type":37,"value":126}," Where you rank in the citation list. Perplexity typically shows 3-6 sources — being first results in 60% more click-through (internal Roibase test data, 2025 Q4). Without position tracking, you don't know the real value of your citation rate.",{"type":32,"tag":33,"props":128,"children":129},{},[130,135],{"type":32,"tag":40,"props":131,"children":132},{},[133],{"type":37,"value":134},"Inference share:",{"type":37,"value":136}," What percentage of the answer comes from your content. If ChatGPT generates a 300-word response and 80 words come from your article paragraph, that's measurable (semantic similarity threshold >0.85). High inference share = the model is using your tone, your framing — this is brand voice propagation.",{"type":32,"tag":33,"props":138,"children":139},{},[140,145],{"type":32,"tag":40,"props":141,"children":142},{},[143],{"type":37,"value":144},"Prompt coverage:",{"type":37,"value":146}," Which query types cite you. You get cited in informational queries (\"What is CDP\") but not commercial queries (\"CDP vendor comparison\")? Coverage analysis directs your editorial strategy — which intent gaps should you fill?",{"type":32,"tag":148,"props":149,"children":151},"h3",{"id":150},"measurement-frequency",[152],{"type":37,"value":153},"Measurement Frequency",{"type":32,"tag":33,"props":155,"children":156},{},[157],{"type":37,"value":158},"These metrics aren't real-time — LLMs are non-deterministic; the same prompt can produce different responses. Weekly batch measurement is sufficient: you automatically trigger 100-200 seed prompts, parse responses, and extract citations. Daily fluctuation is noise; weekly trend is signal.",{"type":32,"tag":46,"props":160,"children":162},{"id":161},"data-collection-architecture",[163],{"type":37,"value":164},"Data Collection Architecture",{"type":32,"tag":33,"props":166,"children":167},{},[168,170,175],{"type":37,"value":169},"Citation tracking needs 3 components: ",{"type":32,"tag":40,"props":171,"children":172},{},[173],{"type":37,"value":174},"prompt pipeline, response parser, attribution engine",{"type":37,"value":176},".",{"type":32,"tag":33,"props":178,"children":179},{},[180,185],{"type":32,"tag":40,"props":181,"children":182},{},[183],{"type":37,"value":184},"Prompt pipeline:",{"type":37,"value":186}," You send your seed keyword set (top 50-100 queries from GSC by impression) to each model API in parallel. Use n8n workflow or Airflow DAG triggered weekly. Model parameters must be fixed for each prompt — temperature=0.3, top_p=0.9 — otherwise results won't be reproducible.",{"type":32,"tag":33,"props":188,"children":189},{},[190],{"type":37,"value":191},"API cost estimate: ChatGPT-4o API ~$0.005\u002Fquery (500 token input + 1,500 token output average), Gemini Pro ~$0.003, Claude Sonnet ~$0.006. 100 prompts × 3 models × 4 weeks = 1,200 requests = $6-7\u002Fmonth. This budget is reasonable for weekly snapshots, though not real-time tracking.",{"type":32,"tag":33,"props":193,"children":194},{},[195,200,202,208,210,216],{"type":32,"tag":40,"props":196,"children":197},{},[198],{"type":37,"value":199},"Response parser:",{"type":37,"value":201}," You need to convert LLM output to structured data. Citation format varies by model — ChatGPT uses ",{"type":32,"tag":109,"props":203,"children":205},{"className":204},[],[206],{"type":37,"value":207},"[1]",{"type":37,"value":209},", Perplexity uses ",{"type":32,"tag":109,"props":211,"children":213},{"className":212},[],[214],{"type":37,"value":215},"[^1]",{"type":37,"value":217},", Claude uses markdown footnotes. Combine regex + NER (Named Entity Recognition): first extract citation markers, then match domain\u002Fbrand names. Python example:",{"type":32,"tag":219,"props":220,"children":224},"pre",{"className":221,"code":222,"language":223,"meta":16,"style":16},"language-python shiki shiki-themes github-dark","import re\nfrom urllib.parse import urlparse\n\ndef extract_citations(response_text):\n    # Citation pattern: [1], [^2], etc.\n    pattern = r'\\[(\\^?\\d+)\\]'\n    markers = re.findall(pattern, response_text)\n    \n    # Source URL extraction (model-specific)\n    sources = re.findall(r'https?:\u002F\u002F[^\\s\\)]+', response_text)\n    \n    citations = []\n    for idx, url in enumerate(sources):\n        domain = urlparse(url).netloc\n        citations.append({\n            'position': idx + 1,\n            'domain': domain,\n            'is_own_brand': 'roibase.com.tr' in domain\n        })\n    \n    return citations\n","python",[225],{"type":32,"tag":109,"props":226,"children":227},{"__ignoreMap":16},[228,246,269,279,299,309,381,399,407,416,496,504,522,551,569,578,606,620,649,658,666],{"type":32,"tag":229,"props":230,"children":233},"span",{"class":231,"line":232},"line",1,[234,240],{"type":32,"tag":229,"props":235,"children":237},{"style":236},"--shiki-default:#F97583",[238],{"type":37,"value":239},"import",{"type":32,"tag":229,"props":241,"children":243},{"style":242},"--shiki-default:#E1E4E8",[244],{"type":37,"value":245}," re\n",{"type":32,"tag":229,"props":247,"children":249},{"class":231,"line":248},2,[250,255,260,264],{"type":32,"tag":229,"props":251,"children":252},{"style":236},[253],{"type":37,"value":254},"from",{"type":32,"tag":229,"props":256,"children":257},{"style":242},[258],{"type":37,"value":259}," urllib.parse ",{"type":32,"tag":229,"props":261,"children":262},{"style":236},[263],{"type":37,"value":239},{"type":32,"tag":229,"props":265,"children":266},{"style":242},[267],{"type":37,"value":268}," urlparse\n",{"type":32,"tag":229,"props":270,"children":272},{"class":231,"line":271},3,[273],{"type":32,"tag":229,"props":274,"children":276},{"emptyLinePlaceholder":275},true,[277],{"type":37,"value":278},"\n",{"type":32,"tag":229,"props":280,"children":282},{"class":231,"line":281},4,[283,288,294],{"type":32,"tag":229,"props":284,"children":285},{"style":236},[286],{"type":37,"value":287},"def",{"type":32,"tag":229,"props":289,"children":291},{"style":290},"--shiki-default:#B392F0",[292],{"type":37,"value":293}," extract_citations",{"type":32,"tag":229,"props":295,"children":296},{"style":242},[297],{"type":37,"value":298},"(response_text):\n",{"type":32,"tag":229,"props":300,"children":302},{"class":231,"line":301},5,[303],{"type":32,"tag":229,"props":304,"children":306},{"style":305},"--shiki-default:#6A737D",[307],{"type":37,"value":308},"    # Citation pattern: [1], [^2], etc.\n",{"type":32,"tag":229,"props":310,"children":312},{"class":231,"line":311},6,[313,318,323,328,334,340,346,351,356,361,366,371,376],{"type":32,"tag":229,"props":314,"children":315},{"style":242},[316],{"type":37,"value":317},"    pattern ",{"type":32,"tag":229,"props":319,"children":320},{"style":236},[321],{"type":37,"value":322},"=",{"type":32,"tag":229,"props":324,"children":325},{"style":236},[326],{"type":37,"value":327}," r",{"type":32,"tag":229,"props":329,"children":331},{"style":330},"--shiki-default:#9ECBFF",[332],{"type":37,"value":333},"'",{"type":32,"tag":229,"props":335,"children":337},{"style":336},"--shiki-default:#85E89D;--shiki-default-font-weight:bold",[338],{"type":37,"value":339},"\\[",{"type":32,"tag":229,"props":341,"children":343},{"style":342},"--shiki-default:#79B8FF",[344],{"type":37,"value":345},"(",{"type":32,"tag":229,"props":347,"children":348},{"style":336},[349],{"type":37,"value":350},"\\^",{"type":32,"tag":229,"props":352,"children":353},{"style":236},[354],{"type":37,"value":355},"?",{"type":32,"tag":229,"props":357,"children":358},{"style":342},[359],{"type":37,"value":360},"\\d",{"type":32,"tag":229,"props":362,"children":363},{"style":236},[364],{"type":37,"value":365},"+",{"type":32,"tag":229,"props":367,"children":368},{"style":342},[369],{"type":37,"value":370},")",{"type":32,"tag":229,"props":372,"children":373},{"style":336},[374],{"type":37,"value":375},"\\]",{"type":32,"tag":229,"props":377,"children":378},{"style":330},[379],{"type":37,"value":380},"'\n",{"type":32,"tag":229,"props":382,"children":384},{"class":231,"line":383},7,[385,390,394],{"type":32,"tag":229,"props":386,"children":387},{"style":242},[388],{"type":37,"value":389},"    markers ",{"type":32,"tag":229,"props":391,"children":392},{"style":236},[393],{"type":37,"value":322},{"type":32,"tag":229,"props":395,"children":396},{"style":242},[397],{"type":37,"value":398}," re.findall(pattern, response_text)\n",{"type":32,"tag":229,"props":400,"children":401},{"class":231,"line":26},[402],{"type":32,"tag":229,"props":403,"children":404},{"style":242},[405],{"type":37,"value":406},"    \n",{"type":32,"tag":229,"props":408,"children":410},{"class":231,"line":409},9,[411],{"type":32,"tag":229,"props":412,"children":413},{"style":305},[414],{"type":37,"value":415},"    # Source URL extraction (model-specific)\n",{"type":32,"tag":229,"props":417,"children":419},{"class":231,"line":418},10,[420,425,429,434,439,443,449,453,458,463,468,473,478,483,487,491],{"type":32,"tag":229,"props":421,"children":422},{"style":242},[423],{"type":37,"value":424},"    sources ",{"type":32,"tag":229,"props":426,"children":427},{"style":236},[428],{"type":37,"value":322},{"type":32,"tag":229,"props":430,"children":431},{"style":242},[432],{"type":37,"value":433}," re.findall(",{"type":32,"tag":229,"props":435,"children":436},{"style":236},[437],{"type":37,"value":438},"r",{"type":32,"tag":229,"props":440,"children":441},{"style":330},[442],{"type":37,"value":333},{"type":32,"tag":229,"props":444,"children":446},{"style":445},"--shiki-default:#DBEDFF",[447],{"type":37,"value":448},"https",{"type":32,"tag":229,"props":450,"children":451},{"style":236},[452],{"type":37,"value":355},{"type":32,"tag":229,"props":454,"children":455},{"style":445},[456],{"type":37,"value":457},":\u002F\u002F",{"type":32,"tag":229,"props":459,"children":460},{"style":342},[461],{"type":37,"value":462},"[",{"type":32,"tag":229,"props":464,"children":465},{"style":236},[466],{"type":37,"value":467},"^",{"type":32,"tag":229,"props":469,"children":470},{"style":342},[471],{"type":37,"value":472},"\\s",{"type":32,"tag":229,"props":474,"children":475},{"style":336},[476],{"type":37,"value":477},"\\)",{"type":32,"tag":229,"props":479,"children":480},{"style":342},[481],{"type":37,"value":482},"]",{"type":32,"tag":229,"props":484,"children":485},{"style":236},[486],{"type":37,"value":365},{"type":32,"tag":229,"props":488,"children":489},{"style":330},[490],{"type":37,"value":333},{"type":32,"tag":229,"props":492,"children":493},{"style":242},[494],{"type":37,"value":495},", response_text)\n",{"type":32,"tag":229,"props":497,"children":499},{"class":231,"line":498},11,[500],{"type":32,"tag":229,"props":501,"children":502},{"style":242},[503],{"type":37,"value":406},{"type":32,"tag":229,"props":505,"children":507},{"class":231,"line":506},12,[508,513,517],{"type":32,"tag":229,"props":509,"children":510},{"style":242},[511],{"type":37,"value":512},"    citations ",{"type":32,"tag":229,"props":514,"children":515},{"style":236},[516],{"type":37,"value":322},{"type":32,"tag":229,"props":518,"children":519},{"style":242},[520],{"type":37,"value":521}," []\n",{"type":32,"tag":229,"props":523,"children":525},{"class":231,"line":524},13,[526,531,536,541,546],{"type":32,"tag":229,"props":527,"children":528},{"style":236},[529],{"type":37,"value":530},"    for",{"type":32,"tag":229,"props":532,"children":533},{"style":242},[534],{"type":37,"value":535}," idx, url ",{"type":32,"tag":229,"props":537,"children":538},{"style":236},[539],{"type":37,"value":540},"in",{"type":32,"tag":229,"props":542,"children":543},{"style":342},[544],{"type":37,"value":545}," enumerate",{"type":32,"tag":229,"props":547,"children":548},{"style":242},[549],{"type":37,"value":550},"(sources):\n",{"type":32,"tag":229,"props":552,"children":554},{"class":231,"line":553},14,[555,560,564],{"type":32,"tag":229,"props":556,"children":557},{"style":242},[558],{"type":37,"value":559},"        domain ",{"type":32,"tag":229,"props":561,"children":562},{"style":236},[563],{"type":37,"value":322},{"type":32,"tag":229,"props":565,"children":566},{"style":242},[567],{"type":37,"value":568}," urlparse(url).netloc\n",{"type":32,"tag":229,"props":570,"children":572},{"class":231,"line":571},15,[573],{"type":32,"tag":229,"props":574,"children":575},{"style":242},[576],{"type":37,"value":577},"        citations.append({\n",{"type":32,"tag":229,"props":579,"children":581},{"class":231,"line":580},16,[582,587,592,596,601],{"type":32,"tag":229,"props":583,"children":584},{"style":330},[585],{"type":37,"value":586},"            'position'",{"type":32,"tag":229,"props":588,"children":589},{"style":242},[590],{"type":37,"value":591},": idx ",{"type":32,"tag":229,"props":593,"children":594},{"style":236},[595],{"type":37,"value":365},{"type":32,"tag":229,"props":597,"children":598},{"style":342},[599],{"type":37,"value":600}," 1",{"type":32,"tag":229,"props":602,"children":603},{"style":242},[604],{"type":37,"value":605},",\n",{"type":32,"tag":229,"props":607,"children":609},{"class":231,"line":608},17,[610,615],{"type":32,"tag":229,"props":611,"children":612},{"style":330},[613],{"type":37,"value":614},"            'domain'",{"type":32,"tag":229,"props":616,"children":617},{"style":242},[618],{"type":37,"value":619},": domain,\n",{"type":32,"tag":229,"props":621,"children":623},{"class":231,"line":622},18,[624,629,634,639,644],{"type":32,"tag":229,"props":625,"children":626},{"style":330},[627],{"type":37,"value":628},"            'is_own_brand'",{"type":32,"tag":229,"props":630,"children":631},{"style":242},[632],{"type":37,"value":633},": ",{"type":32,"tag":229,"props":635,"children":636},{"style":330},[637],{"type":37,"value":638},"'roibase.com.tr'",{"type":32,"tag":229,"props":640,"children":641},{"style":236},[642],{"type":37,"value":643}," in",{"type":32,"tag":229,"props":645,"children":646},{"style":242},[647],{"type":37,"value":648}," domain\n",{"type":32,"tag":229,"props":650,"children":652},{"class":231,"line":651},19,[653],{"type":32,"tag":229,"props":654,"children":655},{"style":242},[656],{"type":37,"value":657},"        })\n",{"type":32,"tag":229,"props":659,"children":661},{"class":231,"line":660},20,[662],{"type":32,"tag":229,"props":663,"children":664},{"style":242},[665],{"type":37,"value":406},{"type":32,"tag":229,"props":667,"children":669},{"class":231,"line":668},21,[670,675],{"type":32,"tag":229,"props":671,"children":672},{"style":236},[673],{"type":37,"value":674},"    return",{"type":32,"tag":229,"props":676,"children":677},{"style":242},[678],{"type":37,"value":679}," citations\n",{"type":32,"tag":33,"props":681,"children":682},{},[683],{"type":37,"value":684},"This simple parser gives ~85% accuracy — edge cases (embedded links, paywalled sources) need periodic manual QA.",{"type":32,"tag":33,"props":686,"children":687},{},[688,693],{"type":32,"tag":40,"props":689,"children":690},{},[691],{"type":37,"value":692},"Attribution engine:",{"type":37,"value":694}," You write extracted citations to a warehouse and calculate aggregate metrics. BigQuery or Snowflake table schema:",{"type":32,"tag":696,"props":697,"children":698},"table",{},[699,723],{"type":32,"tag":700,"props":701,"children":702},"thead",{},[703],{"type":32,"tag":704,"props":705,"children":706},"tr",{},[707,713,718],{"type":32,"tag":708,"props":709,"children":710},"th",{},[711],{"type":37,"value":712},"Column",{"type":32,"tag":708,"props":714,"children":715},{},[716],{"type":37,"value":717},"Type",{"type":32,"tag":708,"props":719,"children":720},{},[721],{"type":37,"value":722},"Description",{"type":32,"tag":724,"props":725,"children":726},"tbody",{},[727,746,763,780,797,815,833],{"type":32,"tag":704,"props":728,"children":729},{},[730,736,741],{"type":32,"tag":731,"props":732,"children":733},"td",{},[734],{"type":37,"value":735},"query_text",{"type":32,"tag":731,"props":737,"children":738},{},[739],{"type":37,"value":740},"STRING",{"type":32,"tag":731,"props":742,"children":743},{},[744],{"type":37,"value":745},"Seed prompt",{"type":32,"tag":704,"props":747,"children":748},{},[749,754,758],{"type":32,"tag":731,"props":750,"children":751},{},[752],{"type":37,"value":753},"model_name",{"type":32,"tag":731,"props":755,"children":756},{},[757],{"type":37,"value":740},{"type":32,"tag":731,"props":759,"children":760},{},[761],{"type":37,"value":762},"chatgpt-4o, gemini-pro, claude-sonnet",{"type":32,"tag":704,"props":764,"children":765},{},[766,771,775],{"type":32,"tag":731,"props":767,"children":768},{},[769],{"type":37,"value":770},"response_id",{"type":32,"tag":731,"props":772,"children":773},{},[774],{"type":37,"value":740},{"type":32,"tag":731,"props":776,"children":777},{},[778],{"type":37,"value":779},"Unique identifier",{"type":32,"tag":704,"props":781,"children":782},{},[783,788,792],{"type":32,"tag":731,"props":784,"children":785},{},[786],{"type":37,"value":787},"citation_domain",{"type":32,"tag":731,"props":789,"children":790},{},[791],{"type":37,"value":740},{"type":32,"tag":731,"props":793,"children":794},{},[795],{"type":37,"value":796},"Cited domain",{"type":32,"tag":704,"props":798,"children":799},{},[800,805,810],{"type":32,"tag":731,"props":801,"children":802},{},[803],{"type":37,"value":804},"citation_position",{"type":32,"tag":731,"props":806,"children":807},{},[808],{"type":37,"value":809},"INTEGER",{"type":32,"tag":731,"props":811,"children":812},{},[813],{"type":37,"value":814},"Rank in source list",{"type":32,"tag":704,"props":816,"children":817},{},[818,823,828],{"type":32,"tag":731,"props":819,"children":820},{},[821],{"type":37,"value":822},"inference_similarity",{"type":32,"tag":731,"props":824,"children":825},{},[826],{"type":37,"value":827},"FLOAT",{"type":32,"tag":731,"props":829,"children":830},{},[831],{"type":37,"value":832},"Semantic overlap (0-1)",{"type":32,"tag":704,"props":834,"children":835},{},[836,841,846],{"type":32,"tag":731,"props":837,"children":838},{},[839],{"type":37,"value":840},"measured_at",{"type":32,"tag":731,"props":842,"children":843},{},[844],{"type":37,"value":845},"TIMESTAMP",{"type":32,"tag":731,"props":847,"children":848},{},[849],{"type":37,"value":850},"Measurement timestamp",{"type":32,"tag":33,"props":852,"children":853},{},[854],{"type":37,"value":855},"Build a weekly aggregate view on this table:",{"type":32,"tag":219,"props":857,"children":861},{"className":858,"code":859,"language":860,"meta":16,"style":16},"language-sql shiki shiki-themes github-dark","SELECT \n  model_name,\n  COUNT(DISTINCT query_text) AS total_queries,\n  SUM(CASE WHEN citation_domain LIKE '%roibase%' THEN 1 ELSE 0 END) AS own_citations,\n  AVG(CASE WHEN citation_domain LIKE '%roibase%' THEN citation_position ELSE NULL END) AS avg_position\nFROM citation_log\nWHERE measured_at >= CURRENT_DATE() - 7\nGROUP BY model_name;\n","sql",[862],{"type":32,"tag":109,"props":863,"children":864},{"__ignoreMap":16},[865,878,886,918,993,1061,1074,1107],{"type":32,"tag":229,"props":866,"children":867},{"class":231,"line":232},[868,873],{"type":32,"tag":229,"props":869,"children":870},{"style":236},[871],{"type":37,"value":872},"SELECT",{"type":32,"tag":229,"props":874,"children":875},{"style":242},[876],{"type":37,"value":877}," \n",{"type":32,"tag":229,"props":879,"children":880},{"class":231,"line":248},[881],{"type":32,"tag":229,"props":882,"children":883},{"style":242},[884],{"type":37,"value":885},"  model_name,\n",{"type":32,"tag":229,"props":887,"children":888},{"class":231,"line":271},[889,894,898,903,908,913],{"type":32,"tag":229,"props":890,"children":891},{"style":342},[892],{"type":37,"value":893},"  COUNT",{"type":32,"tag":229,"props":895,"children":896},{"style":242},[897],{"type":37,"value":345},{"type":32,"tag":229,"props":899,"children":900},{"style":236},[901],{"type":37,"value":902},"DISTINCT",{"type":32,"tag":229,"props":904,"children":905},{"style":242},[906],{"type":37,"value":907}," query_text) ",{"type":32,"tag":229,"props":909,"children":910},{"style":236},[911],{"type":37,"value":912},"AS",{"type":32,"tag":229,"props":914,"children":915},{"style":242},[916],{"type":37,"value":917}," total_queries,\n",{"type":32,"tag":229,"props":919,"children":920},{"class":231,"line":281},[921,926,930,935,940,945,950,955,960,964,969,974,979,984,988],{"type":32,"tag":229,"props":922,"children":923},{"style":342},[924],{"type":37,"value":925},"  SUM",{"type":32,"tag":229,"props":927,"children":928},{"style":242},[929],{"type":37,"value":345},{"type":32,"tag":229,"props":931,"children":932},{"style":236},[933],{"type":37,"value":934},"CASE",{"type":32,"tag":229,"props":936,"children":937},{"style":236},[938],{"type":37,"value":939}," WHEN",{"type":32,"tag":229,"props":941,"children":942},{"style":242},[943],{"type":37,"value":944}," citation_domain ",{"type":32,"tag":229,"props":946,"children":947},{"style":236},[948],{"type":37,"value":949},"LIKE",{"type":32,"tag":229,"props":951,"children":952},{"style":330},[953],{"type":37,"value":954}," '%roibase%'",{"type":32,"tag":229,"props":956,"children":957},{"style":236},[958],{"type":37,"value":959}," THEN",{"type":32,"tag":229,"props":961,"children":962},{"style":342},[963],{"type":37,"value":600},{"type":32,"tag":229,"props":965,"children":966},{"style":236},[967],{"type":37,"value":968}," ELSE",{"type":32,"tag":229,"props":970,"children":971},{"style":342},[972],{"type":37,"value":973}," 0",{"type":32,"tag":229,"props":975,"children":976},{"style":236},[977],{"type":37,"value":978}," END",{"type":32,"tag":229,"props":980,"children":981},{"style":242},[982],{"type":37,"value":983},") ",{"type":32,"tag":229,"props":985,"children":986},{"style":236},[987],{"type":37,"value":912},{"type":32,"tag":229,"props":989,"children":990},{"style":242},[991],{"type":37,"value":992}," own_citations,\n",{"type":32,"tag":229,"props":994,"children":995},{"class":231,"line":301},[996,1001,1005,1009,1013,1017,1021,1025,1029,1034,1039,1044,1048,1052,1056],{"type":32,"tag":229,"props":997,"children":998},{"style":342},[999],{"type":37,"value":1000},"  AVG",{"type":32,"tag":229,"props":1002,"children":1003},{"style":242},[1004],{"type":37,"value":345},{"type":32,"tag":229,"props":1006,"children":1007},{"style":236},[1008],{"type":37,"value":934},{"type":32,"tag":229,"props":1010,"children":1011},{"style":236},[1012],{"type":37,"value":939},{"type":32,"tag":229,"props":1014,"children":1015},{"style":242},[1016],{"type":37,"value":944},{"type":32,"tag":229,"props":1018,"children":1019},{"style":236},[1020],{"type":37,"value":949},{"type":32,"tag":229,"props":1022,"children":1023},{"style":330},[1024],{"type":37,"value":954},{"type":32,"tag":229,"props":1026,"children":1027},{"style":236},[1028],{"type":37,"value":959},{"type":32,"tag":229,"props":1030,"children":1031},{"style":242},[1032],{"type":37,"value":1033}," citation_position ",{"type":32,"tag":229,"props":1035,"children":1036},{"style":236},[1037],{"type":37,"value":1038},"ELSE",{"type":32,"tag":229,"props":1040,"children":1041},{"style":236},[1042],{"type":37,"value":1043}," NULL",{"type":32,"tag":229,"props":1045,"children":1046},{"style":236},[1047],{"type":37,"value":978},{"type":32,"tag":229,"props":1049,"children":1050},{"style":242},[1051],{"type":37,"value":983},{"type":32,"tag":229,"props":1053,"children":1054},{"style":236},[1055],{"type":37,"value":912},{"type":32,"tag":229,"props":1057,"children":1058},{"style":242},[1059],{"type":37,"value":1060}," avg_position\n",{"type":32,"tag":229,"props":1062,"children":1063},{"class":231,"line":311},[1064,1069],{"type":32,"tag":229,"props":1065,"children":1066},{"style":236},[1067],{"type":37,"value":1068},"FROM",{"type":32,"tag":229,"props":1070,"children":1071},{"style":242},[1072],{"type":37,"value":1073}," citation_log\n",{"type":32,"tag":229,"props":1075,"children":1076},{"class":231,"line":383},[1077,1082,1087,1092,1097,1102],{"type":32,"tag":229,"props":1078,"children":1079},{"style":236},[1080],{"type":37,"value":1081},"WHERE",{"type":32,"tag":229,"props":1083,"children":1084},{"style":242},[1085],{"type":37,"value":1086}," measured_at ",{"type":32,"tag":229,"props":1088,"children":1089},{"style":236},[1090],{"type":37,"value":1091},">=",{"type":32,"tag":229,"props":1093,"children":1094},{"style":242},[1095],{"type":37,"value":1096}," CURRENT_DATE() ",{"type":32,"tag":229,"props":1098,"children":1099},{"style":236},[1100],{"type":37,"value":1101},"-",{"type":32,"tag":229,"props":1103,"children":1104},{"style":342},[1105],{"type":37,"value":1106}," 7\n",{"type":32,"tag":229,"props":1108,"children":1109},{"class":231,"line":26},[1110,1115],{"type":32,"tag":229,"props":1111,"children":1112},{"style":236},[1113],{"type":37,"value":1114},"GROUP BY",{"type":32,"tag":229,"props":1116,"children":1117},{"style":242},[1118],{"type":37,"value":1119}," model_name;\n",{"type":32,"tag":33,"props":1121,"children":1122},{},[1123,1125,1134],{"type":37,"value":1124},"Output: 14% citation rate on ChatGPT, 8% on Gemini, 19% on Claude — these differences relate to model training data cut-off dates and retrieval strategy. Once you have this insight, you can model-specifically optimize your ",{"type":32,"tag":1126,"props":1127,"children":1131},"a",{"href":1128,"rel":1129},"https:\u002F\u002Fwww.roibase.com.tr\u002Fru\u002Fgeo",[1130],"nofollow",[1132],{"type":37,"value":1133},"Generative Engine Optimization",{"type":37,"value":1135}," strategy.",{"type":32,"tag":46,"props":1137,"children":1139},{"id":1138},"inference-share-calculation",[1140],{"type":37,"value":1141},"Inference Share Calculation",{"type":32,"tag":33,"props":1143,"children":1144},{},[1145,1147,1152],{"type":37,"value":1146},"Citation rate measures your visibility; inference share measures ",{"type":32,"tag":40,"props":1148,"children":1149},{},[1150],{"type":37,"value":1151},"how much of your content is actually used",{"type":37,"value":1153},". Method: semantic embedding similarity.",{"type":32,"tag":33,"props":1155,"children":1156},{},[1157],{"type":32,"tag":40,"props":1158,"children":1159},{},[1160],{"type":37,"value":1161},"Steps:",{"type":32,"tag":1163,"props":1164,"children":1165},"ol",{},[1166,1172,1177,1182,1187],{"type":32,"tag":1167,"props":1168,"children":1169},"li",{},[1170],{"type":37,"value":1171},"Chunk your source content (blog post, whitepaper) by sentence\u002Fparagraph",{"type":32,"tag":1167,"props":1173,"children":1174},{},[1175],{"type":37,"value":1176},"Chunk the LLM response the same way",{"type":32,"tag":1167,"props":1178,"children":1179},{},[1180],{"type":37,"value":1181},"For each response chunk, find the source chunk with highest similarity (cosine similarity)",{"type":32,"tag":1167,"props":1183,"children":1184},{},[1185],{"type":37,"value":1186},"Count matches above threshold (typically >0.85)",{"type":32,"tag":1167,"props":1188,"children":1189},{},[1190],{"type":37,"value":1191},"Inference share = (matched response chunks) \u002F (total response chunks)",{"type":32,"tag":33,"props":1193,"children":1194},{},[1195],{"type":37,"value":1196},"Python implementation with sentence-transformers:",{"type":32,"tag":219,"props":1198,"children":1200},{"className":221,"code":1199,"language":223,"meta":16,"style":16},"from sentence_transformers import SentenceTransformer, util\n\nmodel = SentenceTransformer('all-MiniLM-L6-v2')\n\nsource_chunks = [\"CDP collects first-party data...\", \"Attribution window is 7 days...\"]\nresponse_chunks = [\"CDP gathers customer data...\", \"Conversion window is typically 7 days...\"]\n\nsource_embeddings = model.encode(source_chunks)\nresponse_embeddings = model.encode(response_chunks)\n\nmatched = 0\nfor resp_emb in response_embeddings:\n    similarities = util.cos_sim(resp_emb, source_embeddings)\n    if similarities.max() > 0.85:\n        matched += 1\n\ninference_share = matched \u002F len(response_chunks)\n",[1201],{"type":32,"tag":109,"props":1202,"children":1203},{"__ignoreMap":16},[1204,1225,1232,1259,1266,1303,1337,1344,1361,1378,1385,1402,1424,1441,1469,1487,1494],{"type":32,"tag":229,"props":1205,"children":1206},{"class":231,"line":232},[1207,1211,1216,1220],{"type":32,"tag":229,"props":1208,"children":1209},{"style":236},[1210],{"type":37,"value":254},{"type":32,"tag":229,"props":1212,"children":1213},{"style":242},[1214],{"type":37,"value":1215}," sentence_transformers ",{"type":32,"tag":229,"props":1217,"children":1218},{"style":236},[1219],{"type":37,"value":239},{"type":32,"tag":229,"props":1221,"children":1222},{"style":242},[1223],{"type":37,"value":1224}," SentenceTransformer, util\n",{"type":32,"tag":229,"props":1226,"children":1227},{"class":231,"line":248},[1228],{"type":32,"tag":229,"props":1229,"children":1230},{"emptyLinePlaceholder":275},[1231],{"type":37,"value":278},{"type":32,"tag":229,"props":1233,"children":1234},{"class":231,"line":271},[1235,1240,1244,1249,1254],{"type":32,"tag":229,"props":1236,"children":1237},{"style":242},[1238],{"type":37,"value":1239},"model ",{"type":32,"tag":229,"props":1241,"children":1242},{"style":236},[1243],{"type":37,"value":322},{"type":32,"tag":229,"props":1245,"children":1246},{"style":242},[1247],{"type":37,"value":1248}," SentenceTransformer(",{"type":32,"tag":229,"props":1250,"children":1251},{"style":330},[1252],{"type":37,"value":1253},"'all-MiniLM-L6-v2'",{"type":32,"tag":229,"props":1255,"children":1256},{"style":242},[1257],{"type":37,"value":1258},")\n",{"type":32,"tag":229,"props":1260,"children":1261},{"class":231,"line":281},[1262],{"type":32,"tag":229,"props":1263,"children":1264},{"emptyLinePlaceholder":275},[1265],{"type":37,"value":278},{"type":32,"tag":229,"props":1267,"children":1268},{"class":231,"line":301},[1269,1274,1278,1283,1288,1293,1298],{"type":32,"tag":229,"props":1270,"children":1271},{"style":242},[1272],{"type":37,"value":1273},"source_chunks ",{"type":32,"tag":229,"props":1275,"children":1276},{"style":236},[1277],{"type":37,"value":322},{"type":32,"tag":229,"props":1279,"children":1280},{"style":242},[1281],{"type":37,"value":1282}," [",{"type":32,"tag":229,"props":1284,"children":1285},{"style":330},[1286],{"type":37,"value":1287},"\"CDP collects first-party data...\"",{"type":32,"tag":229,"props":1289,"children":1290},{"style":242},[1291],{"type":37,"value":1292},", ",{"type":32,"tag":229,"props":1294,"children":1295},{"style":330},[1296],{"type":37,"value":1297},"\"Attribution window is 7 days...\"",{"type":32,"tag":229,"props":1299,"children":1300},{"style":242},[1301],{"type":37,"value":1302},"]\n",{"type":32,"tag":229,"props":1304,"children":1305},{"class":231,"line":311},[1306,1311,1315,1319,1324,1328,1333],{"type":32,"tag":229,"props":1307,"children":1308},{"style":242},[1309],{"type":37,"value":1310},"response_chunks ",{"type":32,"tag":229,"props":1312,"children":1313},{"style":236},[1314],{"type":37,"value":322},{"type":32,"tag":229,"props":1316,"children":1317},{"style":242},[1318],{"type":37,"value":1282},{"type":32,"tag":229,"props":1320,"children":1321},{"style":330},[1322],{"type":37,"value":1323},"\"CDP gathers customer data...\"",{"type":32,"tag":229,"props":1325,"children":1326},{"style":242},[1327],{"type":37,"value":1292},{"type":32,"tag":229,"props":1329,"children":1330},{"style":330},[1331],{"type":37,"value":1332},"\"Conversion window is typically 7 days...\"",{"type":32,"tag":229,"props":1334,"children":1335},{"style":242},[1336],{"type":37,"value":1302},{"type":32,"tag":229,"props":1338,"children":1339},{"class":231,"line":383},[1340],{"type":32,"tag":229,"props":1341,"children":1342},{"emptyLinePlaceholder":275},[1343],{"type":37,"value":278},{"type":32,"tag":229,"props":1345,"children":1346},{"class":231,"line":26},[1347,1352,1356],{"type":32,"tag":229,"props":1348,"children":1349},{"style":242},[1350],{"type":37,"value":1351},"source_embeddings ",{"type":32,"tag":229,"props":1353,"children":1354},{"style":236},[1355],{"type":37,"value":322},{"type":32,"tag":229,"props":1357,"children":1358},{"style":242},[1359],{"type":37,"value":1360}," model.encode(source_chunks)\n",{"type":32,"tag":229,"props":1362,"children":1363},{"class":231,"line":409},[1364,1369,1373],{"type":32,"tag":229,"props":1365,"children":1366},{"style":242},[1367],{"type":37,"value":1368},"response_embeddings ",{"type":32,"tag":229,"props":1370,"children":1371},{"style":236},[1372],{"type":37,"value":322},{"type":32,"tag":229,"props":1374,"children":1375},{"style":242},[1376],{"type":37,"value":1377}," model.encode(response_chunks)\n",{"type":32,"tag":229,"props":1379,"children":1380},{"class":231,"line":418},[1381],{"type":32,"tag":229,"props":1382,"children":1383},{"emptyLinePlaceholder":275},[1384],{"type":37,"value":278},{"type":32,"tag":229,"props":1386,"children":1387},{"class":231,"line":498},[1388,1393,1397],{"type":32,"tag":229,"props":1389,"children":1390},{"style":242},[1391],{"type":37,"value":1392},"matched ",{"type":32,"tag":229,"props":1394,"children":1395},{"style":236},[1396],{"type":37,"value":322},{"type":32,"tag":229,"props":1398,"children":1399},{"style":342},[1400],{"type":37,"value":1401}," 0\n",{"type":32,"tag":229,"props":1403,"children":1404},{"class":231,"line":506},[1405,1410,1415,1419],{"type":32,"tag":229,"props":1406,"children":1407},{"style":236},[1408],{"type":37,"value":1409},"for",{"type":32,"tag":229,"props":1411,"children":1412},{"style":242},[1413],{"type":37,"value":1414}," resp_emb ",{"type":32,"tag":229,"props":1416,"children":1417},{"style":236},[1418],{"type":37,"value":540},{"type":32,"tag":229,"props":1420,"children":1421},{"style":242},[1422],{"type":37,"value":1423}," response_embeddings:\n",{"type":32,"tag":229,"props":1425,"children":1426},{"class":231,"line":524},[1427,1432,1436],{"type":32,"tag":229,"props":1428,"children":1429},{"style":242},[1430],{"type":37,"value":1431},"    similarities ",{"type":32,"tag":229,"props":1433,"children":1434},{"style":236},[1435],{"type":37,"value":322},{"type":32,"tag":229,"props":1437,"children":1438},{"style":242},[1439],{"type":37,"value":1440}," util.cos_sim(resp_emb, source_embeddings)\n",{"type":32,"tag":229,"props":1442,"children":1443},{"class":231,"line":553},[1444,1449,1454,1459,1464],{"type":32,"tag":229,"props":1445,"children":1446},{"style":236},[1447],{"type":37,"value":1448},"    if",{"type":32,"tag":229,"props":1450,"children":1451},{"style":242},[1452],{"type":37,"value":1453}," similarities.max() ",{"type":32,"tag":229,"props":1455,"children":1456},{"style":236},[1457],{"type":37,"value":1458},">",{"type":32,"tag":229,"props":1460,"children":1461},{"style":342},[1462],{"type":37,"value":1463}," 0.85",{"type":32,"tag":229,"props":1465,"children":1466},{"style":242},[1467],{"type":37,"value":1468},":\n",{"type":32,"tag":229,"props":1470,"children":1471},{"class":231,"line":571},[1472,1477,1482],{"type":32,"tag":229,"props":1473,"children":1474},{"style":242},[1475],{"type":37,"value":1476},"        matched ",{"type":32,"tag":229,"props":1478,"children":1479},{"style":236},[1480],{"type":37,"value":1481},"+=",{"type":32,"tag":229,"props":1483,"children":1484},{"style":342},[1485],{"type":37,"value":1486}," 1\n",{"type":32,"tag":229,"props":1488,"children":1489},{"class":231,"line":580},[1490],{"type":32,"tag":229,"props":1491,"children":1492},{"emptyLinePlaceholder":275},[1493],{"type":37,"value":278},{"type":32,"tag":229,"props":1495,"children":1496},{"class":231,"line":608},[1497,1502,1506,1511,1516,1521],{"type":32,"tag":229,"props":1498,"children":1499},{"style":242},[1500],{"type":37,"value":1501},"inference_share ",{"type":32,"tag":229,"props":1503,"children":1504},{"style":236},[1505],{"type":37,"value":322},{"type":32,"tag":229,"props":1507,"children":1508},{"style":242},[1509],{"type":37,"value":1510}," matched ",{"type":32,"tag":229,"props":1512,"children":1513},{"style":236},[1514],{"type":37,"value":1515},"\u002F",{"type":32,"tag":229,"props":1517,"children":1518},{"style":342},[1519],{"type":37,"value":1520}," len",{"type":32,"tag":229,"props":1522,"children":1523},{"style":242},[1524],{"type":37,"value":1525},"(response_chunks)\n",{"type":32,"tag":33,"props":1527,"children":1528},{},[1529],{"type":37,"value":1530},"Above 60% inference share = the LLM is heavily repurposing your content. This is both positive (brand authority) and negative (direct traffic loss) — you should highlight this tradeoff on your exec dashboard.",{"type":32,"tag":46,"props":1532,"children":1534},{"id":1533},"prompt-coverage-analysis",[1535],{"type":37,"value":1536},"Prompt Coverage Analysis",{"type":32,"tag":33,"props":1538,"children":1539},{},[1540],{"type":37,"value":1541},"How does your citation performance vary across intent types? Measure separately: informational (\"What is CDP\"), navigational (\"Shopify CDP integration\"), commercial (\"best CDP vendors\"), transactional (\"request CDP demo\") queries.",{"type":32,"tag":33,"props":1543,"children":1544},{},[1545],{"type":37,"value":1546},"Coverage gap example: In e-commerce, you get 18% citation on informational queries but 3% on commercial. This gap tells your content strategy to add \"vendor comparison,\" \"pricing breakdown,\" \"implementation checklist\" pieces.",{"type":32,"tag":33,"props":1548,"children":1549},{},[1550],{"type":37,"value":1551},"Segmentation table example:",{"type":32,"tag":696,"props":1553,"children":1554},{},[1555,1581],{"type":32,"tag":700,"props":1556,"children":1557},{},[1558],{"type":32,"tag":704,"props":1559,"children":1560},{},[1561,1566,1571,1576],{"type":32,"tag":708,"props":1562,"children":1563},{},[1564],{"type":37,"value":1565},"Intent Type",{"type":32,"tag":708,"props":1567,"children":1568},{},[1569],{"type":37,"value":1570},"Query Count",{"type":32,"tag":708,"props":1572,"children":1573},{},[1574],{"type":37,"value":1575},"Citation Rate",{"type":32,"tag":708,"props":1577,"children":1578},{},[1579],{"type":37,"value":1580},"Avg Position",{"type":32,"tag":724,"props":1582,"children":1583},{},[1584,1607,1630,1653],{"type":32,"tag":704,"props":1585,"children":1586},{},[1587,1592,1597,1602],{"type":32,"tag":731,"props":1588,"children":1589},{},[1590],{"type":37,"value":1591},"Informational",{"type":32,"tag":731,"props":1593,"children":1594},{},[1595],{"type":37,"value":1596},"120",{"type":32,"tag":731,"props":1598,"children":1599},{},[1600],{"type":37,"value":1601},"18%",{"type":32,"tag":731,"props":1603,"children":1604},{},[1605],{"type":37,"value":1606},"2.1",{"type":32,"tag":704,"props":1608,"children":1609},{},[1610,1615,1620,1625],{"type":32,"tag":731,"props":1611,"children":1612},{},[1613],{"type":37,"value":1614},"Commercial",{"type":32,"tag":731,"props":1616,"children":1617},{},[1618],{"type":37,"value":1619},"80",{"type":32,"tag":731,"props":1621,"children":1622},{},[1623],{"type":37,"value":1624},"3%",{"type":32,"tag":731,"props":1626,"children":1627},{},[1628],{"type":37,"value":1629},"4.5",{"type":32,"tag":704,"props":1631,"children":1632},{},[1633,1638,1643,1648],{"type":32,"tag":731,"props":1634,"children":1635},{},[1636],{"type":37,"value":1637},"Navigational",{"type":32,"tag":731,"props":1639,"children":1640},{},[1641],{"type":37,"value":1642},"40",{"type":32,"tag":731,"props":1644,"children":1645},{},[1646],{"type":37,"value":1647},"25%",{"type":32,"tag":731,"props":1649,"children":1650},{},[1651],{"type":37,"value":1652},"1.8",{"type":32,"tag":704,"props":1654,"children":1655},{},[1656,1661,1666,1671],{"type":32,"tag":731,"props":1657,"children":1658},{},[1659],{"type":37,"value":1660},"Transactional",{"type":32,"tag":731,"props":1662,"children":1663},{},[1664],{"type":37,"value":1665},"20",{"type":32,"tag":731,"props":1667,"children":1668},{},[1669],{"type":37,"value":1670},"0%",{"type":32,"tag":731,"props":1672,"children":1673},{},[1674],{"type":37,"value":1675},"N\u002FA",{"type":32,"tag":33,"props":1677,"children":1678},{},[1679],{"type":37,"value":1680},"0% citation on transactional is normal — LLMs can't execute sales directly, so they don't cite sources for \"request demo\" queries. But that low commercial rate is actionable.",{"type":32,"tag":46,"props":1682,"children":1684},{"id":1683},"dashboard-and-alert-system",[1685],{"type":37,"value":1686},"Dashboard and Alert System",{"type":32,"tag":33,"props":1688,"children":1689},{},[1690],{"type":37,"value":1691},"Collecting metrics without reporting creates zero operational value. Weekly citation report template:",{"type":32,"tag":33,"props":1693,"children":1694},{},[1695],{"type":32,"tag":40,"props":1696,"children":1697},{},[1698],{"type":37,"value":1699},"Executive Summary (one slide):",{"type":32,"tag":1701,"props":1702,"children":1703},"ul",{},[1704,1709,1714,1719],{"type":32,"tag":1167,"props":1705,"children":1706},{},[1707],{"type":37,"value":1708},"Overall citation rate trend (last 12 weeks)",{"type":32,"tag":1167,"props":1710,"children":1711},{},[1712],{"type":37,"value":1713},"Model breakdown (ChatGPT\u002FGemini\u002FClaude bar chart)",{"type":32,"tag":1167,"props":1715,"children":1716},{},[1717],{"type":37,"value":1718},"Top 5 cited content pieces",{"type":32,"tag":1167,"props":1720,"children":1721},{},[1722],{"type":37,"value":1723},"Coverage gaps (which intent types underperform)",{"type":32,"tag":33,"props":1725,"children":1726},{},[1727],{"type":32,"tag":40,"props":1728,"children":1729},{},[1730],{"type":37,"value":1731},"Alert rules (Slack\u002Femail):",{"type":32,"tag":1701,"props":1733,"children":1734},{},[1735,1740,1745],{"type":32,"tag":1167,"props":1736,"children":1737},{},[1738],{"type":37,"value":1739},"Citation rate drops below 20% → editorial team review triggered",{"type":32,"tag":1167,"props":1741,"children":1742},{},[1743],{"type":37,"value":1744},"Competitor citation rate exceeds yours (separate tracking pipeline needed) → strategic response plan",{"type":32,"tag":1167,"props":1746,"children":1747},{},[1748],{"type":37,"value":1749},"New high-performing keyword cluster detected → content production prioritized",{"type":32,"tag":33,"props":1751,"children":1752},{},[1753,1755,1762],{"type":37,"value":1754},"These alerts fit within ",{"type":32,"tag":1126,"props":1756,"children":1759},{"href":1757,"rel":1758},"https:\u002F\u002Fwww.roibase.com.tr\u002Fru\u002Fverianalizi",[1130],[1760],{"type":37,"value":1761},"Data Analysis & Insight Engineering",{"type":37,"value":1763}," — raw metrics become actionable signals through data engineering.",{"type":32,"tag":46,"props":1765,"children":1767},{"id":1766},"geo-strategy-connection",[1768],{"type":37,"value":1769},"GEO Strategy Connection",{"type":32,"tag":33,"props":1771,"children":1772},{},[1773],{"type":37,"value":1774},"Citation measurement isn't just reporting; it's input for optimization. If inference share is low, make your content LLM-friendly: chunked paragraphs, clear header hierarchy, higher factual statement density. If citation position is low, strengthen authoritativeness signals: boost backlink quality, leverage domain age, optimize content freshness.",{"type":32,"tag":33,"props":1776,"children":1777},{},[1778],{"type":37,"value":1779},"The difference between GEO and classical SEO: you used to optimize keyword density; now optimize semantic cluster coverage. LLMs look at concept overlap, not n-gram matching — repeating the same keyword 10 times doesn't matter; covering related concepts does.",{"type":32,"tag":1781,"props":1782,"children":1783},"hr",{},[],{"type":32,"tag":33,"props":1785,"children":1786},{},[1787],{"type":37,"value":1788},"LLM citation tracking isn't optional in 2026; it's mandatory. If your brand isn't visible in generative engines, you've been cut out of the next generation's decision process. Citation rate, inference share, prompt coverage — without these three metrics on your dashboard, your SEO strategy is incomplete. Now pick your first 50 keywords, build the pipeline, and get your first weekly snapshot — in 3 months, while competitors are still staring at Google Analytics, you'll be reading real signal from your attribution graph.",{"type":32,"tag":1790,"props":1791,"children":1792},"style",{},[1793],{"type":37,"value":1794},"html .default .shiki span {color: var(--shiki-default);background: var(--shiki-default-bg);font-style: var(--shiki-default-font-style);font-weight: var(--shiki-default-font-weight);text-decoration: var(--shiki-default-text-decoration);}html .shiki span {color: var(--shiki-default);background: var(--shiki-default-bg);font-style: var(--shiki-default-font-style);font-weight: var(--shiki-default-font-weight);text-decoration: var(--shiki-default-text-decoration);}",{"title":16,"searchDepth":271,"depth":271,"links":1796},[1797,1798,1801,1802,1803,1804,1805],{"id":48,"depth":248,"text":51},{"id":89,"depth":248,"text":92,"children":1799},[1800],{"id":150,"depth":271,"text":153},{"id":161,"depth":248,"text":164},{"id":1138,"depth":248,"text":1141},{"id":1533,"depth":248,"text":1536},{"id":1683,"depth":248,"text":1686},{"id":1766,"depth":248,"text":1769},"markdown","content:ru:ai:llm-citation-measurement.md","content","ru\u002Fai\u002Fllm-citation-measurement.md","ru\u002Fai\u002Fllm-citation-measurement","md",1785103546838]