{"id":13877,"date":"2017-12-01T20:54:18","date_gmt":"2017-12-01T18:54:18","guid":{"rendered":"https:\/\/www.dase-analytics.com\/blog\/?p=13877"},"modified":"2019-02-12T13:56:13","modified_gmt":"2019-02-12T11:56:13","slug":"ako-analyzovat-data-analyticky-framework","status":"publish","type":"post","link":"https:\/\/www.dase-analytics.com\/blog\/sk\/ako-analyzovat-data-analyticky-framework\/","title":{"rendered":"Ako analyzova\u0165 d\u00e1ta? ( Analytick\u00fd Framework)"},"content":{"rendered":"

Mo\u017enosti Google Analytics s\u00fa skoro neobmedzen\u00e9. \u010ci u\u017e si prajete analyzova\u0165 technick\u00fa str\u00e1nku svojej webovej str\u00e1nky, pr\u00edpadne pou\u017e\u00edvate\u013eov alebo ich spr\u00e1vanie, m\u00f4\u017ee sa \u010dlovek mnohokr\u00e1t c\u00edti\u0165 straten\u00fd. Niekedy n\u00e1jdenie probl\u00e9mu je ako h\u013eada\u0165 ihlu v kope sena.<\/strong><\/p>\n

Ako a kde teda za\u010da\u0165 s anal\u00fdzou d\u00e1t?<\/span><\/p>\n

Najd\u00f4le\u017eitej\u0161ie je polo\u017ei\u0165 spr\u00e1vnu ot\u00e1zku<\/span><\/h2>\n

Vyhodnoti\u0165 d\u00e1ta, analyzova\u0165 marketingov\u00e9 kan\u00e1ly, segmentova\u0165 pou\u017e\u00edvate\u013eov?<\/span><\/p>\n

By\u0165 dobr\u00fdm webov\u00fdm analytikom znamen\u00e1 vedie\u0165 polo\u017ei\u0165 spr\u00e1vne ot\u00e1zky a rozv\u00edja\u0165 ich.<\/span><\/p>\n

Dobr\u00e1 ot\u00e1zka mus\u00ed by\u0165:<\/span><\/p>\n

    \n
  • \u0161pecifick\u00e1 <\/span><\/li>\n
  • merate\u013en\u00e1<\/span><\/li>\n
  • \u010dasovo ohrani\u010den\u00e1<\/span><\/li>\n
  • relevantn\u00e1<\/span><\/li>\n<\/ul>\n

    Met\u00f3da Kaizen<\/span><\/h2>\n

    Na to, aby sme sa po\u010das anal\u00fdzy d\u00e1t nestratili a pri\u0161li od polo\u017eenia dobrej ot\u00e1zky k anal\u00fdze samotnej, je dobr\u00e9 postupova\u0165 \u0161trukturovane, pod\u013ea met\u00f3dy. V praxi sa mi osved\u010dila <\/span>met\u00f3da<\/b> Kaizen.<\/b><\/p>\n

    Pozost\u00e1va z piatich krokov:<\/span><\/p>\n

      \n
    1. \u00a0<\/span> Anomality<\/b>: Kde nach\u00e1dzame anomality v d\u00e1tach? S\u00fa to dobr\u00e9 alebo zl\u00e9 ukazovatele? Je nie\u010do nezvy\u010dajn\u00e9 nezvy\u010dajn\u00e9 v mojich d\u00e1tach?<\/span><\/li>\n
    2. \u00a0<\/span> Anal\u00fdzy<\/b>: Ke\u010f sme identifikovali anomalitu, m\u00f4\u017eeme ju analyzova\u0165.<\/span><\/li>\n
    3. \u00a0<\/span> Rie\u0161enia<\/b>: Vygenerujeme rie\u0161enia pre anomality, ktor\u00e9 sme analyzovali.<\/span><\/li>\n
    4. \u00a0<\/span> Implement\u00e1cie:<\/b> Najlep\u0161ie rie\u0161enie implementujeme.<\/span><\/li>\n
    5. \u00a0<\/span> Vyhodnotenie<\/b>: Priniesli rie\u0161enia \u017eelan\u00e9 v\u00fdsledky? Boli tieto \u0161tatistiky v\u00fdznamn\u00e9?\u00a0\"\" <\/span><\/li>\n<\/ol>\n

      Tento proces je teda premena inform\u00e1cie, ktor\u00fa dostaneme z d\u00e1t (anomality \u2013 polo\u017eenia spr\u00e1vnej ot\u00e1zky) na akciu (rie\u0161enie, ktor\u00e9 pri\u0161lo po anal\u00fdze d\u00e1t, a od ktor\u00e9ho d\u00fafame zlep\u0161enie).<\/span><\/p>\n

      Rozoznanie anomal\u00edt<\/span><\/h2>\n

      Najd\u00f4le\u017eitej\u0161\u00edm a \u010dasto najkomplikovanej\u0161\u00edm krokom celej Kaizen met\u00f3dy je identifikovanie anomal\u00edt. Pre\u010do? Preto\u017ee Google Analytics n\u00e1m pon\u00faka tak\u00e9 mno\u017estvo d\u00e1t, grafov a tabuliek, \u017ee \u010dasto nie je evidentn\u00e9, \u017ee nie\u010do nesed\u00ed, \u017ee je nie\u010do divn\u00e9.\u00a0 \u00a0<\/span><\/p>\n

      Existuj\u00fa v\u0161ak 4 kroky, ktor\u00e9 n\u00e1m pom\u00f4\u017eu analyzova\u0165 anomality.<\/span><\/p>\n

      Na za\u010diatku je potrebn\u00e9 definova\u0165 si \u010dasov\u00fd rozsah. Potom je treba s\u00fastredi\u0165 sa na technol\u00f3giu a vstupn\u00e9 dimenzie. Najviac anomal\u00edt m\u00f4\u017ee by\u0165 vysvetlen\u00fdch na z\u00e1klade anal\u00fdzy t\u00fdchto metr\u00edk.<\/span><\/p>\n

      Ak tieto kroky nepom\u00f4\u017eu pri rie\u0161en\u00ed v\u00e1\u0161ho probl\u00e9mu, je spr\u00e1vny \u010das ponori\u0165 sa do hlb\u0161ej anal\u00fdzy spr\u00e1vania u\u017e\u00edvate\u013ea. <\/span><\/p>\n

      \"\"
      \n1. \u010casov\u00fd rozsah<\/span><\/h3>\n

      Zvoli\u0165 si spr\u00e1vny \u010dasov\u00fd rozsah je najd\u00f4le\u017eitej\u0161ie pre efekt\u00edvnu anal\u00fdzu.<\/span><\/p>\n

      \"\"<\/p>\n

      Anomality sa m\u00f4\u017eu vyskytova\u0165 bu\u010f skokovo, po\u010das \u010dasov\u00e9ho \u00faseku, alebo sa anomalita vyskytuje v porovnan\u00ed s predch\u00e1dzaj\u00facim \u010dasov\u00fdm obdob\u00edm. D\u00f4le\u017eit\u00e9 je vykresli\u0165 spr\u00e1vne d\u00e1ta v dostato\u010dnom \u010dasovom obdob\u00ed a s\u00fastredi\u0165 sa presne na toto obdobie.<\/span><\/p>\n

      Pr\u00edklad: <\/b>V\u0161imnete si napr\u00edklad, \u017ee v marci v\u00e1m klesol po\u010det u\u017e\u00edvate\u013eov o 40 %. Je tento trend podobn\u00fd s d\u00e1tami z predch\u00e1dzaj\u00faceho roka? Nastal tento jav skokovo? Je to sp\u00f4soben\u00e9 t\u00fdm, \u017ee skon\u010dilo leto a prich\u00e1dza zima (m\u00f4j e-shop napr\u00edklad pred\u00e1va plavky)?<\/span><\/p>\n

      2. Technol\u00f3gia<\/span><\/h3>\n

      Ve\u013ea anomal\u00edt je sp\u00f4soben\u00fdch technol\u00f3giou. V tomto pr\u00edpade je rozumn\u00e9 vytvori\u0165 segment zalo\u017een\u00fd na dimenzi\u00e1ch, ako:<\/span><\/p>\n

        \n
      • druh zariadenia (mobil, tablet, PC),<\/span><\/li>\n
      • verzia prehliada\u010da (Chrome, Safari, IE…),<\/span><\/li>\n
      • sie\u0165,<\/span><\/li>\n
      • verzia aplik\u00e1cie.<\/span><\/li>\n<\/ul>\n

        Pr\u00edklad: <\/b>Klesol po\u010det pou\u017e\u00edvate\u013eov v porovnan\u00ed s predch\u00e1dzaj\u00facim mesiacom. Aplikovan\u00edm segmentu a anal\u00fdzou technol\u00f3gie zist\u00edme, \u017ee sa tak stalo na PC s prehliada\u010dom Internet Explorer.<\/span><\/p>\n

        M\u00f4\u017ee to by\u0165 sp\u00f4soben\u00e9 probl\u00e9mami pri na\u010d\u00edtavan\u00ed, preto\u017ee ur\u010dit\u00e1 verzia Internet Exploreru na PC nedok\u00e1\u017ee spracova\u0165 JavaScript umiestnen\u00fd na va\u0161ej str\u00e1nke. V tomto pr\u00edpade m\u00f4\u017eeme napr\u00edklad proakt\u00edvne upozorni\u0165 u\u017e\u00edvate\u013eov str\u00e1nky, aby si aktualizovali verziu Internet Exploreru.<\/span><\/p>\n

        3. Vstupn\u00e9 dimenzie<\/span><\/h3>\n

        Po definovan\u00ed \u010dasov\u00e9ho obdobia a identifikovan\u00ed eventu\u00e1lnych technick\u00fdch parametrov sa m\u00f4\u017eeme s\u00fastredi\u0165 na vstupn\u00e9 dimenzie. <\/span><\/p>\n

        Tu je p\u00e1r metr\u00edk, na ktor\u00e9 by sme sa mali prim\u00e1rne s\u00fastredi\u0165:<\/span><\/p>\n

          \n
        • zdroj n\u00e1v\u0161tevnosti,<\/span><\/li>\n
        • landing page,<\/span><\/li>\n
        • miera okam\u017eit\u00fdch odchodov,<\/span><\/li>\n
        • refferal.<\/span><\/li>\n<\/ul>\n

          Ve\u013ea anomal\u00edt m\u00f4\u017ee by\u0165 sp\u00f4soben\u00fdch ur\u010dit\u00fdmi zdrojmi n\u00e1v\u0161tevnosti. Napr\u00edklad platen\u00fdmi a neplaten\u00fdmi n\u00e1v\u0161tevami.<\/span><\/p>\n

          4. Spr\u00e1vanie pou\u017e\u00edvate\u013ea<\/span><\/h3>\n

          Cez tieto metriky sa sna\u017e\u00edme porozumie\u0165 tomu, ako sa pou\u017e\u00edvate\u013e na na\u0161ej str\u00e1nke spr\u00e1va. Patria sme hlavne tieto metriky:<\/span><\/p>\n

            \n
          • \u010das na str\u00e1nke,<\/span><\/li>\n
          • konverzie a konverzn\u00fd pomer,<\/span><\/li>\n
          • udalosti,<\/span><\/li>\n
          • pou\u017e\u00edvanie vyh\u013ead\u00e1vania.<\/span><\/li>\n<\/ul>\n

            Ke\u010f sa n\u00e1m podar\u00ed \u00faspe\u0161ne identifikova\u0165 anomality, m\u00f4\u017eeme sa hlb\u0161ie ponori\u0165 do \u010fal\u0161\u00edch krokov met\u00f3dy Kaizen. V\u017edy je v\u0161ak potrebn\u00e9 rozm\u00fd\u0161\u013ea\u0165 o tom, \u010di s\u00fa na\u0161e d\u00e1ta a zistenia \u0161tatisticky relevantn\u00e9. <\/span><\/p>\n

            O praktickom vyu\u017eit\u00ed \u0161tatistiky v anal\u00fdze d\u00e1t \u010doskoro nap\u00ed\u0161eme \u010fal\u0161\u00ed blog \ud83d\ude42<\/span><\/p>\n","protected":false},"excerpt":{"rendered":"

            Mo\u017enosti Google Analytics s\u00fa skoro neobmedzen\u00e9. \u010ci u\u017e si prajete analyzova\u0165 technick\u00fa str\u00e1nku svojej webovej str\u00e1nky, pr\u00edpadne pou\u017e\u00edvate\u013eov…<\/p>\n","protected":false},"author":64,"featured_media":13878,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":[],"categories":[639,637],"tags":[],"_links":{"self":[{"href":"https:\/\/www.dase-analytics.com\/blog\/sk\/wp-json\/wp\/v2\/posts\/13877"}],"collection":[{"href":"https:\/\/www.dase-analytics.com\/blog\/sk\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.dase-analytics.com\/blog\/sk\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.dase-analytics.com\/blog\/sk\/wp-json\/wp\/v2\/users\/64"}],"replies":[{"embeddable":true,"href":"https:\/\/www.dase-analytics.com\/blog\/sk\/wp-json\/wp\/v2\/comments?post=13877"}],"version-history":[{"count":3,"href":"https:\/\/www.dase-analytics.com\/blog\/sk\/wp-json\/wp\/v2\/posts\/13877\/revisions"}],"predecessor-version":[{"id":13884,"href":"https:\/\/www.dase-analytics.com\/blog\/sk\/wp-json\/wp\/v2\/posts\/13877\/revisions\/13884"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.dase-analytics.com\/blog\/sk\/wp-json\/wp\/v2\/media\/13878"}],"wp:attachment":[{"href":"https:\/\/www.dase-analytics.com\/blog\/sk\/wp-json\/wp\/v2\/media?parent=13877"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.dase-analytics.com\/blog\/sk\/wp-json\/wp\/v2\/categories?post=13877"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.dase-analytics.com\/blog\/sk\/wp-json\/wp\/v2\/tags?post=13877"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}