{"id":17468,"date":"2026-10-19T18:18:16","date_gmt":"2026-10-19T15:18:16","guid":{"rendered":"https:\/\/www.stradiji.com\/?p=17468"},"modified":"2026-10-19T18:18:16","modified_gmt":"2026-10-19T15:18:16","slug":"search-console-ve-bing-ile-ucretsiz-geo-olcum-hunisi","status":"publish","type":"post","link":"https:\/\/www.stradiji.com\/tr\/search-console-ve-bing-ile-ucretsiz-geo-olcum-hunisi\/","title":{"rendered":"Search Console ve Bing Webmaster Tools’u Birle\u015ftirdim: \u00dccretsiz GEO \u00d6l\u00e7\u00fcm Hunisi"},"content":{"rendered":"

GEO Yol Haritas\u0131<\/a> \u203a A\u015fama 7: \u00d6l\u00e7mek<\/a><\/p>\n

Yapay zek\u00e2 g\u00f6r\u00fcn\u00fcrl\u00fc\u011f\u00fcn\u00fc \u00f6l\u00e7meye ba\u015flamak i\u00e7in \u00fccretli bir araca ihtiyac\u0131n\u0131z yok. Google Search Console’un \u00fcretken yapay zek\u00e2 raporu ve Bing Webmaster Tools’un AI Performance raporu bu veriyi bedavaya ve birinci elden veriyor. \u0130ki rapor birbirinin bo\u015flu\u011funu dolduruyor. Onlar\u0131 klasik arama verisiyle ayn\u0131 d\u00f6neme hizalay\u0131nca d\u00f6rt soruluk bir GEO hunisi \u00e7\u0131k\u0131yor: talep, ta\u015f\u0131nma, pay ve de\u011fer.<\/p>\n

Bu yaz\u0131da her raporun ne g\u00f6sterdi\u011fini, huniyi ad\u0131m ad\u0131m, kendi sitemde ilk okumam\u0131n beni nas\u0131l yan\u0131ltt\u0131\u011f\u0131n\u0131, ayn\u0131 huninin ABD’deki bir SaaS sitesinde neler g\u00f6sterdi\u011fini ve bu hafta 30 dakikada kurabilece\u011finiz s\u00fcr\u00fcm\u00fcn\u00fc anlat\u0131yorum.<\/p>\n

Bug\u00fcn piyasada yapay zek\u00e2 g\u00f6r\u00fcn\u00fcrl\u00fc\u011f\u00fcn\u00fc \u00f6l\u00e7mek i\u00e7in onlarca \u00fccretli ara\u00e7 var. \u00c7o\u011fu ayn\u0131 \u015feyi yap\u0131yor: prompt’lar\u0131 kendileri yaz\u0131yor, cevaplar\u0131 kendileri topluyor, size bir skor veriyor. Oysa iki arama motoru, kendi yapay zek\u00e2lar\u0131nda sitenizin nas\u0131l g\u00f6r\u00fcnd\u00fc\u011f\u00fcn\u00fc art\u0131k size do\u011frudan s\u00f6yl\u00fcyor. Ben de stradiji.com i\u00e7in ikisini klasik arama verisiyle ayn\u0131 \u00fc\u00e7 aya hizalad\u0131m.<\/p>\n

Ana tez: Her rapor size resmin yar\u0131s\u0131n\u0131 veriyor<\/h2>\n

Google, talebi ve yapay zek\u00e2daki g\u00f6r\u00fcn\u00fcrl\u00fc\u011f\u00fc g\u00f6steriyor ama sorguyu ve rakiplerle payla\u015f\u0131m\u0131 g\u00f6stermiyor. Bing, yapay zek\u00e2n\u0131n kulland\u0131\u011f\u0131 sorguyu ve sizin at\u0131f pay\u0131n\u0131z\u0131 g\u00f6steriyor ama talebi g\u00f6stermiyor. \u0130kisini birle\u015ftirince d\u00f6rt soruluk \u00fccretsiz bir GEO hunisi \u00e7\u0131k\u0131yor:<\/p>\n

    \n
  1. Aran\u0131yor muyum?<\/strong><\/li>\n
  2. Arand\u0131\u011f\u0131mda yapay zek\u00e2ya giriyor muyum?<\/strong><\/li>\n
  3. Yapay zek\u00e2 cevap verirken konu\u015fman\u0131n ne kadar\u0131 benim?<\/strong><\/li>\n
  4. Bu g\u00f6r\u00fcn\u00fcrl\u00fck bana de\u011fer getiriyor mu?<\/strong><\/li>\n<\/ol>\n

    Huniyi kendi siteme uygulad\u0131m. \u0130lk tablom beni yan\u0131ltt\u0131, ikinci okuma \u00e7ok daha net bir i\u015f listesi verdi.<\/p>\n

    \"\u0130nfografik:<\/p>\n

    1. Neden iki rapor birlikte?<\/h2>\n

    \u00d6nce kim neyi veriyor, ona bakal\u0131m:<\/p>\n

    \n\n\n\n\n\n\n\n
    Rapor<\/th>\nNe g\u00f6steriyor?<\/th>\nNe g\u00f6stermiyor?<\/th>\n<\/tr>\n<\/thead>\n
    Google Search Console, klasik performans<\/td>\nTalep: sorgu, sayfa, g\u00f6sterim, t\u0131klama<\/td>\nYapay zek\u00e2ya dair hi\u00e7bir \u015fey<\/td>\n<\/tr>\n
    Google Search Console, \u00fcretken yapay zek\u00e2 raporu<\/td>\nSayfa baz\u0131nda yapay zek\u00e2 \u00f6zetleri (AI Overviews) ve AI Modu g\u00f6sterimi<\/td>\nSorgu ve t\u0131klama<\/td>\n<\/tr>\n
    Bing Webmaster Tools, AI Performance<\/td>\nCopilot at\u0131flar\u0131, her atf\u0131n arkas\u0131ndaki dayanak (grounding) sorgusu, at\u0131f pay\u0131n\u0131z<\/td>\nTalep; Bing veriyi “\u00f6rneklem” olarak tan\u0131ml\u0131yor<\/td>\n<\/tr>\n<\/tbody>\n<\/table>
    Google’da sorgu yok, Bing’de talep yok. Birinin bo\u015flu\u011funu \u00f6b\u00fcr\u00fc dolduruyor.<\/figcaption><\/figure>\n

    Mike King, 15 Ekim’deki \u00fcretken SERP yaz\u0131s\u0131nda<\/a> “getirilen” ve “al\u0131nt\u0131lanan” kaynaklar\u0131 ayr\u0131 \u00f6l\u00e7meyi \u00f6neriyor. Bu iki \u00fccretsiz rapor, o ayr\u0131m\u0131n bug\u00fcn elimizdeki en yak\u0131n birinci taraf kar\u015f\u0131l\u0131\u011f\u0131.<\/p>\n

    stradiji.com i\u00e7in 18 Temmuz ile 17 Ekim 2026 aras\u0131n\u0131 hizalad\u0131m: Google klasik aramada 112.019 g\u00f6sterim ve 508 t\u0131klama, \u00fcretken yapay zek\u00e2 raporunda 6.356 g\u00f6sterim. Bing’de ayn\u0131 d\u00f6nemde 215 Copilot atf\u0131. Her i\u00e7eri\u011fi bir kez saymak i\u00e7in adresleri \/tr\/<\/code>, \/en\/<\/code> ve sondaki e\u011fik \u00e7izgiden ar\u0131nd\u0131r\u0131p i\u00e7erik d\u00fczeyinde birle\u015ftirdim.<\/p>\n

    \"Search<\/p>\n

    2. Huni: D\u00f6rt soru, \u00fc\u00e7 rapor<\/h2>\n
      \n
    1. Talep: Aran\u0131yor muyum?<\/strong> Kaynak: klasik Search Console g\u00f6sterimi.<\/li>\n
    2. Ta\u015f\u0131nma: Arand\u0131\u011f\u0131mda yapay zek\u00e2ya giriyor muyum?<\/strong> Kaynak: \u00fcretken yapay zek\u00e2 g\u00f6sterimi b\u00f6l\u00fc klasik g\u00f6sterim. Ben buna ta\u015f\u0131nma oran\u0131<\/strong> diyorum ve binde olarak okuyorum. Yan\u0131na bir de kapsama<\/strong> koyuyorum: i\u00e7eriklerinizin y\u00fczde ka\u00e7\u0131 yapay zek\u00e2da en az bir kez g\u00f6r\u00fcn\u00fcyor?<\/li>\n
    3. Pay: Yapay zek\u00e2 cevap verirken konu\u015fman\u0131n ne kadar\u0131 benim?<\/strong> Kaynak: Bing. At\u0131f say\u0131n\u0131z\u0131 at\u0131f pay\u0131n\u0131za b\u00f6l\u00fcnce o sorgudaki toplam at\u0131f pastas\u0131<\/strong> \u00e7\u0131k\u0131yor.<\/li>\n
    4. De\u011fer: Bu g\u00f6r\u00fcn\u00fcrl\u00fck t\u0131klamaya ve sat\u0131\u015fa d\u00f6n\u00fcyor mu?<\/strong> Kaynak: Search Console t\u0131klama oran\u0131, sonra GA4 ve CRM.<\/li>\n<\/ol>\n

      stradiji.com’un genel resmi \u015f\u00f6yle: her 1.000 klasik g\u00f6sterime 57 yapay zek\u00e2 g\u00f6sterimi d\u00fc\u015f\u00fcyor. 795 i\u00e7eri\u011fin %36’s\u0131 yapay zek\u00e2da en az bir kez g\u00f6r\u00fcnm\u00fc\u015f. As\u0131l hik\u00e2ye ise bu ortalamalar\u0131n alt\u0131nda.<\/p>\n

      3. Talep ve ta\u015f\u0131nma: \u0130lk tablom beni nas\u0131l yan\u0131ltt\u0131?<\/h2>\n

      Hunisiz ilk okumam \u015fuydu: Google’daki yapay zek\u00e2 g\u00f6sterimlerimin %53,5’i s\u00f6zl\u00fck maddelerine, sadece %1,9’u \u00f6zg\u00fcn ara\u015ft\u0131rmalar\u0131ma gidiyor. Bing’de ara\u015ft\u0131rman\u0131n pay\u0131 %0,5. Yaz\u0131y\u0131 neredeyse “Yapay zek\u00e2 ara\u015ft\u0131rmam\u0131 de\u011fil, s\u00f6zl\u00fc\u011f\u00fcm\u00fc al\u0131nt\u0131l\u0131yor” ba\u015fl\u0131\u011f\u0131yla yay\u0131ml\u0131yordum.<\/p>\n

      Sonra huninin ilk iki katman\u0131n\u0131 ekledim:<\/p>\n

      \n\n\n\n\n\n\n\n\n
      \u0130\u00e7erik t\u00fcr\u00fc<\/th>\n\u0130\u00e7erik say\u0131s\u0131<\/th>\nKapsama<\/th>\nKlasik aramadaki pay<\/th>\nYapay zek\u00e2daki pay<\/th>\nTa\u015f\u0131nma oran\u0131 (binde)<\/th>\n<\/tr>\n<\/thead>\n
      S\u00f6zl\u00fck<\/td>\n148<\/td>\n%43<\/td>\n%27,2<\/td>\n%53,5<\/td>\n112<\/td>\n<\/tr>\n
      Rehber ve haber<\/td>\n604<\/td>\n%32<\/td>\n%68,3<\/td>\n%40,3<\/td>\n34<\/td>\n<\/tr>\n
      Kurumsal<\/td>\n29<\/td>\n%72<\/td>\n%3,6<\/td>\n%4,2<\/td>\n65<\/td>\n<\/tr>\n
      \u00d6zg\u00fcn ara\u015ft\u0131rma<\/td>\n14<\/td>\n%79<\/td>\n%0,8<\/td>\n%1,9<\/td>\n133<\/td>\n<\/tr>\n<\/tbody>\n<\/table>
      Kaynak: stradiji.com, Google Search Console \u00fcretken yapay zek\u00e2 raporu ve klasik performans (API), 18 Temmuz ile 17 Ekim 2026 aras\u0131.<\/figcaption><\/figure>\n

      Ara\u015ft\u0131rman\u0131n k\u00fc\u00e7\u00fck pay\u0131n\u0131n sebebi yapay zek\u00e2 de\u011fil, talep. Ara\u015ft\u0131rmalar klasik aramada da g\u00f6sterimlerin sadece %0,8’ini al\u0131yor. Arand\u0131klar\u0131nda ise yapay zek\u00e2ya en y\u00fcksek oranda ta\u015f\u0131nan t\u00fcr onlar. 14 ara\u015ft\u0131rman\u0131n 11’i en az bir kez yapay zek\u00e2da g\u00f6r\u00fcnm\u00fc\u015f.<\/p>\n

      Bir parantez a\u00e7ay\u0131m: bu ortalamay\u0131 tek bir i\u00e7erik yukar\u0131 \u00e7ekiyor, yapay zek\u00e2 \u00f6zetlerinin T\u00fcrkiye’de aramay\u0131 nas\u0131l de\u011fi\u015ftirdi\u011fini inceledi\u011fim \u00e7al\u0131\u015fma<\/a>. Onu \u00e7\u0131kar\u0131nca oran binde 79’a iniyor, yine rehberlerin iki kat\u0131ndan fazla. 14 i\u00e7erik k\u00fc\u00e7\u00fck bir \u00f6rneklem. Mike King’in 17 Eyl\u00fcl’deki kesinlik uyar\u0131s\u0131n\u0131<\/a> hat\u0131rlayarak bunu kanun gibi de\u011fil, hipotez gibi okuyorum.<\/p>\n

      T\u00fcr i\u00e7indeki fark t\u00fcrler aras\u0131ndaki farktan da b\u00fcy\u00fck. “Semantik HTML nedir” maddesi binde 614 ta\u015f\u0131n\u0131yor. “SERP nedir” maddesi 1.984 klasik g\u00f6sterim alm\u0131\u015f, yapay zek\u00e2da s\u0131f\u0131r. Ayn\u0131 \u015fablon, ayn\u0131 yazar. Belirleyici olan sayfan\u0131n kal\u0131b\u0131 de\u011fil, o sorunun yapay zek\u00e2 cevab\u0131 a\u00e7\u0131p a\u00e7mad\u0131\u011f\u0131. Peec AI’\u0131n 6 A\u011fustos’taki yeniden s\u0131ralay\u0131c\u0131 (reranker) analizi<\/a> de bunu s\u00f6yl\u00fcyor: tan\u0131m ve liste, e\u015fle\u015fen niyette \u00f6ne \u00e7\u0131k\u0131yor; genel bir bi\u00e7im tercihi yok.<\/p>\n

      4. Pay: Bing size her sorgunun pastas\u0131n\u0131 veriyor<\/h2>\n

      Bing’in raporunda neredeyse kimsenin dikkat etmedi\u011fi bir s\u00fctun var: at\u0131f pay\u0131<\/strong>. Copilot o sorguda toplam ka\u00e7 kaynak g\u00f6sterdiyse, sizinki onun y\u00fczde ka\u00e7\u0131, bunu s\u00f6yl\u00fcyor. At\u0131f\u0131n\u0131z\u0131 bu paya b\u00f6l\u00fcnce o sorgudaki toplam pastay\u0131 buluyorsunuz.<\/p>\n

      Benim sitemde iki u\u00e7 \u00f6rnek var:<\/p>\n

      \n\n\n\n\n\n\n
      Dayanak sorgusu<\/th>\nAtf\u0131m<\/th>\nAt\u0131f pay\u0131m<\/th>\nToplam pasta (yakla\u015f\u0131k)<\/th>\n<\/tr>\n<\/thead>\n
      “arama kriteri”<\/td>\n40<\/td>\n%26<\/td>\n150<\/td>\n<\/tr>\n
      “bing webmaster tools”<\/td>\n7<\/td>\n%0,2<\/td>\n3.300<\/td>\n<\/tr>\n<\/tbody>\n<\/table>
      Kaynak: Bing Webmaster Tools AI Performance, stradiji.com.<\/figcaption><\/figure>\n

      \u0130lki k\u00fc\u00e7\u00fck bir pasta ama d\u00f6rtte biri benim. \u0130kincisi b\u00fcy\u00fck bir pasta ama i\u00e7inde bir k\u0131r\u0131nt\u0131y\u0131m. Bu ikisi farkl\u0131 stratejiler ister. K\u00fc\u00e7\u00fck pastada sahip oldu\u011funuz yeri korursunuz. B\u00fcy\u00fck pastada pay almak i\u00e7in sitenizin d\u0131\u015f\u0131nda da konu\u015fulman\u0131z gerekir.<\/p>\n

      5. De\u011fer: G\u00f6r\u00fcn\u00fcrl\u00fck t\u0131klamaya d\u00f6n\u00fcyor mu?<\/h2>\n

      Huninin son katman\u0131nda iki \u015fey \u00e7\u0131kt\u0131.<\/p>\n

      S\u0131f\u0131r t\u0131klama.<\/strong> Alt\u0131 ayl\u0131k klasik veride (19 Nisan ile 17 Ekim 2026 aras\u0131) s\u00f6zl\u00fck sayfalar\u0131m\u0131n t\u0131klama oran\u0131 %0,17, di\u011fer sayfalar\u0131m\u0131n %1,03. Copilot “chat gpt g\u00f6rsel olu\u015fturma” sorgusunda beni 58 kez kaynak g\u00f6stermi\u015f. Google’da ayn\u0131 sorgu alt\u0131 ayda 901 g\u00f6sterim alm\u0131\u015f, t\u0131klama tek.<\/p>\n

      Yanl\u0131\u015f kimlik.<\/strong> Google’daki yapay zek\u00e2 g\u00f6sterimlerimin %56’s\u0131 \u00fc\u00e7 i\u00e7erikte toplan\u0131yor: PWA nedir (1.905), ChatGPT g\u00f6rsel komutlar\u0131 (925) ve SEO ara\u00e7lar\u0131 listesi (718). \u00dc\u00e7\u00fc de as\u0131l i\u015fimden, yani kurumsal markalara SEO ve GEO dan\u0131\u015fmanl\u0131\u011f\u0131ndan uzak. Yapay zek\u00e2n\u0131n g\u00f6z\u00fcnde stradiji.com bir GEO dan\u0131\u015fman\u0131ndan \u00e7ok bir PWA kayna\u011f\u0131. Bing’in 13 Ekim’de duyurdu\u011fu Topic Portfolio<\/a> raporu da ayn\u0131 resmi \u00e7iziyor: \u00f6nerdi\u011fi “ai-driven search visibility expert help” gibi tam karar sorgusunda ben yokum.<\/p>\n

      De\u011fer katman\u0131n\u0131n sorusu basit: yapay zek\u00e2 g\u00f6sterimlerinizin y\u00fczde ka\u00e7\u0131 satmak istedi\u011finiz konuya d\u00fc\u015f\u00fcyor?<\/p>\n

      6. \u00c7apraz okuma 1: \u0130ki motor, iki farkl\u0131 site<\/h2>\n

      \u0130ki raporu yan yana koyman\u0131n en beklenmedik faydas\u0131, iki motorun sitenizin farkl\u0131 foto\u011fraflar\u0131n\u0131 okudu\u011funu g\u00f6rmek.<\/p>\n