[{"data":1,"prerenderedAt":1771},["ShallowReactive",2],{"article-alternates":3,"article-\u002Fes\u002Fai\u002Fmedicion-de-citacion-llm":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":8,"_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":1765,"_id":1766,"_source":1767,"_file":1768,"_stem":1769,"_extension":1770},"ai",false,"","Medición de Citación en LLM — Tu Nuevo Conjunto de Métricas SEO","¿Cómo medir la tasa de citación de tu marca en Perplexity, ChatGPT y Gemini? Guía para configurar métricas críticas para GEO.","2026-07-25",[21,22,23,24,25],"llm-citation","geo","seo-metrics","ai-search","attribution",8,"Roibase",{"type":29,"children":30,"toc":1752},"root",[31,47,54,66,71,83,89,94,108,154,190,232,253,259,264,271,435,443,462,467,473,485,603,613,619,624,769,774,780,793,798,819,832,876,908,977,989,995,1015,1023,1039,1047,1072,1077,1098,1265,1274,1280,1292,1297,1314,1334,1693,1706,1726,1732,1746],{"type":32,"tag":33,"props":34,"children":35},"element","p",{},[36,39,45],{"type":37,"value":38},"text","El tráfico orgánico cae, el tráfico directo sube en Google Analytics, pero no sabes qué consultas ahora se responden en ChatGPT sin que los usuarios lleguen a tu sitio. A mediados de 2026, los LLM capturaron el 23% del tráfico de búsqueda (datos Q2 2026 de SimilarWeb). En lugar de recuperar ese tráfico, necesitas empezar a medir la tasa de ",{"type":32,"tag":40,"props":41,"children":42},"strong",{},[43],{"type":37,"value":44},"citación de tu marca como fuente",{"type":37,"value":46}," en los LLM. Añade una nueva capa a tus métricas SEO: citation rate, source prominence, retrieval frequency.",{"type":32,"tag":48,"props":49,"children":51},"h2",{"id":50},"qué-es-citación-en-llm-y-por-qué-medirla-ahora",[52],{"type":37,"value":53},"Qué es Citación en LLM y por Qué Medirla Ahora",{"type":32,"tag":33,"props":55,"children":56},{},[57,59,64],{"type":37,"value":58},"La citación en LLM es el porcentaje de veces que un modelo generativo ",{"type":32,"tag":40,"props":60,"children":61},{},[62],{"type":37,"value":63},"referencia tu marca o contenido como fuente",{"type":37,"value":65}," al responder a una consulta de usuario. Si ChatGPT escribe \"Fuente: roibase.com.tr\", Perplexity proporciona un enlace en línea, o Gemini lista tu sitio en una nota al pie, obtuviste una citación.",{"type":32,"tag":33,"props":67,"children":68},{},[69],{"type":37,"value":70},"En el SEO clásico había \"ranking\" — estar en la posición 3 en Google. En la era de los LLM hay \"citation prominence\" — si el modelo muestra 4 fuentes, ¿cuál es tu participación? ¿Eres la primera fuente o estás al final de una lista de \"fuentes relacionadas\"? Esta diferencia puede cambiar la tasa de clics en un 300% (datos internos de Perplexity Labs, Q1 2026).",{"type":32,"tag":33,"props":72,"children":73},{},[74,76,81],{"type":37,"value":75},"Si no empiezas a medir ahora, no podrás establecer una línea base. En 6 meses no podrás responder \"¿funcionó nuestra estrategia GEO?\" El primer paso es crear un ",{"type":32,"tag":40,"props":77,"children":78},{},[79],{"type":37,"value":80},"conjunto de consultas sintéticas",{"type":37,"value":82}," e interrogar regularmente los LLM.",{"type":32,"tag":48,"props":84,"children":86},{"id":85},"configurar-la-arquitectura-de-medición-pipeline-de-consultas-sintéticas",[87],{"type":37,"value":88},"Configurar la Arquitectura de Medición: Pipeline de Consultas Sintéticas",{"type":32,"tag":33,"props":90,"children":91},{},[92],{"type":37,"value":93},"Medir citación en LLM no es suficiente con pruebas manuales. Necesitas hacer las mismas 50-100 preguntas a Perplexity, ChatGPT y Gemini cada día, analizando las referencias de fuentes en las respuestas. Hacemos esto con un pipeline de 3 capas:",{"type":32,"tag":33,"props":95,"children":96},{},[97,102,106],{"type":32,"tag":40,"props":98,"children":99},{},[100],{"type":37,"value":101},"Capa 1: Diseño del Conjunto de Consultas",{"type":32,"tag":103,"props":104,"children":105},"br",{},[],{"type":37,"value":107},"\nExtrae de GSC las consultas de los últimos 90 días con impresiones, posición 1-20 y CTR por debajo del 5%. Estas consultas significan \"aparecemos en Google pero no nos hacen clic\" — es probable que los LLM ya estén respondiendo. Selecciona 50-100 consultas. No solo branded queries, sino una mezcla de informativas y transaccionales. Ejemplo: \"duración de cookie en server-side GTM\", \"optimización de costos en BigQuery\".",{"type":32,"tag":33,"props":109,"children":110},{},[111,116,119,121,128,130,136,138,144,146,152],{"type":32,"tag":40,"props":112,"children":113},{},[114],{"type":37,"value":115},"Capa 2: Interrogación Automatizada",{"type":32,"tag":103,"props":117,"children":118},{},[],{"type":37,"value":120},"\nUsa un workflow de n8n para interrogar la API de cada LLM una vez al día. Perplexity con parámetro ",{"type":32,"tag":122,"props":123,"children":125},"code",{"className":124},[],[126],{"type":37,"value":127},"model: sonar-pro",{"type":37,"value":129},", ChatGPT en modo ",{"type":32,"tag":122,"props":131,"children":133},{"className":132},[],[134],{"type":37,"value":135},"browsing: true",{"type":37,"value":137},", Gemini con ",{"type":32,"tag":122,"props":139,"children":141},{"className":140},[],[142],{"type":37,"value":143},"grounding: web",{"type":37,"value":145},". Guarda la respuesta como JSON — tanto el cuerpo como el array de ",{"type":32,"tag":122,"props":147,"children":149},{"className":148},[],[150],{"type":37,"value":151},"sources",{"type":37,"value":153},". Importante: gestiona los límites de tasa (Perplexity free tier 5 req\u002Fmin, ChatGPT Plus 40 req\u002F3 horas).",{"type":32,"tag":33,"props":155,"children":156},{},[157,162,165,167,172,174,180,182,188],{"type":32,"tag":40,"props":158,"children":159},{},[160],{"type":37,"value":161},"Capa 3: Parser de Citaciones",{"type":32,"tag":103,"props":163,"children":164},{},[],{"type":37,"value":166},"\nEn el JSON de respuesta, busca el key ",{"type":32,"tag":122,"props":168,"children":170},{"className":169},[],[171],{"type":37,"value":151},{"type":37,"value":173}," — si existe, itera el array y compara dominios (",{"type":32,"tag":122,"props":175,"children":177},{"className":176},[],[178],{"type":37,"value":179},"roibase.com.tr",{"type":37,"value":181}," o subdominio). Si no hay sources, busca en el cuerpo enlaces en línea (",{"type":32,"tag":122,"props":183,"children":185},{"className":184},[],[186],{"type":37,"value":187},"[roibase](...)",{"type":37,"value":189},") o URL simples (con regex). Para cada consulta, registra 3 métricas:",{"type":32,"tag":191,"props":192,"children":193},"ol",{},[194,205,222],{"type":32,"tag":195,"props":196,"children":197},"li",{},[198,203],{"type":32,"tag":40,"props":199,"children":200},{},[201],{"type":37,"value":202},"¿Hay citación:",{"type":37,"value":204}," booleano (0\u002F1)",{"type":32,"tag":195,"props":206,"children":207},{},[208,213,215,220],{"type":32,"tag":40,"props":209,"children":210},{},[211],{"type":37,"value":212},"Rango:",{"type":37,"value":214}," en qué posición del array ",{"type":32,"tag":122,"props":216,"children":218},{"className":217},[],[219],{"type":37,"value":151},{"type":37,"value":221}," (1-5, nulo si ausente)",{"type":32,"tag":195,"props":223,"children":224},{},[225,230],{"type":32,"tag":40,"props":226,"children":227},{},[228],{"type":37,"value":229},"Prominencia:",{"type":37,"value":231}," ¿está en línea o solo en nota al pie? (en línea = 2, nota al pie = 1, ausente = 0)",{"type":32,"tag":33,"props":233,"children":234},{},[235,237,243,245,251],{"type":37,"value":236},"Escribe estos datos en BigQuery en la tabla ",{"type":32,"tag":122,"props":238,"children":240},{"className":239},[],[241],{"type":37,"value":242},"llm_citations",{"type":37,"value":244}," — esquema: ",{"type":32,"tag":122,"props":246,"children":248},{"className":247},[],[249],{"type":37,"value":250},"query_id, llm_provider, date, cited, rank, prominence",{"type":37,"value":252},".",{"type":32,"tag":48,"props":254,"children":256},{"id":255},"calcular-citation-rate-y-benchmark",[257],{"type":37,"value":258},"Calcular Citation Rate y Benchmark",{"type":32,"tag":33,"props":260,"children":261},{},[262],{"type":37,"value":263},"Si ejecutas 50 consultas una vez al día, 30 días, con 3 LLM, tienes 50 × 3 × 30 = 4.500 filas de datos. Ahora calcula las métricas:",{"type":32,"tag":265,"props":266,"children":268},"h3",{"id":267},"_1-citation-rate-general",[269],{"type":37,"value":270},"1. Citation Rate General",{"type":32,"tag":272,"props":273,"children":277},"pre",{"className":274,"code":275,"language":276,"meta":16,"style":16},"language-sql shiki shiki-themes github-dark","SELECT \n  llm_provider,\n  COUNTIF(cited = 1) \u002F COUNT(*) AS citation_rate\nFROM `project.dataset.llm_citations`\nWHERE date >= DATE_SUB(CURRENT_DATE(), INTERVAL 30 DAY)\nGROUP BY llm_provider;\n","sql",[278],{"type":32,"tag":122,"props":279,"children":280},{"__ignoreMap":16},[281,299,308,367,382,421],{"type":32,"tag":282,"props":283,"children":286},"span",{"class":284,"line":285},"line",1,[287,293],{"type":32,"tag":282,"props":288,"children":290},{"style":289},"--shiki-default:#F97583",[291],{"type":37,"value":292},"SELECT",{"type":32,"tag":282,"props":294,"children":296},{"style":295},"--shiki-default:#E1E4E8",[297],{"type":37,"value":298}," \n",{"type":32,"tag":282,"props":300,"children":302},{"class":284,"line":301},2,[303],{"type":32,"tag":282,"props":304,"children":305},{"style":295},[306],{"type":37,"value":307},"  llm_provider,\n",{"type":32,"tag":282,"props":309,"children":311},{"class":284,"line":310},3,[312,317,322,328,333,338,343,348,353,357,362],{"type":32,"tag":282,"props":313,"children":314},{"style":295},[315],{"type":37,"value":316},"  COUNTIF(cited ",{"type":32,"tag":282,"props":318,"children":319},{"style":289},[320],{"type":37,"value":321},"=",{"type":32,"tag":282,"props":323,"children":325},{"style":324},"--shiki-default:#79B8FF",[326],{"type":37,"value":327}," 1",{"type":32,"tag":282,"props":329,"children":330},{"style":295},[331],{"type":37,"value":332},") ",{"type":32,"tag":282,"props":334,"children":335},{"style":289},[336],{"type":37,"value":337},"\u002F",{"type":32,"tag":282,"props":339,"children":340},{"style":324},[341],{"type":37,"value":342}," COUNT",{"type":32,"tag":282,"props":344,"children":345},{"style":295},[346],{"type":37,"value":347},"(",{"type":32,"tag":282,"props":349,"children":350},{"style":289},[351],{"type":37,"value":352},"*",{"type":32,"tag":282,"props":354,"children":355},{"style":295},[356],{"type":37,"value":332},{"type":32,"tag":282,"props":358,"children":359},{"style":289},[360],{"type":37,"value":361},"AS",{"type":32,"tag":282,"props":363,"children":364},{"style":295},[365],{"type":37,"value":366}," citation_rate\n",{"type":32,"tag":282,"props":368,"children":370},{"class":284,"line":369},4,[371,376],{"type":32,"tag":282,"props":372,"children":373},{"style":289},[374],{"type":37,"value":375},"FROM",{"type":32,"tag":282,"props":377,"children":379},{"style":378},"--shiki-default:#9ECBFF",[380],{"type":37,"value":381}," `project.dataset.llm_citations`\n",{"type":32,"tag":282,"props":383,"children":385},{"class":284,"line":384},5,[386,391,396,401,406,411,416],{"type":32,"tag":282,"props":387,"children":388},{"style":289},[389],{"type":37,"value":390},"WHERE",{"type":32,"tag":282,"props":392,"children":393},{"style":289},[394],{"type":37,"value":395}," date",{"type":32,"tag":282,"props":397,"children":398},{"style":289},[399],{"type":37,"value":400}," >=",{"type":32,"tag":282,"props":402,"children":403},{"style":295},[404],{"type":37,"value":405}," DATE_SUB(CURRENT_DATE(), INTERVAL ",{"type":32,"tag":282,"props":407,"children":408},{"style":324},[409],{"type":37,"value":410},"30",{"type":32,"tag":282,"props":412,"children":413},{"style":289},[414],{"type":37,"value":415}," DAY",{"type":32,"tag":282,"props":417,"children":418},{"style":295},[419],{"type":37,"value":420},")\n",{"type":32,"tag":282,"props":422,"children":424},{"class":284,"line":423},6,[425,430],{"type":32,"tag":282,"props":426,"children":427},{"style":289},[428],{"type":37,"value":429},"GROUP BY",{"type":32,"tag":282,"props":431,"children":432},{"style":295},[433],{"type":37,"value":434}," llm_provider;\n",{"type":32,"tag":33,"props":436,"children":437},{},[438],{"type":32,"tag":40,"props":439,"children":440},{},[441],{"type":37,"value":442},"Benchmark (Q2 2026, promedio B2B SaaS):",{"type":32,"tag":444,"props":445,"children":446},"ul",{},[447,452,457],{"type":32,"tag":195,"props":448,"children":449},{},[450],{"type":37,"value":451},"Perplexity: 18-24%",{"type":32,"tag":195,"props":453,"children":454},{},[455],{"type":37,"value":456},"ChatGPT browsing: 12-16%",{"type":32,"tag":195,"props":458,"children":459},{},[460],{"type":37,"value":461},"Gemini grounding: 8-14%",{"type":32,"tag":33,"props":463,"children":464},{},[465],{"type":37,"value":466},"Si estás por debajo del 12% en Perplexity, hay una deficiencia GEO — tu contenido no está estructurado para retrieval.",{"type":32,"tag":265,"props":468,"children":470},{"id":469},"_2-primary-source-rate",[471],{"type":37,"value":472},"2. Primary Source Rate",{"type":32,"tag":33,"props":474,"children":475},{},[476,478,483],{"type":37,"value":477},"Cuántas veces que te citan eres la ",{"type":32,"tag":40,"props":479,"children":480},{},[481],{"type":37,"value":482},"primera fuente",{"type":37,"value":484},":",{"type":32,"tag":272,"props":486,"children":488},{"className":274,"code":487,"language":276,"meta":16,"style":16},"SELECT \n  llm_provider,\n  COUNTIF(rank = 1) \u002F COUNTIF(cited = 1) AS primary_rate\nFROM `project.dataset.llm_citations`\nWHERE cited = 1\nGROUP BY llm_provider;\n",[489],{"type":32,"tag":122,"props":490,"children":491},{"__ignoreMap":16},[492,503,510,560,571,592],{"type":32,"tag":282,"props":493,"children":494},{"class":284,"line":285},[495,499],{"type":32,"tag":282,"props":496,"children":497},{"style":289},[498],{"type":37,"value":292},{"type":32,"tag":282,"props":500,"children":501},{"style":295},[502],{"type":37,"value":298},{"type":32,"tag":282,"props":504,"children":505},{"class":284,"line":301},[506],{"type":32,"tag":282,"props":507,"children":508},{"style":295},[509],{"type":37,"value":307},{"type":32,"tag":282,"props":511,"children":512},{"class":284,"line":310},[513,518,522,526,530,534,539,543,547,551,555],{"type":32,"tag":282,"props":514,"children":515},{"style":295},[516],{"type":37,"value":517},"  COUNTIF(rank ",{"type":32,"tag":282,"props":519,"children":520},{"style":289},[521],{"type":37,"value":321},{"type":32,"tag":282,"props":523,"children":524},{"style":324},[525],{"type":37,"value":327},{"type":32,"tag":282,"props":527,"children":528},{"style":295},[529],{"type":37,"value":332},{"type":32,"tag":282,"props":531,"children":532},{"style":289},[533],{"type":37,"value":337},{"type":32,"tag":282,"props":535,"children":536},{"style":295},[537],{"type":37,"value":538}," COUNTIF(cited ",{"type":32,"tag":282,"props":540,"children":541},{"style":289},[542],{"type":37,"value":321},{"type":32,"tag":282,"props":544,"children":545},{"style":324},[546],{"type":37,"value":327},{"type":32,"tag":282,"props":548,"children":549},{"style":295},[550],{"type":37,"value":332},{"type":32,"tag":282,"props":552,"children":553},{"style":289},[554],{"type":37,"value":361},{"type":32,"tag":282,"props":556,"children":557},{"style":295},[558],{"type":37,"value":559}," primary_rate\n",{"type":32,"tag":282,"props":561,"children":562},{"class":284,"line":369},[563,567],{"type":32,"tag":282,"props":564,"children":565},{"style":289},[566],{"type":37,"value":375},{"type":32,"tag":282,"props":568,"children":569},{"style":378},[570],{"type":37,"value":381},{"type":32,"tag":282,"props":572,"children":573},{"class":284,"line":384},[574,578,583,587],{"type":32,"tag":282,"props":575,"children":576},{"style":289},[577],{"type":37,"value":390},{"type":32,"tag":282,"props":579,"children":580},{"style":295},[581],{"type":37,"value":582}," cited ",{"type":32,"tag":282,"props":584,"children":585},{"style":289},[586],{"type":37,"value":321},{"type":32,"tag":282,"props":588,"children":589},{"style":324},[590],{"type":37,"value":591}," 1\n",{"type":32,"tag":282,"props":593,"children":594},{"class":284,"line":423},[595,599],{"type":32,"tag":282,"props":596,"children":597},{"style":289},[598],{"type":37,"value":429},{"type":32,"tag":282,"props":600,"children":601},{"style":295},[602],{"type":37,"value":434},{"type":32,"tag":33,"props":604,"children":605},{},[606,611],{"type":32,"tag":40,"props":607,"children":608},{},[609],{"type":37,"value":610},"Meta:",{"type":37,"value":612}," 40%+ (si te citan, deberías ser la primera fuente en 4 de cada 10 casos). Por debajo del 20% significa que la \"relevance signal\" es débil — probablemente baja similaridad de embedding en el retrieval.",{"type":32,"tag":265,"props":614,"children":616},{"id":615},"_3-volatilidad-a-nivel-de-consulta",[617],{"type":37,"value":618},"3. Volatilidad a Nivel de Consulta",{"type":32,"tag":33,"props":620,"children":621},{},[622],{"type":37,"value":623},"Para cada consulta, calcula la varianza de citación en 30 días — si siempre te citan, la volatilidad es baja; si a veces sí y a veces no, es alta. Alta volatilidad significa que el LLM actualiza su índice frecuentemente o el contenido competidor te ha dejado atrás.",{"type":32,"tag":272,"props":625,"children":627},{"className":274,"code":626,"language":276,"meta":16,"style":16},"SELECT \n  query_id,\n  STDDEV(cited) AS citation_volatility\nFROM `project.dataset.llm_citations`\nWHERE llm_provider = 'perplexity'\nGROUP BY query_id\nHAVING COUNT(*) >= 20\nORDER BY citation_volatility DESC;\n",[628],{"type":32,"tag":122,"props":629,"children":630},{"__ignoreMap":16},[631,642,650,667,678,699,711,746],{"type":32,"tag":282,"props":632,"children":633},{"class":284,"line":285},[634,638],{"type":32,"tag":282,"props":635,"children":636},{"style":289},[637],{"type":37,"value":292},{"type":32,"tag":282,"props":639,"children":640},{"style":295},[641],{"type":37,"value":298},{"type":32,"tag":282,"props":643,"children":644},{"class":284,"line":301},[645],{"type":32,"tag":282,"props":646,"children":647},{"style":295},[648],{"type":37,"value":649},"  query_id,\n",{"type":32,"tag":282,"props":651,"children":652},{"class":284,"line":310},[653,658,662],{"type":32,"tag":282,"props":654,"children":655},{"style":295},[656],{"type":37,"value":657},"  STDDEV(cited) ",{"type":32,"tag":282,"props":659,"children":660},{"style":289},[661],{"type":37,"value":361},{"type":32,"tag":282,"props":663,"children":664},{"style":295},[665],{"type":37,"value":666}," citation_volatility\n",{"type":32,"tag":282,"props":668,"children":669},{"class":284,"line":369},[670,674],{"type":32,"tag":282,"props":671,"children":672},{"style":289},[673],{"type":37,"value":375},{"type":32,"tag":282,"props":675,"children":676},{"style":378},[677],{"type":37,"value":381},{"type":32,"tag":282,"props":679,"children":680},{"class":284,"line":384},[681,685,690,694],{"type":32,"tag":282,"props":682,"children":683},{"style":289},[684],{"type":37,"value":390},{"type":32,"tag":282,"props":686,"children":687},{"style":295},[688],{"type":37,"value":689}," llm_provider ",{"type":32,"tag":282,"props":691,"children":692},{"style":289},[693],{"type":37,"value":321},{"type":32,"tag":282,"props":695,"children":696},{"style":378},[697],{"type":37,"value":698}," 'perplexity'\n",{"type":32,"tag":282,"props":700,"children":701},{"class":284,"line":423},[702,706],{"type":32,"tag":282,"props":703,"children":704},{"style":289},[705],{"type":37,"value":429},{"type":32,"tag":282,"props":707,"children":708},{"style":295},[709],{"type":37,"value":710}," query_id\n",{"type":32,"tag":282,"props":712,"children":714},{"class":284,"line":713},7,[715,720,724,728,732,736,741],{"type":32,"tag":282,"props":716,"children":717},{"style":289},[718],{"type":37,"value":719},"HAVING",{"type":32,"tag":282,"props":721,"children":722},{"style":324},[723],{"type":37,"value":342},{"type":32,"tag":282,"props":725,"children":726},{"style":295},[727],{"type":37,"value":347},{"type":32,"tag":282,"props":729,"children":730},{"style":289},[731],{"type":37,"value":352},{"type":32,"tag":282,"props":733,"children":734},{"style":295},[735],{"type":37,"value":332},{"type":32,"tag":282,"props":737,"children":738},{"style":289},[739],{"type":37,"value":740},">=",{"type":32,"tag":282,"props":742,"children":743},{"style":324},[744],{"type":37,"value":745}," 20\n",{"type":32,"tag":282,"props":747,"children":748},{"class":284,"line":26},[749,754,759,764],{"type":32,"tag":282,"props":750,"children":751},{"style":289},[752],{"type":37,"value":753},"ORDER BY",{"type":32,"tag":282,"props":755,"children":756},{"style":295},[757],{"type":37,"value":758}," citation_volatility ",{"type":32,"tag":282,"props":760,"children":761},{"style":289},[762],{"type":37,"value":763},"DESC",{"type":32,"tag":282,"props":765,"children":766},{"style":295},[767],{"type":37,"value":768},";\n",{"type":32,"tag":33,"props":770,"children":771},{},[772],{"type":37,"value":773},"Si volatility > 0.4, revisa manualmente — probablemente hay un problema de \"freshness\" (tu contenido se publicó hace 6 meses, el LLM prefiere contenido más reciente).",{"type":32,"tag":48,"props":775,"children":777},{"id":776},"tradeoff-de-atribución-tráfico-directo-o-referencia-de-llm",[778],{"type":37,"value":779},"Tradeoff de Atribución: ¿Tráfico Directo o Referencia de LLM?",{"type":32,"tag":33,"props":781,"children":782},{},[783,785,791],{"type":37,"value":784},"Hay un efecto secundario de obtener citaciones en LLM: en Google Analytics sube el tráfico directo pero no puedes saber que vino de un LLM. Porque los clics desde la interfaz web de ChatGPT aparecen como ",{"type":32,"tag":122,"props":786,"children":788},{"className":787},[],[789],{"type":37,"value":790},"(direct) \u002F (none)",{"type":37,"value":792}," — sin header de referrer.",{"type":32,"tag":33,"props":794,"children":795},{},[796],{"type":37,"value":797},"Para resolver esto, hay 2 métodos:",{"type":32,"tag":33,"props":799,"children":800},{},[801,806,809,811,817],{"type":32,"tag":40,"props":802,"children":803},{},[804],{"type":37,"value":805},"Método 1: Inyección de UTM (en API de LLM)",{"type":32,"tag":103,"props":807,"children":808},{},[],{"type":37,"value":810},"\nSi envías contenido a la API de un LLM (por ejemplo, Perplexity Publisher API), añade ",{"type":32,"tag":122,"props":812,"children":814},{"className":813},[],[815],{"type":37,"value":816},"?utm_source=perplexity&utm_medium=llm&utm_campaign=citation",{"type":37,"value":818}," a tus URLs. Así el source aparece en GA4. Pero este método solo funciona en LLM que usan APIs — en ChatGPT web crawling no puedes inyectar UTM.",{"type":32,"tag":33,"props":820,"children":821},{},[822,827,830],{"type":32,"tag":40,"props":823,"children":824},{},[825],{"type":37,"value":826},"Método 2: Fingerprinting del Lado del Servidor",{"type":32,"tag":103,"props":828,"children":829},{},[],{"type":37,"value":831},"\nLos bots de LLM usan patrones específicos de user-agent:",{"type":32,"tag":444,"props":833,"children":834},{},[835,846,865],{"type":32,"tag":195,"props":836,"children":837},{},[838,840],{"type":37,"value":839},"Perplexity: ",{"type":32,"tag":122,"props":841,"children":843},{"className":842},[],[844],{"type":37,"value":845},"PerplexityBot",{"type":32,"tag":195,"props":847,"children":848},{},[849,851,857,859],{"type":37,"value":850},"ChatGPT: ",{"type":32,"tag":122,"props":852,"children":854},{"className":853},[],[855],{"type":37,"value":856},"ChatGPT-User",{"type":37,"value":858}," o ",{"type":32,"tag":122,"props":860,"children":862},{"className":861},[],[863],{"type":37,"value":864},"GPTBot",{"type":32,"tag":195,"props":866,"children":867},{},[868,870],{"type":37,"value":869},"Gemini: ",{"type":32,"tag":122,"props":871,"children":873},{"className":872},[],[874],{"type":37,"value":875},"Google-Extended",{"type":32,"tag":33,"props":877,"children":878},{},[879,881,890,892,898,900,906],{"type":37,"value":880},"Filtra estos user-agent en los logs del servidor e integra con ",{"type":32,"tag":882,"props":883,"children":887},"a",{"href":884,"rel":885},"https:\u002F\u002Fwww.roibase.com.tr\u002Fes\u002Ffirstparty",[886],"nofollow",[888],{"type":37,"value":889},"Arquitectura de Medición First-Party",{"type":37,"value":891}," para enviar eventos a GA4 del lado del servidor. Nombre del evento: ",{"type":32,"tag":122,"props":893,"children":895},{"className":894},[],[896],{"type":37,"value":897},"llm_visit",{"type":37,"value":899},", parámetro: ",{"type":32,"tag":122,"props":901,"children":903},{"className":902},[],[904],{"type":37,"value":905},"llm_provider",{"type":37,"value":907},". De esta forma puedes distinguir los LLM dentro del tráfico \"directo\".",{"type":32,"tag":909,"props":910,"children":911},"table",{},[912,936],{"type":32,"tag":913,"props":914,"children":915},"thead",{},[916],{"type":32,"tag":917,"props":918,"children":919},"tr",{},[920,926,931],{"type":32,"tag":921,"props":922,"children":923},"th",{},[924],{"type":37,"value":925},"Método",{"type":32,"tag":921,"props":927,"children":928},{},[929],{"type":37,"value":930},"Ventaja",{"type":32,"tag":921,"props":932,"children":933},{},[934],{"type":37,"value":935},"Desventaja",{"type":32,"tag":937,"props":938,"children":939},"tbody",{},[940,959],{"type":32,"tag":917,"props":941,"children":942},{},[943,949,954],{"type":32,"tag":944,"props":945,"children":946},"td",{},[947],{"type":37,"value":948},"Inyección de UTM",{"type":32,"tag":944,"props":950,"children":951},{},[952],{"type":37,"value":953},"Source automático en GA4",{"type":32,"tag":944,"props":955,"children":956},{},[957],{"type":37,"value":958},"Solo en APIs",{"type":32,"tag":917,"props":960,"children":961},{},[962,967,972],{"type":32,"tag":944,"props":963,"children":964},{},[965],{"type":37,"value":966},"Fingerprint Servidor",{"type":32,"tag":944,"props":968,"children":969},{},[970],{"type":37,"value":971},"Funciona en todos los LLM",{"type":32,"tag":944,"props":973,"children":974},{},[975],{"type":37,"value":976},"Requiere parsing de logs",{"type":32,"tag":33,"props":978,"children":979},{},[980,982,987],{"type":37,"value":981},"Sea cuál sea el que elijas, el objetivo es: ",{"type":32,"tag":40,"props":983,"children":984},{},[985],{"type":37,"value":986},"ver la correlación entre citation rate y tráfico referido de LLM",{"type":37,"value":988},". Si la citación sube el 20% pero el tráfico LLM no sube, significa que aunque te citan, los usuarios no hacen clic — problema de prominencia o calidad del snippet.",{"type":32,"tag":48,"props":990,"children":992},{"id":991},"prominencia-de-citación-diferencia-inline-vs-nota-al-pie",[993],{"type":37,"value":994},"Prominencia de Citación: Diferencia Inline vs Nota al Pie",{"type":32,"tag":33,"props":996,"children":997},{},[998,1000,1005,1007,1013],{"type":37,"value":999},"El LLM te citó, pero ",{"type":32,"tag":40,"props":1001,"children":1002},{},[1003],{"type":37,"value":1004},"¿cómo?",{"type":37,"value":1006}," ¿Perplexity te dio un enlace en línea (con ",{"type":32,"tag":122,"props":1008,"children":1010},{"className":1009},[],[1011],{"type":37,"value":1012},"[1]",{"type":37,"value":1014}," dentro de la frase), o solo apareces en una lista \"Fuentes Relacionadas\" al final? Esta diferencia afecta el CTR en un 400% (test A\u002FB interno de Roibase, n=2.300 consultas).",{"type":32,"tag":33,"props":1016,"children":1017},{},[1018],{"type":32,"tag":40,"props":1019,"children":1020},{},[1021],{"type":37,"value":1022},"Ejemplo de citación inline:",{"type":32,"tag":1024,"props":1025,"children":1026},"blockquote",{},[1027],{"type":32,"tag":33,"props":1028,"children":1029},{},[1030,1032,1037],{"type":37,"value":1031},"\"La duración de cookie en server-side GTM se puede aumentar a 730 días ",{"type":32,"tag":882,"props":1033,"children":1035},{"href":1034},"roibase.com.tr\u002F...",[1036],{"type":37,"value":1012},{"type":37,"value":1038},".\"",{"type":32,"tag":33,"props":1040,"children":1041},{},[1042],{"type":32,"tag":40,"props":1043,"children":1044},{},[1045],{"type":37,"value":1046},"Ejemplo de citación en nota al pie:",{"type":32,"tag":1024,"props":1048,"children":1049},{},[1050,1060],{"type":32,"tag":33,"props":1051,"children":1052},{},[1053,1055,1058],{"type":37,"value":1054},"\"...hay varios métodos disponibles.",{"type":32,"tag":103,"props":1056,"children":1057},{},[],{"type":37,"value":1059},"\nFuentes:",{"type":32,"tag":191,"props":1061,"children":1062},{},[1063,1067],{"type":32,"tag":195,"props":1064,"children":1065},{},[1066],{"type":37,"value":1034},{"type":32,"tag":195,"props":1068,"children":1069},{},[1070],{"type":37,"value":1071},"competitor.com\u002F...\"",{"type":32,"tag":33,"props":1073,"children":1074},{},[1075],{"type":37,"value":1076},"En citación inline, el usuario hace clic mientras lee la frase — hay contexto. En nota al pie, solo hace clic si busca \"más detalles\" después de leer la respuesta — intent de conversión bajo.",{"type":32,"tag":33,"props":1078,"children":1079},{},[1080,1085,1088,1090,1096],{"type":32,"tag":40,"props":1081,"children":1082},{},[1083],{"type":37,"value":1084},"Cálculo de prominence score:",{"type":32,"tag":103,"props":1086,"children":1087},{},[],{"type":37,"value":1089},"\nRegistra ",{"type":32,"tag":122,"props":1091,"children":1093},{"className":1092},[],[1094],{"type":37,"value":1095},"position_type",{"type":37,"value":1097}," cada vez que te citen (inline \u002F footnote \u002F sidebar). Toma el promedio de 30 días:",{"type":32,"tag":272,"props":1099,"children":1101},{"className":274,"code":1100,"language":276,"meta":16,"style":16},"SELECT \n  AVG(CASE \n    WHEN position_type = 'inline' THEN 3\n    WHEN position_type = 'footnote' THEN 1\n    ELSE 0\n  END) AS avg_prominence_score\nFROM `project.dataset.llm_citations`\nWHERE cited = 1;\n",[1102],{"type":32,"tag":122,"props":1103,"children":1104},{"__ignoreMap":16},[1105,1116,1137,1169,1197,1210,1231,1242],{"type":32,"tag":282,"props":1106,"children":1107},{"class":284,"line":285},[1108,1112],{"type":32,"tag":282,"props":1109,"children":1110},{"style":289},[1111],{"type":37,"value":292},{"type":32,"tag":282,"props":1113,"children":1114},{"style":295},[1115],{"type":37,"value":298},{"type":32,"tag":282,"props":1117,"children":1118},{"class":284,"line":301},[1119,1124,1128,1133],{"type":32,"tag":282,"props":1120,"children":1121},{"style":324},[1122],{"type":37,"value":1123},"  AVG",{"type":32,"tag":282,"props":1125,"children":1126},{"style":295},[1127],{"type":37,"value":347},{"type":32,"tag":282,"props":1129,"children":1130},{"style":289},[1131],{"type":37,"value":1132},"CASE",{"type":32,"tag":282,"props":1134,"children":1135},{"style":295},[1136],{"type":37,"value":298},{"type":32,"tag":282,"props":1138,"children":1139},{"class":284,"line":310},[1140,1145,1150,1154,1159,1164],{"type":32,"tag":282,"props":1141,"children":1142},{"style":289},[1143],{"type":37,"value":1144},"    WHEN",{"type":32,"tag":282,"props":1146,"children":1147},{"style":295},[1148],{"type":37,"value":1149}," position_type ",{"type":32,"tag":282,"props":1151,"children":1152},{"style":289},[1153],{"type":37,"value":321},{"type":32,"tag":282,"props":1155,"children":1156},{"style":378},[1157],{"type":37,"value":1158}," 'inline'",{"type":32,"tag":282,"props":1160,"children":1161},{"style":289},[1162],{"type":37,"value":1163}," THEN",{"type":32,"tag":282,"props":1165,"children":1166},{"style":324},[1167],{"type":37,"value":1168}," 3\n",{"type":32,"tag":282,"props":1170,"children":1171},{"class":284,"line":369},[1172,1176,1180,1184,1189,1193],{"type":32,"tag":282,"props":1173,"children":1174},{"style":289},[1175],{"type":37,"value":1144},{"type":32,"tag":282,"props":1177,"children":1178},{"style":295},[1179],{"type":37,"value":1149},{"type":32,"tag":282,"props":1181,"children":1182},{"style":289},[1183],{"type":37,"value":321},{"type":32,"tag":282,"props":1185,"children":1186},{"style":378},[1187],{"type":37,"value":1188}," 'footnote'",{"type":32,"tag":282,"props":1190,"children":1191},{"style":289},[1192],{"type":37,"value":1163},{"type":32,"tag":282,"props":1194,"children":1195},{"style":324},[1196],{"type":37,"value":591},{"type":32,"tag":282,"props":1198,"children":1199},{"class":284,"line":384},[1200,1205],{"type":32,"tag":282,"props":1201,"children":1202},{"style":289},[1203],{"type":37,"value":1204},"    ELSE",{"type":32,"tag":282,"props":1206,"children":1207},{"style":324},[1208],{"type":37,"value":1209}," 0\n",{"type":32,"tag":282,"props":1211,"children":1212},{"class":284,"line":423},[1213,1218,1222,1226],{"type":32,"tag":282,"props":1214,"children":1215},{"style":289},[1216],{"type":37,"value":1217},"  END",{"type":32,"tag":282,"props":1219,"children":1220},{"style":295},[1221],{"type":37,"value":332},{"type":32,"tag":282,"props":1223,"children":1224},{"style":289},[1225],{"type":37,"value":361},{"type":32,"tag":282,"props":1227,"children":1228},{"style":295},[1229],{"type":37,"value":1230}," avg_prominence_score\n",{"type":32,"tag":282,"props":1232,"children":1233},{"class":284,"line":713},[1234,1238],{"type":32,"tag":282,"props":1235,"children":1236},{"style":289},[1237],{"type":37,"value":375},{"type":32,"tag":282,"props":1239,"children":1240},{"style":378},[1241],{"type":37,"value":381},{"type":32,"tag":282,"props":1243,"children":1244},{"class":284,"line":26},[1245,1249,1253,1257,1261],{"type":32,"tag":282,"props":1246,"children":1247},{"style":289},[1248],{"type":37,"value":390},{"type":32,"tag":282,"props":1250,"children":1251},{"style":295},[1252],{"type":37,"value":582},{"type":32,"tag":282,"props":1254,"children":1255},{"style":289},[1256],{"type":37,"value":321},{"type":32,"tag":282,"props":1258,"children":1259},{"style":324},[1260],{"type":37,"value":327},{"type":32,"tag":282,"props":1262,"children":1263},{"style":295},[1264],{"type":37,"value":768},{"type":32,"tag":33,"props":1266,"children":1267},{},[1268,1272],{"type":32,"tag":40,"props":1269,"children":1270},{},[1271],{"type":37,"value":610},{"type":37,"value":1273}," 2.0+ (si te citan, más de la mitad debería ser inline). Por debajo de 1.5 significa que el LLM te ve como \"fuente complementaria\", no \"fuente principal\". Solución: estructura tu contenido para que el LLM pueda citar directamente — definiciones de una línea, fact boxes, snippets de código.",{"type":32,"tag":48,"props":1275,"children":1277},{"id":1276},"análisis-de-competencia-solapamiento-de-fuentes-a-nivel-de-consulta",[1278],{"type":37,"value":1279},"Análisis de Competencia: Solapamiento de Fuentes a Nivel de Consulta",{"type":32,"tag":33,"props":1281,"children":1282},{},[1283,1285,1290],{"type":37,"value":1284},"¿En qué consultas no te citan pero tus competidores sí? Para verlo, parsea ",{"type":32,"tag":40,"props":1286,"children":1287},{},[1288],{"type":37,"value":1289},"todas las fuentes",{"type":37,"value":1291}," que muestra el LLM en cada respuesta, no solo la tuya.",{"type":32,"tag":33,"props":1293,"children":1294},{},[1295],{"type":37,"value":1296},"Ejemplo: en la consulta \"optimización de costos en BigQuery\", Perplexity muestra estas fuentes:",{"type":32,"tag":191,"props":1298,"children":1299},{},[1300,1305,1309],{"type":32,"tag":195,"props":1301,"children":1302},{},[1303],{"type":37,"value":1304},"competitor-a.com",{"type":32,"tag":195,"props":1306,"children":1307},{},[1308],{"type":37,"value":179},{"type":32,"tag":195,"props":1310,"children":1311},{},[1312],{"type":37,"value":1313},"competitor-b.com",{"type":32,"tag":33,"props":1315,"children":1316},{},[1317,1319,1325,1326,1332],{"type":37,"value":1318},"Escribe todos los datos en la tabla ",{"type":32,"tag":122,"props":1320,"children":1322},{"className":1321},[],[1323],{"type":37,"value":1324},"llm_all_sources",{"type":37,"value":244},{"type":32,"tag":122,"props":1327,"children":1329},{"className":1328},[],[1330],{"type":37,"value":1331},"query_id, llm_provider, date, source_domain, rank",{"type":37,"value":1333},". Ahora calcula una \"matriz de solapamiento\":",{"type":32,"tag":272,"props":1335,"children":1337},{"className":274,"code":1336,"language":276,"meta":16,"style":16},"SELECT \n  a.source_domain AS source_1,\n  b.source_domain AS source_2,\n  COUNT(DISTINCT a.query_id) AS co_citation_count\nFROM `project.dataset.llm_all_sources` a\nJOIN `project.dataset.llm_all_sources` b \n  ON a.query_id = b.query_id \n  AND a.llm_provider = b.llm_provider\n  AND a.date = b.date\nWHERE a.source_domain != b.source_domain\nGROUP BY source_1, source_2\nHAVING co_citation_count > 5\nORDER BY co_citation_count DESC;\n",[1338],{"type":32,"tag":122,"props":1339,"children":1340},{"__ignoreMap":16},[1341,1352,1379,1404,1448,1465,1482,1524,1561,1599,1637,1650,1673],{"type":32,"tag":282,"props":1342,"children":1343},{"class":284,"line":285},[1344,1348],{"type":32,"tag":282,"props":1345,"children":1346},{"style":289},[1347],{"type":37,"value":292},{"type":32,"tag":282,"props":1349,"children":1350},{"style":295},[1351],{"type":37,"value":298},{"type":32,"tag":282,"props":1353,"children":1354},{"class":284,"line":301},[1355,1360,1364,1369,1374],{"type":32,"tag":282,"props":1356,"children":1357},{"style":324},[1358],{"type":37,"value":1359},"  a",{"type":32,"tag":282,"props":1361,"children":1362},{"style":295},[1363],{"type":37,"value":252},{"type":32,"tag":282,"props":1365,"children":1366},{"style":324},[1367],{"type":37,"value":1368},"source_domain",{"type":32,"tag":282,"props":1370,"children":1371},{"style":289},[1372],{"type":37,"value":1373}," AS",{"type":32,"tag":282,"props":1375,"children":1376},{"style":295},[1377],{"type":37,"value":1378}," source_1,\n",{"type":32,"tag":282,"props":1380,"children":1381},{"class":284,"line":310},[1382,1387,1391,1395,1399],{"type":32,"tag":282,"props":1383,"children":1384},{"style":324},[1385],{"type":37,"value":1386},"  b",{"type":32,"tag":282,"props":1388,"children":1389},{"style":295},[1390],{"type":37,"value":252},{"type":32,"tag":282,"props":1392,"children":1393},{"style":324},[1394],{"type":37,"value":1368},{"type":32,"tag":282,"props":1396,"children":1397},{"style":289},[1398],{"type":37,"value":1373},{"type":32,"tag":282,"props":1400,"children":1401},{"style":295},[1402],{"type":37,"value":1403}," source_2,\n",{"type":32,"tag":282,"props":1405,"children":1406},{"class":284,"line":369},[1407,1412,1416,1421,1426,1430,1435,1439,1443],{"type":32,"tag":282,"props":1408,"children":1409},{"style":324},[1410],{"type":37,"value":1411},"  COUNT",{"type":32,"tag":282,"props":1413,"children":1414},{"style":295},[1415],{"type":37,"value":347},{"type":32,"tag":282,"props":1417,"children":1418},{"style":289},[1419],{"type":37,"value":1420},"DISTINCT",{"type":32,"tag":282,"props":1422,"children":1423},{"style":324},[1424],{"type":37,"value":1425}," a",{"type":32,"tag":282,"props":1427,"children":1428},{"style":295},[1429],{"type":37,"value":252},{"type":32,"tag":282,"props":1431,"children":1432},{"style":324},[1433],{"type":37,"value":1434},"query_id",{"type":32,"tag":282,"props":1436,"children":1437},{"style":295},[1438],{"type":37,"value":332},{"type":32,"tag":282,"props":1440,"children":1441},{"style":289},[1442],{"type":37,"value":361},{"type":32,"tag":282,"props":1444,"children":1445},{"style":295},[1446],{"type":37,"value":1447}," co_citation_count\n",{"type":32,"tag":282,"props":1449,"children":1450},{"class":284,"line":384},[1451,1455,1460],{"type":32,"tag":282,"props":1452,"children":1453},{"style":289},[1454],{"type":37,"value":375},{"type":32,"tag":282,"props":1456,"children":1457},{"style":378},[1458],{"type":37,"value":1459}," `project.dataset.llm_all_sources`",{"type":32,"tag":282,"props":1461,"children":1462},{"style":295},[1463],{"type":37,"value":1464}," a\n",{"type":32,"tag":282,"props":1466,"children":1467},{"class":284,"line":423},[1468,1473,1477],{"type":32,"tag":282,"props":1469,"children":1470},{"style":289},[1471],{"type":37,"value":1472},"JOIN",{"type":32,"tag":282,"props":1474,"children":1475},{"style":378},[1476],{"type":37,"value":1459},{"type":32,"tag":282,"props":1478,"children":1479},{"style":295},[1480],{"type":37,"value":1481}," b \n",{"type":32,"tag":282,"props":1483,"children":1484},{"class":284,"line":713},[1485,1490,1494,1498,1502,1507,1512,1516,1520],{"type":32,"tag":282,"props":1486,"children":1487},{"style":289},[1488],{"type":37,"value":1489},"  ON",{"type":32,"tag":282,"props":1491,"children":1492},{"style":324},[1493],{"type":37,"value":1425},{"type":32,"tag":282,"props":1495,"children":1496},{"style":295},[1497],{"type":37,"value":252},{"type":32,"tag":282,"props":1499,"children":1500},{"style":324},[1501],{"type":37,"value":1434},{"type":32,"tag":282,"props":1503,"children":1504},{"style":289},[1505],{"type":37,"value":1506}," =",{"type":32,"tag":282,"props":1508,"children":1509},{"style":324},[1510],{"type":37,"value":1511}," b",{"type":32,"tag":282,"props":1513,"children":1514},{"style":295},[1515],{"type":37,"value":252},{"type":32,"tag":282,"props":1517,"children":1518},{"style":324},[1519],{"type":37,"value":1434},{"type":32,"tag":282,"props":1521,"children":1522},{"style":295},[1523],{"type":37,"value":298},{"type":32,"tag":282,"props":1525,"children":1526},{"class":284,"line":26},[1527,1532,1536,1540,1544,1548,1552,1556],{"type":32,"tag":282,"props":1528,"children":1529},{"style":289},[1530],{"type":37,"value":1531},"  AND",{"type":32,"tag":282,"props":1533,"children":1534},{"style":324},[1535],{"type":37,"value":1425},{"type":32,"tag":282,"props":1537,"children":1538},{"style":295},[1539],{"type":37,"value":252},{"type":32,"tag":282,"props":1541,"children":1542},{"style":324},[1543],{"type":37,"value":905},{"type":32,"tag":282,"props":1545,"children":1546},{"style":289},[1547],{"type":37,"value":1506},{"type":32,"tag":282,"props":1549,"children":1550},{"style":324},[1551],{"type":37,"value":1511},{"type":32,"tag":282,"props":1553,"children":1554},{"style":295},[1555],{"type":37,"value":252},{"type":32,"tag":282,"props":1557,"children":1558},{"style":324},[1559],{"type":37,"value":1560},"llm_provider\n",{"type":32,"tag":282,"props":1562,"children":1564},{"class":284,"line":1563},9,[1565,1569,1573,1577,1582,1586,1590,1594],{"type":32,"tag":282,"props":1566,"children":1567},{"style":289},[1568],{"type":37,"value":1531},{"type":32,"tag":282,"props":1570,"children":1571},{"style":324},[1572],{"type":37,"value":1425},{"type":32,"tag":282,"props":1574,"children":1575},{"style":295},[1576],{"type":37,"value":252},{"type":32,"tag":282,"props":1578,"children":1579},{"style":324},[1580],{"type":37,"value":1581},"date",{"type":32,"tag":282,"props":1583,"children":1584},{"style":289},[1585],{"type":37,"value":1506},{"type":32,"tag":282,"props":1587,"children":1588},{"style":324},[1589],{"type":37,"value":1511},{"type":32,"tag":282,"props":1591,"children":1592},{"style":295},[1593],{"type":37,"value":252},{"type":32,"tag":282,"props":1595,"children":1596},{"style":324},[1597],{"type":37,"value":1598},"date\n",{"type":32,"tag":282,"props":1600,"children":1602},{"class":284,"line":1601},10,[1603,1607,1611,1615,1619,1624,1628,1632],{"type":32,"tag":282,"props":1604,"children":1605},{"style":289},[1606],{"type":37,"value":390},{"type":32,"tag":282,"props":1608,"children":1609},{"style":324},[1610],{"type":37,"value":1425},{"type":32,"tag":282,"props":1612,"children":1613},{"style":295},[1614],{"type":37,"value":252},{"type":32,"tag":282,"props":1616,"children":1617},{"style":324},[1618],{"type":37,"value":1368},{"type":32,"tag":282,"props":1620,"children":1621},{"style":289},[1622],{"type":37,"value":1623}," !=",{"type":32,"tag":282,"props":1625,"children":1626},{"style":324},[1627],{"type":37,"value":1511},{"type":32,"tag":282,"props":1629,"children":1630},{"style":295},[1631],{"type":37,"value":252},{"type":32,"tag":282,"props":1633,"children":1634},{"style":324},[1635],{"type":37,"value":1636},"source_domain\n",{"type":32,"tag":282,"props":1638,"children":1640},{"class":284,"line":1639},11,[1641,1645],{"type":32,"tag":282,"props":1642,"children":1643},{"style":289},[1644],{"type":37,"value":429},{"type":32,"tag":282,"props":1646,"children":1647},{"style":295},[1648],{"type":37,"value":1649}," source_1, source_2\n",{"type":32,"tag":282,"props":1651,"children":1653},{"class":284,"line":1652},12,[1654,1658,1663,1668],{"type":32,"tag":282,"props":1655,"children":1656},{"style":289},[1657],{"type":37,"value":719},{"type":32,"tag":282,"props":1659,"children":1660},{"style":295},[1661],{"type":37,"value":1662}," co_citation_count ",{"type":32,"tag":282,"props":1664,"children":1665},{"style":289},[1666],{"type":37,"value":1667},">",{"type":32,"tag":282,"props":1669,"children":1670},{"style":324},[1671],{"type":37,"value":1672}," 5\n",{"type":32,"tag":282,"props":1674,"children":1676},{"class":284,"line":1675},13,[1677,1681,1685,1689],{"type":32,"tag":282,"props":1678,"children":1679},{"style":289},[1680],{"type":37,"value":753},{"type":32,"tag":282,"props":1682,"children":1683},{"style":295},[1684],{"type":37,"value":1662},{"type":32,"tag":282,"props":1686,"children":1687},{"style":289},[1688],{"type":37,"value":763},{"type":32,"tag":282,"props":1690,"children":1691},{"style":295},[1692],{"type":37,"value":768},{"type":32,"tag":33,"props":1694,"children":1695},{},[1696,1698,1704],{"type":37,"value":1697},"Esta consulta te muestra: \"somos citados juntos con competitor-a en 47 consultas.\" Ahora divide ",{"type":32,"tag":122,"props":1699,"children":1701},{"className":1700},[],[1702],{"type":37,"value":1703},"co_citation_count",{"type":37,"value":1705}," por el número total de consultas donde competitor-a es citado — esto es tu \"citation overlap ratio\". Si es mayor al 60%, compiten directamente; si es menor al 30%, están en nichos diferentes.",{"type":32,"tag":33,"props":1707,"children":1708},{},[1709,1714,1717,1719,1724],{"type":32,"tag":40,"props":1710,"children":1711},{},[1712],{"type":37,"value":1713},"Convertir a acciones:",{"type":32,"tag":103,"props":1715,"children":1716},{},[],{"type":37,"value":1718},"\nSi el solapamiento es alto pero no te citan (competitor-a sí, tú no), cierra el gap de contenido de esa consulta. Lee la página del competidor — ¿qué facts menciona, qué formato usa (tabla \u002F lista \u002F código)? Da los mismos facts pero ",{"type":32,"tag":40,"props":1720,"children":1721},{},[1722],{"type":37,"value":1723},"más estructurados",{"type":37,"value":1725}," (JSON-LD, tabla, bullets) — el retrieval de LLM prefiere estas estructuras.",{"type":32,"tag":48,"props":1727,"children":1729},{"id":1728},"qué-empezar-a-medir-ahora",[1730],{"type":37,"value":1731},"Qué Empezar a Medir Ahora",{"type":32,"tag":33,"props":1733,"children":1734},{},[1735,1737,1744],{"type":37,"value":1736},"Para configurar métricas de citación en LLM, primero diseña un conjunto de consultas sintéticas — extrae de GSC las consultas de bajo CTR y alta impresión. Luego, crea un pipeline de sondeo diario con n8n, escribe las respuestas en BigQuery. En los primeros 30 días, establece la línea base: citation rate, primary source rate, prominence score. Después, mide el impacto de tus trabajos en ",{"type":32,"tag":882,"props":1738,"children":1741},{"href":1739,"rel":1740},"https:\u002F\u002Fwww.roibase.com.tr\u002Fes\u002Fgeo",[886],[1742],{"type":37,"value":1743},"Generative Engine Optimization",{"type":37,"value":1745}," — qué cambios de contenido subieron el citation rate, cuáles lo bajaron. Si obtuviste citaciones pero no tráfico, es un problema de prominencia — apunta a referencias inline. Analiza los patrones de co-citación para ver dónde te ganan los competidores y cierra esos gaps. Añade estas métricas a tu dashboard SEO — a finales de 2026, mirarás \"tráfico orgánico + visibilidad en LLM\" en lugar de solo tráfico orgánico.",{"type":32,"tag":1747,"props":1748,"children":1749},"style",{},[1750],{"type":37,"value":1751},"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":310,"depth":310,"links":1753},[1754,1755,1756,1761,1762,1763,1764],{"id":50,"depth":301,"text":53},{"id":85,"depth":301,"text":88},{"id":255,"depth":301,"text":258,"children":1757},[1758,1759,1760],{"id":267,"depth":310,"text":270},{"id":469,"depth":310,"text":472},{"id":615,"depth":310,"text":618},{"id":776,"depth":301,"text":779},{"id":991,"depth":301,"text":994},{"id":1276,"depth":301,"text":1279},{"id":1728,"depth":301,"text":1731},"markdown","content:es:ai:medicion-de-citacion-llm.md","content","es\u002Fai\u002Fmedicion-de-citacion-llm.md","es\u002Fai\u002Fmedicion-de-citacion-llm","md",1785103519287]