{"id":8179,"date":"2024-12-12T12:07:27","date_gmt":"2024-12-12T11:07:27","guid":{"rendered":"https:\/\/igloonet.cz\/blog\/?p=8179"},"modified":"2025-04-02T12:11:26","modified_gmt":"2025-04-02T10:11:26","slug":"triangulace-v-marketingu-co-to-je-a-jak-s-tim-pracovat","status":"publish","type":"post","link":"https:\/\/igloonet.cz\/blog\/triangulace-v-marketingu-co-to-je-a-jak-s-tim-pracovat\/","title":{"rendered":"Triangulace v marketingu - co to je a jak s t\u00edm pracovat"},"content":{"rendered":"
<\/div>\n

Co znamen\u00e1 triangulace u vyhodnocov\u00e1n\u00ed v marketingu? Co tvo\u0159\u00ed jej\u00ed sou\u010d\u00e1sti a pro\u010d u\u017e samotn\u00e9 atribuce nesta\u010d\u00ed? Adam sepsal hlavn\u00ed body z prosincov\u00e9ho tu\u010d\u0148\u00e1\u010d\u00edho webin\u00e1\u0159e o vyhodnocov\u00e1n\u00ed.
\n<\/p>\n

Doned\u00e1vna bylo vyhodnocov\u00e1n\u00ed v marketingu zalo\u017een\u00e9 t\u00e9m\u011b\u0159 v\u00fdhradn\u011b na atribuci. Jednoduch\u00fd koncept, kdy se podle n\u011bjak\u00e9ho pravidla rozpo\u010d\u00edt\u00e1v\u00e1, jak\u00fd vliv m\u011bly jednotliv\u00e9 kan\u00e1ly a kampan\u011b na n\u00e1kup.<\/p>\n

Koncept atribuc\u00ed od za\u010d\u00e1tku stoj\u00ed na z\u00e1kladech, kter\u00e9 ale nejde jednodu\u0161e naplnit<\/a>. K probl\u00e9mu s cross-device<\/strong> (lid\u00e9 maj\u00ed takov\u00fd o\u0161kliv\u00fd zvyk, \u017ee str\u00e1nky nav\u0161t\u011bvuj\u00ed z v\u00edce ne\u017e jednoho za\u0159\u00edzen\u00ed) se p\u0159idala z\u00e1konn\u00e1 a technick\u00e1 omezen\u00ed. GDRP n\u00e1m nakazuje, kdy m\u016f\u017eeme m\u011b\u0159it a aby toho nebylo m\u00e1lo, h\u00e1\u017e\u00ed n\u00e1m v tom klacky pod nohy i prohl\u00ed\u017ee\u010de a za\u0159\u00edzen\u00ed. Jeden p\u0159\u00edklad za v\u0161echny p\u0159edstavuje Apple a jeho Intelligent Tracking Prevention<\/strong>.<\/p>\n

To neznamen\u00e1, \u017ee by u\u017e atribuce byla \u00fapln\u011b nepou\u017eiteln\u00e1, jen se na n\u011bkter\u00e9 \u010dinnosti u\u017e ned\u00e1 spolehliv\u011b pou\u017e\u00edt<\/strong>. Nap\u0159\u00edklad na rozd\u011blov\u00e1n\u00ed rozpo\u010dt\u016f.<\/p>\n

Co se s t\u00edm d\u00e1 d\u011blat?<\/strong><\/p>\n

Bohu\u017eel neexistuje jeden nov\u00fd kouzeln\u00fd p\u0159\u00edstup, kter\u00fd by nahradil atribuci a my tak jen p\u0159e\u0161li z jednoho n\u00e1stroje na jin\u00fd. Bude za t\u00edm daleko v\u00edc pr\u00e1ce.<\/p>\n

\"triangulace<\/a><\/p>\n

Odpov\u011bd\u00ed se zd\u00e1 pr\u00e1v\u011b triangulace. Tedy kombinace r\u016fzn\u00fdch metod pro vyhodnocov\u00e1n\u00ed.<\/p>\n

V sou\u010dasn\u00e9 dob\u011b se za jej\u00ed \u010d\u00e1sti pova\u017euj\u00ed inkrement\u00e1ln\u00ed experimenty, media mix modelling a st\u00e1le atribuce.<\/p>\n

Atribuce<\/h2>\n

\"atribuce\"<\/a><\/p>\n

Atribuce je zalo\u017een\u00e1 na jednoduch\u00e9 my\u0161lence: d\u016fvod, pro\u010d se z\u00e1kazn\u00edk rozhodl k n\u00e1kupu, se skr\u00fdv\u00e1 v jeho n\u00e1kupn\u00ed cest\u011b. A bu\u010f se jedn\u00e1 o jednoduch\u00fd model (typu za n\u00e1kup m\u016f\u017ee posledn\u00ed zdroj v jeho cest\u011b), nebo sofistikovan\u011bj\u0161\u00ed \u201cdata-driven\u201d<\/strong> model, kter\u00fd tu hodnotu dopo\u010d\u00edt\u00e1v\u00e1.<\/p>\n

S atribuc\u00ed pracuje ka\u017ed\u00fd n\u00e1stroj webov\u00e9 analytiky a ka\u017ed\u00fd reklamn\u00ed syst\u00e9m. Zp\u016fsob pr\u00e1ce se ale z\u00e1sadn\u011b li\u0161\u00ed.<\/p>\n

Pokud nap\u0159\u00edklad vyhodnocujete kampan\u011b v Met\u011b, nejsp\u00ed\u0161 se v\u00e1m zd\u00e1, \u017ee jsou ta \u010d\u00edsla n\u011bjak nafoukl\u00e1. \u017de si prost\u011b Facebook sna\u017e\u00ed p\u0159isoudit v\u011bt\u0161\u00ed z\u00e1sluhy, ne\u017e jak\u00e9 mu pat\u0159\u00ed. Ten d\u016fvod je celkem jednoduch\u00fd. Meta a priori nevid\u00ed, z jak\u00fdch dal\u0161\u00edch zdroj\u016f se z\u00e1kazn\u00edk na web dost\u00e1v\u00e1. Zaznamen\u00e1, \u017ee u\u017eivatel proklikl kampa\u0148 A na Facebooku a pak, \u017ee za p\u00e1r dn\u00ed nakoupil na str\u00e1nce inzerenta. A to, \u017ee mezit\u00edm pro\u0161el p\u0159es Google, Heur\u00e9ku a Sklik remarketing, u\u017e Meta nev\u00ed. Pokud bychom tedy takto s\u010d\u00edtali konverze z r\u016fzn\u00fdch reklamn\u00edch n\u00e1stroj\u016f, z\u00e1konit\u011b bychom se dostali na v\u011bt\u0161\u00ed \u010d\u00edslo, ne\u017e kolik je celkov\u00fd po\u010det n\u00e1kup\u016f v administraci.<\/p>\n

To je pr\u00e1v\u011b v\u00fdhoda n\u00e1stroj\u016f typu Google Analytics, Mixpanel, Roivenue. T\u00edm, \u017ee po\u010d\u00edtaj\u00ed atribuci ucelen\u011b pro v\u0161echny kan\u00e1ly, d\u011blaj\u00ed tzv. deduplikaci konverz\u00ed<\/strong>. Nem\u016f\u017ee se tak st\u00e1t, \u017ee by byl v GA vy\u0161\u0161\u00ed po\u010det konverz\u00ed, ne\u017e kolik jich re\u00e1ln\u011b p\u0159i\u0161lo. Teda pokud nem\u00e1te n\u011bjakou chybu v m\u011b\u0159en\u00ed.<\/p>\n

Koncept atribuce se \u010dasto p\u0159ipodob\u0148uje k fotbalov\u00e9mu z\u00e1pasu. Nechceme d\u00e1vat kredit za v\u00fdhru jen \u00fato\u010dn\u00edk\u016fm, kte\u0159\u00ed st\u0159\u00edl\u00ed g\u00f3ly, ale chceme l\u00e9pe pochopit, jak se na v\u00fdh\u0159e pod\u00edleli i ostatn\u00ed hr\u00e1\u010di. Jen ta analogie funguje pouze za p\u0159edpokladu, \u017ee vyhodnocujeme prvn\u00ed a druh\u00fd polo\u010das zvl\u00e1\u0161\u0165. Bez toho, \u017ee bychom \u0159e\u0161ili jak\u00fdkoliv vztah mezi nimi. A nav\u00edc n\u00e1m u toho trochu na televizi vypad\u00e1v\u00e1 obraz, tak\u017ee ob\u010das p\u0159ehl\u00e9dneme p\u0159ihr\u00e1vku. Sta\u010d\u00ed n\u00e1m takov\u00fd zp\u016fsob vyhodnocov\u00e1n\u00ed, abychom na tomto z\u00e1klad\u011b d\u011blali rozhodnut\u00ed o zm\u011bn\u00e1ch v t\u00fdmu na dal\u0161\u00ed sez\u00f3nu? S nejv\u011bt\u0161\u00ed pravd\u011bpodobnost\u00ed ne. Ale dost mo\u017en\u00e1 n\u00e1m sta\u010d\u00ed k tomu, abychom v\u011bd\u011bli, koho v dan\u00e9m polo\u010dasu odvolat a poslat na lavi\u010dku.<\/p>\n

A u fotbalov\u00fdch analogi\u00ed je\u0161t\u011b chv\u00edli z\u016fstaneme. Vzpome\u0148te si na fotbal, kter\u00fd jsme hr\u00e1vali na z\u00e1kladce. T\u0159\u00edda se n\u011bjak rozd\u011bl\u00ed do dvou t\u00fdm\u016f a hraje. A nyn\u00ed si p\u0159edstavte Frantu. Ned\u00e1 se o n\u011bm \u0159\u00edct, \u017ee by mu fotbal n\u011bjak \u0161el, ale zato je dostate\u010dn\u011b chytr\u00fd na to, aby si spo\u010d\u00edtal, kter\u00fd t\u00fdm m\u00e1 v\u011bt\u0161\u00ed \u0161anci na v\u00fdhru a k n\u011bmu se p\u0159idal. A kdy\u017e na z\u00e1v\u011br \u0161koln\u00edho roku budeme vyhodnocovat nejlep\u0161\u00edho hr\u00e1\u010de \u010dist\u011b podle toho, kolikr\u00e1t byl \u010dlenem v\u00edt\u011bzn\u00e9ho t\u00fdmu, kdo mysl\u00edte, \u017ee si odnese trofej? P\u0159esn\u011b takto se chovaj\u00ed n\u011bkter\u00e9 platformy \u010di reklamn\u00ed modely. Sna\u017e\u00ed se spo\u010d\u00edtat, jak\u00e1 je \u0161ance, \u017ee u\u017eivatel nakoup\u00ed tak jako tak a zobrazuj\u00ed mu svou reklamu, aby se do jeho n\u00e1kupn\u00ed cesty je\u0161t\u011b zvl\u00e1dly nacpat.<\/p>\n

Co se m\u016f\u017ee zd\u00e1t paradoxn\u00ed, tak jednoduch\u00fd lastclick model (tedy v\u0161echny konverze n\u00e1le\u017e\u00ed posledn\u00edmu zn\u00e1m\u00e9mu zdroji) daleko m\u00ed\u0148 rozb\u00edj\u00ed, \u017ee nejsme schopni nap\u00e1rovat n\u00e1v\u0161t\u011bvy mezi za\u0159\u00edzen\u00edmi a \u017ee n\u00e1m ob\u010das chyb\u00ed data kv\u016fli cookies. Z\u00e1le\u017e\u00ed mu toti\u017e prim\u00e1rn\u011b pouze na t\u00e9 posledn\u00ed n\u00e1v\u0161t\u011bv\u011b a co se d\u011blo p\u0159edt\u00edm, jde mimo n\u011bj.<\/p>\n

Inkrement\u00e1ln\u00ed experimenty<\/h2>\n

\"inkrement\u00e1ln\u00ed<\/a><\/p>\n

Oproti atribuci jim v\u016fbec nez\u00e1le\u017e\u00ed na tom jak vypad\u00e1 n\u00e1kupn\u00ed cesta, kolik je na n\u00ed touchpoint\u016f ani \u017e\u00e1dn\u00e9 slo\u017eit\u00e9 rozpo\u010d\u00edt\u00e1v\u00e1n\u00ed. Zaj\u00edm\u00e1 je \u010dist\u011b, zda by ke konverzi do\u0161lo, \u010di nedo\u0161lo, kdybychom danou kampa\u0148 ned\u011blali - jak\u00fd je inkrement\u00e1ln\u00ed p\u0159\u00edr\u016fstek dan\u00e9 kampan\u011b<\/strong>. Co\u017e je hrozn\u011b d\u016fle\u017eit\u00fd rozd\u00edl oproti atribuci, kter\u00e1 automaticky rozpo\u010d\u00edt\u00e1v\u00e1 v\u0161echny konverze. Nejen ty nav\u00edc.<\/p>\n

Inkrement\u00e1ln\u00ed experimenty jsou zalo\u017een\u00e9 na causal inference a m\u016f\u017eeme k nim p\u0159istoupit v\u00edcero zp\u016fsoby. Zlat\u00fd standard jsou randomized controlled trials<\/strong>, tedy star\u00e9 dobr\u00e9 AB testov\u00e1n\u00ed. Jen to bohu\u017eel m\u00e1lokde u vyhodnocov\u00e1n\u00ed kampan\u00ed m\u016f\u017eeme vyu\u017e\u00edt (v\u00fdjimkou je nap\u0159. Meta conversion lift) a mus\u00edme p\u0159istoupit k jin\u00fdm metod\u00e1m, u kter\u00fdch si kontroln\u00ed skupinu mus\u00edme n\u011bjak vytvo\u0159it.<\/p>\n

Jednu z mo\u017enost\u00ed jak si vytvo\u0159it kontroln\u00ed skupinu vyu\u017e\u00edv\u00e1 CausalImpact<\/a>. Dejme tomu, \u017ee v \u0159\u00edjnu d\u011bl\u00e1me velkou akci, kterou chceme vyhodnotit. CausalImpact nejd\u0159\u00edv jakoby zapomene \u0159\u00edjnov\u00e1 data a predikuje, jak by \u0159\u00edjen dopadl, kdybychom akci ned\u011blali. Tuto predikci pak srovn\u00e1 s re\u00e1ln\u00fdmi daty a odhaduje jak velk\u00fd m\u011bla akce doopravdy vliv.<\/p>\n

\"causalimpact\"<\/a><\/p>\n

Opravdu elegantn\u00ed zp\u016fsob jak vyhodnotit dopad akce. Jedin\u00e1 nev\u00fdhoda je, \u017ee CausalImpact je zalo\u017een\u00fd na predik\u010dn\u00edm modelu, kter\u00fd m\u011bl v\u011bt\u0161\u00ed aktualizaci naposledy p\u0159ed 6 lety a mo\u017enosti predikce se za posledn\u00edch p\u00e1r let celkem posunuly. Proto mi d\u00e1v\u00e1 v\u011bt\u0161\u00ed smysl vyu\u017e\u00edt funkcionalitu Causal Impactu a aplikovat ji na p\u0159esn\u011bj\u0161\u00edm predik\u010dn\u00edm modelu.<\/p>\n

Jin\u00fd zp\u016fsob jak vytvo\u0159it kontroln\u00ed skupinu vyu\u017e\u00edv\u00e1 GeoLift<\/a>. Ten srovn\u00e1v\u00e1 v\u00fdsledky experimentu s tzv. syntetickou kontroln\u00ed skupinou<\/strong> zalo\u017eenou na \u00fazemn\u00edch celc\u00edch. Vybereme st\u00e1t \/ kraj \/ okres, na kter\u00e9m chceme experiment prov\u00e9st a GeoLift n\u00e1m dopo\u010d\u00edt\u00e1 takovou kombinaci ostatn\u00edch \u00fazem\u00ed, aby byly tomu p\u016fvodn\u00edmu p\u0159ed proveden\u00edm experimentu co nejpodobn\u011bj\u0161\u00ed.<\/p>\n

\"geolift\"<\/a><\/p>\n

T\u00edm se \u0159e\u0161\u00ed nap\u0159\u00edklad probl\u00e9m se siln\u00fdm hlavn\u00edm m\u011bstem<\/strong>. V kontroln\u00edm mixu jej pou\u017eijeme t\u0159eba jen s v\u00e1hou 0,1 tak, abychom chov\u00e1n\u00ed v hlavn\u00edm m\u011bst\u011b zohlednili, ale z\u00e1rove\u0148 aby n\u00e1m \u00fapln\u011b nezkreslilo v\u00fdsledky.<\/p>\n

Pr\u00e1ce s okresy a kraji je kv\u016fli spolehlivosti c\u00edlen\u00ed v\u017edycky slo\u017eit\u011bj\u0161\u00ed, a proto se s n\u00ed u n\u00e1s moc nepracuje. Pokud jsme ale do designu experimentu schopni dostat ten \u00fazemn\u00ed prvek, je interval spolehlivosti zpravidla v\u00fdrazn\u011b u\u017e\u0161\u00ed ne\u017e u CausalImpactu<\/strong>. Bude tedy p\u0159esn\u011bj\u0161\u00ed v odhadu efektu.<\/p>\n

Nejn\u00e1ro\u010dn\u011bj\u0161\u00ed na cel\u00e9m experimentu je p\u0159\u00edprava dat a zejm\u00e9na jeho spr\u00e1vn\u00fd design. V\u017edy je daleko jednodu\u0161\u0161\u00ed \u0159e\u0161it, jak experiment vyhodnocovat p\u0159edt\u00edm, ne\u017e dojde ke spu\u0161t\u011bn\u00ed kampan\u011b \/ p\u0159ekop\u00e1n\u00ed marketingov\u00e9ho rozpo\u010dtu ne\u017e ex post. \u010casto si tak m\u016f\u017eete u\u0161et\u0159it dost pr\u00e1ce s oh\u00fdb\u00e1n\u00edm modelu, kdy\u017e v\u0161echno od za\u010d\u00e1tku odlad\u00edte<\/a>.<\/p>\n

Media Mix Modelling<\/h2>\n

\"mmm\"<\/a><\/p>\n

Posledn\u00ed d\u00edl  triangu-skl\u00e1da\u010dky tvo\u0159\u00ed media mix modelling (MMM). O n\u011bm jste nejsp\u00ed\u0161 sly\u0161eli poprv\u00e9 v posledn\u00edch 3 letech, ale je zalo\u017een\u00fd na ekonometrick\u00fdch metod\u00e1ch, kter\u00e9 tu jsou s n\u00e1mi u\u017e p\u00e1r dek\u00e1d. Nen\u00ed to tak \u017e\u00e1dn\u00e1 rychlokva\u0161ka. Jen je te\u010f d\u00edky v\u011bt\u0161\u00ed v\u00fdpo\u010detn\u00ed s\u00edle daleko dostupn\u011bj\u0161\u00ed i pro b\u011b\u017en\u00e9 firmy a ne jen pro t\u011bch p\u00e1r nejv\u011bt\u0161\u00edch.<\/p>\n

P\u0159\u00edslib MMM je op\u011bt velmi elegantn\u00ed. Nepot\u0159ebujeme zn\u00e1t cestu z\u00e1kazn\u00edka, nejsme z\u00e1visl\u00ed na dom\u011b\u0159en\u00ed a propojen\u00ed ka\u017ed\u00e9 n\u00e1v\u0161t\u011bvy<\/strong>. Sta\u010d\u00ed n\u00e1m n\u011bjak\u00e9 exposure metriky (nej\u010dast\u011bji se pracuje s impresemi \u010di dosahem), \u00fatrata v r\u00e1mci dan\u00e9ho kan\u00e1lu a celkov\u00e9 tr\u017eby. Ne tr\u017eby dle kan\u00e1l\u016f pomoc\u00ed n\u011bjak\u00e9 atribuce, ale opravdu celkov\u00e9 tr\u017eby, kter\u00e9 v dan\u00e9m obdob\u00ed byly.<\/p>\n

A mezi t\u011bmito dv\u011bma stranami rovnice se nach\u00e1z\u00ed model, kter\u00fd se sna\u017e\u00ed namodelovat z\u00e1vislosti. Jak\u00fd vliv na celkov\u00e9 tr\u017eby m\u00e1 zv\u00fd\u0161en\u00ed impres\u00ed v Google Ads o 30 %? Jak\u00fd vliv bude m\u00edt sn\u00ed\u017een\u00ed rozpo\u010dtu pro Metu o 10 %?<\/p>\n

Zn\u00ed to trochu \u201ctoo good to be true\u201d. Nejsme z\u00e1visl\u00ed na cookies ani jin\u00e9m trackov\u00e1n\u00ed, odpadaj\u00ed tak probl\u00e9my s p\u0159\u00edli\u0161n\u00fdm z\u00e1sahem do ochrany osobn\u00edch \u00fadaj\u016f a sta\u010d\u00ed n\u00e1m \u010dist\u011b ta data, kter\u00e1 bereme z intern\u00edho syst\u00e9mu a z reklamn\u00edch platforem. Nav\u00edc oproti atribuci se sna\u017e\u00ed vy\u010d\u00edslit baseline (co bychom m\u011bli \u010dist\u011b setrva\u010dnost\u00ed a silou na\u0161\u00ed zna\u010dky) a co nav\u00edc n\u00e1m kampan\u011b p\u0159in\u00e1\u0161ej\u00ed. To p\u0159ece nem\u016f\u017ee b\u00fdt tak jednoduch\u00e9?<\/p>\n

Moje dv\u011b hlavn\u00ed p\u0159ipom\u00ednky od za\u010d\u00e1tku byly: v\u0161echny marketingov\u00e9 aktivity maj\u00ed n\u011bjak\u00e9 zpo\u017ed\u011bn\u00ed a m\u00edra zpo\u017ed\u011bn\u00ed se nav\u00edc u jednotliv\u00fdch kan\u00e1l\u016f dost li\u0161\u00ed (rychleji mi za\u010dnou p\u0159ich\u00e1zet tr\u017eby z kampan\u00ed ve vyhled\u00e1v\u00e1n\u00ed ne\u017e z televize, na druhou stranu ty z televize budou m\u00edt nejsp\u00ed\u0161 del\u0161\u00ed \u201cdojezd\u201d) a kampan\u011b nen\u00ed mo\u017en\u00e9 \u0161k\u00e1lovat do nekone\u010dna. Budou pracovat s n\u011bjakou m\u00edrou saturace, kter\u00e1 se bude postupn\u011b zv\u011bt\u0161ovat. Nesta\u010d\u00ed mi tedy prolo\u017eit na jedn\u00e9 stran\u011b imprese s \u00fatratou a na druh\u00e9 tr\u017eby a \u010dekat, \u017ee dostanu relevantn\u00ed v\u00fdsledek.<\/p>\n

\"carry<\/a><\/p>\n

Ale i s carry-over efektem<\/strong> (zpo\u017ed\u011bn\u00ed + setrva\u010dnost) a m\u00edrou saturace<\/strong> na\u0161t\u011bst\u00ed MMM po\u010d\u00edt\u00e1.<\/p>\n

Jen to u\u017e dost zvy\u0161uje slo\u017eitost modelu, a tak aby mohl n\u011bco relevantn\u00edho napo\u010d\u00edtat a ty z\u00e1vislosti trefit spr\u00e1vn\u011b, pot\u0159ebuje k tomu spoustu dat. V MMM se proto nikdy nedostaneme na takovou granularitu, s jakou pracujeme v r\u00e1mci atribuce<\/strong>. Pracujeme s n\u00edm tedy bu\u010f na \u00farovni zdroj\u016f, nebo logick\u00fdch kampa\u0148ov\u00fdch celk\u016f (nap\u0159. remarketing, vyhled\u00e1v\u00e1n\u00ed, social organic) a abychom k tomu dostali rozumn\u00e1 sez\u00f3nn\u00ed a trendov\u00e1 data, je naprost\u00e9 minimum p\u0159ipravovat model na alespo\u0148 2 letech dat<\/strong>. D\u00edky tomu, \u017ee n\u00e1m sta\u010d\u00ed data z reklamn\u00edch platforem a intern\u00edho syst\u00e9mu,se nemus\u00edme na\u0161t\u011bst\u00ed zaob\u00edrat takov\u00fdmi detaily, jako je nekonzistence mezi Universal Analytics a GA4 daty.<\/p>\n

MMM zn\u00ed skv\u011ble, pro\u010d ho tedy nevyu\u017e\u00edt jako n\u00e1hradu atribuce? Hlavn\u00ed d\u016fvod je ta n\u00edzk\u00e1 granularita (na \u00farovni kampan\u00ed n\u00e1m moc nepom\u016f\u017ee) a relativn\u011b dlouh\u00e1 doba reakce na zm\u011bny. Pokud nenastane opravdu velk\u00fd v\u00fdkyv, na denn\u00ed \u00farovni n\u00e1m MMM moc s vyhodnocov\u00e1n\u00edm nepom\u016f\u017ee.<\/p>\n

Triangulace<\/h2>\n

\"triangulace<\/a><\/p>\n

Co tedy s t\u00edm? Atribuce se nehod\u00ed pro v\u011bt\u0161\u00ed zm\u011bny typu rozd\u011blov\u00e1n\u00ed rozpo\u010dtu mezi kan\u00e1ly, inkrement\u00e1ln\u00ed experiment je n\u00e1ro\u010dn\u00fd na spu\u0161t\u011bn\u00ed a vyhodnocen\u00ed a MMM zase nevyu\u017eiji na b\u011b\u017enou optimalizaci kampan\u00ed.<\/p>\n

Odpov\u011bd\u00ed nen\u00ed jeden jednoduch\u00fd p\u0159\u00edstup, kter\u00fd n\u00e1m vy\u0159e\u0161\u00ed v\u0161echny ot\u00e1zky vyhodnocov\u00e1n\u00ed, ale um\u011bt si v\u017edy vybrat, kter\u00fd p\u0159\u00edstup se nejv\u00edc hod\u00ed pro probl\u00e9m, kter\u00fd \u0159e\u0161\u00edm<\/strong>.<\/p>\n

Pot\u0159ebuji optimalizovat kampa\u0148? Atribuce mi naprosto sta\u010d\u00ed, proto\u017ee prim\u00e1rn\u011b \u0159e\u0161\u00edm, jestli se zlep\u0161ila \/ zhor\u0161ila, \u010di jak se j\u00ed da\u0159\u00ed oproti stejn\u00e9mu typu kampan\u00ed (jakmile za\u010dnu porovn\u00e1vat odli\u0161n\u00e9 typy a kampan\u011b v r\u016fzn\u00fdch f\u00e1z\u00edch n\u00e1kupn\u00edho cyklu, u\u017eite\u010dnost jde celkem rychle k nule).<\/p>\n

\u0158e\u0161\u00edm efektivitu jednotliv\u00fdch kan\u00e1l\u016f \/ typ\u016f kampan\u00ed a chci dos\u00e1hnout lep\u0161\u00edho v\u00fdtlaku t\u00edm, \u017ee l\u00edp p\u0159esunu rozpo\u010det? Pou\u017eiji media mix modeling.<\/p>\n

V\u0161echno jsou to ale modely a nem\u011bli bychom jejich doporu\u010den\u00ed br\u00e1t slep\u011b jako bernou minci. S jejich ov\u011b\u0159en\u00edm a zp\u0159esn\u011bn\u00ed do budoucna n\u00e1m pom\u016f\u017eou inkrement\u00e1ln\u00ed experimenty. Tvrd\u00ed mi MMM, \u017ee Meta je o 40 % podhodnocen\u00e1? Spust\u00edm inkrement\u00e1ln\u00ed experiment, zjist\u00edm, \u017ee sice podhodnocen\u00e1 je, ale re\u00e1ln\u011b je to sp\u00ed\u0161 10-20 %. Uprav\u00edm d\u00e1l MMM model, aby Metu tolik nenadhodnocoval. Z toho mi pravd\u011bpodobn\u011b vypadne, \u017ee je podhodnocen\u00fd jin\u00fd kan\u00e1l a cel\u00fd proces m\u016f\u017eu zopakovat.<\/p>\n

tldr<\/h2>\n
    \n
  1. atribuce u\u017e nesta\u010d\u00ed na v\u011bt\u0161\u00ed \u00fakoly vyhodnocov\u00e1n\u00ed typu \u00fapravy rozpo\u010dt\u016f<\/li>\n
  2. nen\u00ed jin\u00fd kouzeln\u00fd n\u00e1stroj, kter\u00fd by ji nahradil ve v\u0161ech \u010d\u00e1stech vyhodnocov\u00e1n\u00ed<\/li>\n
  3. proto se pou\u017e\u00edv\u00e1 triangulace - kombinace p\u0159\u00edstup\u016f<\/li>\n
  4. atribuci st\u00e1le pou\u017e\u00edv\u00e1me na denn\u00ed optimalizaci kampan\u00ed<\/li>\n
  5. MMM m\u016f\u017ee zv\u00fd\u0161it celkovou efektivitu d\u00edky lep\u0161\u00edmu rozd\u011blen\u00ed marketingov\u00e9ho rozpo\u010dtu<\/li>\n
  6. inkrement\u00e1ln\u00ed experimenty ov\u011b\u0159\u00ed re\u00e1ln\u00fd dopad kan\u00e1lu \/ kampan\u011b a d\u00edky nim tak m\u016f\u017eeme zp\u0159es\u0148ovat atribu\u010dn\u00ed i media mix model<\/li>\n<\/ol>\n

    A d\u00e1l?<\/h2>\n

    Na webin\u00e1\u0159i jsme prob\u00edrali je\u0161t\u011b alternativy k atribu\u010dn\u00edmu modelov\u00e1n\u00ed, ale jako \u00favod do triangulace je to dostate\u010dn\u011b hutn\u00e9, tak si to nech\u00e1m zase pro dal\u0161\u00ed \u010dl\u00e1nek.<\/p>\n

    Nechcete si nechat uj\u00edt dal\u0161\u00ed tu\u010d\u0148\u00e1\u010d\u00ed webin\u00e1\u0159?<\/strong> Po datov\u00e9 p\u0159\u00edprav\u011b na expanzi<\/a> s Martinem Anderkem<\/strong> z GymBeam <\/strong>pokra\u010dujeme Media Mix Modellingem. P\u0159ihl\u00e1sit se m\u016f\u017eete tady<\/strong><\/a>.<\/p>\n","protected":false},"excerpt":{"rendered":"

    Co znamen\u00e1 triangulace u vyhodnocov\u00e1n\u00ed v marketingu? Co tvo\u0159\u00ed jej\u00ed sou\u010d\u00e1sti a pro\u010d u\u017e samotn\u00e9 atribuce nesta\u010d\u00ed? Adam sepsal hlavn\u00ed body z prosincov\u00e9ho tu\u010d\u0148\u00e1\u010d\u00edho webin\u00e1\u0159e o vyhodnocov\u00e1n\u00ed.<\/p>\n","protected":false},"author":7,"featured_media":8197,"comment_status":"open","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"wp_typography_post_enhancements_disabled":false,"footnotes":""},"categories":[370,70],"tags":[74,340,432,485,367],"class_list":["post-8179","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-analytika","category-marketing","tag-analytika","tag-data","tag-data-driven","tag-webinare","tag-webova-analytika"],"yoast_head":"\nTriangulace v marketingu - co to je a jak s t\u00edm pracovat | igloonet blog<\/title>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/igloonet.cz\/blog\/triangulace-v-marketingu-co-to-je-a-jak-s-tim-pracovat\/\" \/>\n<meta property=\"og:locale\" content=\"cs_CZ\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"Triangulace v marketingu - co to je a jak s t\u00edm pracovat | igloonet blog\" \/>\n<meta property=\"og:description\" content=\"Co znamen\u00e1 triangulace u vyhodnocov\u00e1n\u00ed v marketingu? Co tvo\u0159\u00ed jej\u00ed sou\u010d\u00e1sti a pro\u010d u\u017e samotn\u00e9 atribuce nesta\u010d\u00ed? Adam sepsal hlavn\u00ed body z prosincov\u00e9ho tu\u010d\u0148\u00e1\u010d\u00edho webin\u00e1\u0159e o vyhodnocov\u00e1n\u00ed.\" \/>\n<meta property=\"og:url\" content=\"https:\/\/igloonet.cz\/blog\/triangulace-v-marketingu-co-to-je-a-jak-s-tim-pracovat\/\" \/>\n<meta property=\"og:site_name\" content=\"igloonet blog\" \/>\n<meta property=\"article:publisher\" content=\"https:\/\/www.facebook.com\/igloonet\" \/>\n<meta property=\"article:published_time\" content=\"2024-12-12T11:07:27+00:00\" \/>\n<meta property=\"article:modified_time\" content=\"2025-04-02T10:11:26+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/igloonet.cz\/blog\/wp-content\/uploads\/2024\/12\/adam.png\" \/>\n\t<meta property=\"og:image:width\" content=\"1200\" \/>\n\t<meta property=\"og:image:height\" content=\"628\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/png\" \/>\n<meta name=\"author\" content=\"Adam \u0160ilhan\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:creator\" content=\"@NEmarketak\" \/>\n<meta name=\"twitter:site\" content=\"@igloonet\" \/>\n<meta name=\"twitter:label1\" content=\"Napsal(a)\" \/>\n\t<meta name=\"twitter:data1\" content=\"Adam \u0160ilhan\" \/>\n\t<meta name=\"twitter:label2\" content=\"Odhadovan\u00e1 doba \u010dten\u00ed\" \/>\n\t<meta name=\"twitter:data2\" content=\"12 minut\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\/\/schema.org\",\"@graph\":[{\"@type\":\"Article\",\"@id\":\"https:\/\/igloonet.cz\/blog\/triangulace-v-marketingu-co-to-je-a-jak-s-tim-pracovat\/#article\",\"isPartOf\":{\"@id\":\"https:\/\/igloonet.cz\/blog\/triangulace-v-marketingu-co-to-je-a-jak-s-tim-pracovat\/\"},\"author\":{\"name\":\"Adam \u0160ilhan\",\"@id\":\"https:\/\/igloonet.cz\/blog\/#\/schema\/person\/a1aada43c3b360cc63f8e2c01809d429\"},\"headline\":\"Triangulace v marketingu - co to je a jak s t\u00edm pracovat\",\"datePublished\":\"2024-12-12T11:07:27+00:00\",\"dateModified\":\"2025-04-02T10:11:26+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\/\/igloonet.cz\/blog\/triangulace-v-marketingu-co-to-je-a-jak-s-tim-pracovat\/\"},\"wordCount\":2522,\"commentCount\":0,\"publisher\":{\"@id\":\"https:\/\/igloonet.cz\/blog\/#organization\"},\"image\":{\"@id\":\"https:\/\/igloonet.cz\/blog\/triangulace-v-marketingu-co-to-je-a-jak-s-tim-pracovat\/#primaryimage\"},\"thumbnailUrl\":\"https:\/\/igloonet.cz\/blog\/wp-content\/uploads\/2024\/12\/adam.png\",\"keywords\":[\"Analytika\",\"data\",\"data-driven\",\"webin\u00e1\u0159e\",\"webov\u00e1 analytika\"],\"articleSection\":[\"Analytika\",\"Marketing\"],\"inLanguage\":\"cs\",\"potentialAction\":[{\"@type\":\"CommentAction\",\"name\":\"Comment\",\"target\":[\"https:\/\/igloonet.cz\/blog\/triangulace-v-marketingu-co-to-je-a-jak-s-tim-pracovat\/#respond\"]}]},{\"@type\":\"WebPage\",\"@id\":\"https:\/\/igloonet.cz\/blog\/triangulace-v-marketingu-co-to-je-a-jak-s-tim-pracovat\/\",\"url\":\"https:\/\/igloonet.cz\/blog\/triangulace-v-marketingu-co-to-je-a-jak-s-tim-pracovat\/\",\"name\":\"Triangulace v marketingu - co to je a jak s t\u00edm pracovat | igloonet blog\",\"isPartOf\":{\"@id\":\"https:\/\/igloonet.cz\/blog\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\/\/igloonet.cz\/blog\/triangulace-v-marketingu-co-to-je-a-jak-s-tim-pracovat\/#primaryimage\"},\"image\":{\"@id\":\"https:\/\/igloonet.cz\/blog\/triangulace-v-marketingu-co-to-je-a-jak-s-tim-pracovat\/#primaryimage\"},\"thumbnailUrl\":\"https:\/\/igloonet.cz\/blog\/wp-content\/uploads\/2024\/12\/adam.png\",\"datePublished\":\"2024-12-12T11:07:27+00:00\",\"dateModified\":\"2025-04-02T10:11:26+00:00\",\"breadcrumb\":{\"@id\":\"https:\/\/igloonet.cz\/blog\/triangulace-v-marketingu-co-to-je-a-jak-s-tim-pracovat\/#breadcrumb\"},\"inLanguage\":\"cs\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\/\/igloonet.cz\/blog\/triangulace-v-marketingu-co-to-je-a-jak-s-tim-pracovat\/\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"cs\",\"@id\":\"https:\/\/igloonet.cz\/blog\/triangulace-v-marketingu-co-to-je-a-jak-s-tim-pracovat\/#primaryimage\",\"url\":\"https:\/\/igloonet.cz\/blog\/wp-content\/uploads\/2024\/12\/adam.png\",\"contentUrl\":\"https:\/\/igloonet.cz\/blog\/wp-content\/uploads\/2024\/12\/adam.png\",\"width\":1200,\"height\":628,\"caption\":\"Triangulace v marketingu\"},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\/\/igloonet.cz\/blog\/triangulace-v-marketingu-co-to-je-a-jak-s-tim-pracovat\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Analytika\",\"item\":\"https:\/\/igloonet.cz\/blog\/rubriky\/analytika\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"Triangulace v marketingu – co to je a jak s t\u00edm pracovat\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\/\/igloonet.cz\/blog\/#website\",\"url\":\"https:\/\/igloonet.cz\/blog\/\",\"name\":\"igloonet blog\",\"description\":\"\",\"publisher\":{\"@id\":\"https:\/\/igloonet.cz\/blog\/#organization\"},\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\/\/igloonet.cz\/blog\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"cs\"},{\"@type\":\"Organization\",\"@id\":\"https:\/\/igloonet.cz\/blog\/#organization\",\"name\":\"igloonet\",\"url\":\"https:\/\/igloonet.cz\/blog\/\",\"logo\":{\"@type\":\"ImageObject\",\"inLanguage\":\"cs\",\"@id\":\"https:\/\/igloonet.cz\/blog\/#\/schema\/logo\/image\/\",\"url\":\"https:\/\/igloonet.cz\/blog\/wp-content\/uploads\/2015\/07\/RGB_LVPC_igloonet.png\",\"contentUrl\":\"https:\/\/igloonet.cz\/blog\/wp-content\/uploads\/2015\/07\/RGB_LVPC_igloonet.png\",\"width\":650,\"height\":465,\"caption\":\"igloonet\"},\"image\":{\"@id\":\"https:\/\/igloonet.cz\/blog\/#\/schema\/logo\/image\/\"},\"sameAs\":[\"https:\/\/www.facebook.com\/igloonet\",\"https:\/\/x.com\/igloonet\",\"https:\/\/instagram.com\/igloonet\/\",\"https:\/\/www.linkedin.com\/company\/igloonet\"]},{\"@type\":\"Person\",\"@id\":\"https:\/\/igloonet.cz\/blog\/#\/schema\/person\/a1aada43c3b360cc63f8e2c01809d429\",\"name\":\"Adam \u0160ilhan\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"cs\",\"@id\":\"https:\/\/igloonet.cz\/blog\/#\/schema\/person\/image\/\",\"url\":\"https:\/\/secure.gravatar.com\/avatar\/aa03ce5f493a168de15b09b9835eb6f92c752b564ac1f18cf0ff09c3863bcba2?s=96&d=retro&r=g\",\"contentUrl\":\"https:\/\/secure.gravatar.com\/avatar\/aa03ce5f493a168de15b09b9835eb6f92c752b564ac1f18cf0ff09c3863bcba2?s=96&d=retro&r=g\",\"caption\":\"Adam \u0160ilhan\"},\"description\":\"Vedu igloo t\u00fdm analytik\u016f a s klienty \u0159e\u0161\u00edm tu z\u00e1bavnou \u010d\u00e1st na pomez\u00ed dat a marketingu. V posledn\u00ed dob\u011b m\u011b chytnul freediving, tak m\u011b ob\u010das m\u016f\u017eete vid\u011bt topit.\",\"sameAs\":[\"https:\/\/igloonet.cz\/adam\",\"https:\/\/www.linkedin.com\/in\/adamsilhan\/\",\"https:\/\/x.com\/NEmarketak\"],\"url\":\"https:\/\/igloonet.cz\/blog\/author\/adam\/\"}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"Triangulace v marketingu - co to je a jak s t\u00edm pracovat | igloonet blog","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/igloonet.cz\/blog\/triangulace-v-marketingu-co-to-je-a-jak-s-tim-pracovat\/","og_locale":"cs_CZ","og_type":"article","og_title":"Triangulace v marketingu - co to je a jak s t\u00edm pracovat | igloonet blog","og_description":"Co znamen\u00e1 triangulace u vyhodnocov\u00e1n\u00ed v marketingu? Co tvo\u0159\u00ed jej\u00ed sou\u010d\u00e1sti a pro\u010d u\u017e samotn\u00e9 atribuce nesta\u010d\u00ed? Adam sepsal hlavn\u00ed body z prosincov\u00e9ho tu\u010d\u0148\u00e1\u010d\u00edho webin\u00e1\u0159e o vyhodnocov\u00e1n\u00ed.","og_url":"https:\/\/igloonet.cz\/blog\/triangulace-v-marketingu-co-to-je-a-jak-s-tim-pracovat\/","og_site_name":"igloonet blog","article_publisher":"https:\/\/www.facebook.com\/igloonet","article_published_time":"2024-12-12T11:07:27+00:00","article_modified_time":"2025-04-02T10:11:26+00:00","og_image":[{"width":1200,"height":628,"url":"https:\/\/igloonet.cz\/blog\/wp-content\/uploads\/2024\/12\/adam.png","type":"image\/png"}],"author":"Adam \u0160ilhan","twitter_card":"summary_large_image","twitter_creator":"@NEmarketak","twitter_site":"@igloonet","twitter_misc":{"Napsal(a)":"Adam \u0160ilhan","Odhadovan\u00e1 doba \u010dten\u00ed":"12 minut"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"Article","@id":"https:\/\/igloonet.cz\/blog\/triangulace-v-marketingu-co-to-je-a-jak-s-tim-pracovat\/#article","isPartOf":{"@id":"https:\/\/igloonet.cz\/blog\/triangulace-v-marketingu-co-to-je-a-jak-s-tim-pracovat\/"},"author":{"name":"Adam \u0160ilhan","@id":"https:\/\/igloonet.cz\/blog\/#\/schema\/person\/a1aada43c3b360cc63f8e2c01809d429"},"headline":"Triangulace v marketingu - co to je a jak s t\u00edm pracovat","datePublished":"2024-12-12T11:07:27+00:00","dateModified":"2025-04-02T10:11:26+00:00","mainEntityOfPage":{"@id":"https:\/\/igloonet.cz\/blog\/triangulace-v-marketingu-co-to-je-a-jak-s-tim-pracovat\/"},"wordCount":2522,"commentCount":0,"publisher":{"@id":"https:\/\/igloonet.cz\/blog\/#organization"},"image":{"@id":"https:\/\/igloonet.cz\/blog\/triangulace-v-marketingu-co-to-je-a-jak-s-tim-pracovat\/#primaryimage"},"thumbnailUrl":"https:\/\/igloonet.cz\/blog\/wp-content\/uploads\/2024\/12\/adam.png","keywords":["Analytika","data","data-driven","webin\u00e1\u0159e","webov\u00e1 analytika"],"articleSection":["Analytika","Marketing"],"inLanguage":"cs","potentialAction":[{"@type":"CommentAction","name":"Comment","target":["https:\/\/igloonet.cz\/blog\/triangulace-v-marketingu-co-to-je-a-jak-s-tim-pracovat\/#respond"]}]},{"@type":"WebPage","@id":"https:\/\/igloonet.cz\/blog\/triangulace-v-marketingu-co-to-je-a-jak-s-tim-pracovat\/","url":"https:\/\/igloonet.cz\/blog\/triangulace-v-marketingu-co-to-je-a-jak-s-tim-pracovat\/","name":"Triangulace v marketingu - co to je a jak s t\u00edm pracovat | igloonet blog","isPartOf":{"@id":"https:\/\/igloonet.cz\/blog\/#website"},"primaryImageOfPage":{"@id":"https:\/\/igloonet.cz\/blog\/triangulace-v-marketingu-co-to-je-a-jak-s-tim-pracovat\/#primaryimage"},"image":{"@id":"https:\/\/igloonet.cz\/blog\/triangulace-v-marketingu-co-to-je-a-jak-s-tim-pracovat\/#primaryimage"},"thumbnailUrl":"https:\/\/igloonet.cz\/blog\/wp-content\/uploads\/2024\/12\/adam.png","datePublished":"2024-12-12T11:07:27+00:00","dateModified":"2025-04-02T10:11:26+00:00","breadcrumb":{"@id":"https:\/\/igloonet.cz\/blog\/triangulace-v-marketingu-co-to-je-a-jak-s-tim-pracovat\/#breadcrumb"},"inLanguage":"cs","potentialAction":[{"@type":"ReadAction","target":["https:\/\/igloonet.cz\/blog\/triangulace-v-marketingu-co-to-je-a-jak-s-tim-pracovat\/"]}]},{"@type":"ImageObject","inLanguage":"cs","@id":"https:\/\/igloonet.cz\/blog\/triangulace-v-marketingu-co-to-je-a-jak-s-tim-pracovat\/#primaryimage","url":"https:\/\/igloonet.cz\/blog\/wp-content\/uploads\/2024\/12\/adam.png","contentUrl":"https:\/\/igloonet.cz\/blog\/wp-content\/uploads\/2024\/12\/adam.png","width":1200,"height":628,"caption":"Triangulace v marketingu"},{"@type":"BreadcrumbList","@id":"https:\/\/igloonet.cz\/blog\/triangulace-v-marketingu-co-to-je-a-jak-s-tim-pracovat\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Analytika","item":"https:\/\/igloonet.cz\/blog\/rubriky\/analytika\/"},{"@type":"ListItem","position":2,"name":"Triangulace v marketingu – co to je a jak s t\u00edm pracovat"}]},{"@type":"WebSite","@id":"https:\/\/igloonet.cz\/blog\/#website","url":"https:\/\/igloonet.cz\/blog\/","name":"igloonet blog","description":"","publisher":{"@id":"https:\/\/igloonet.cz\/blog\/#organization"},"potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/igloonet.cz\/blog\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"cs"},{"@type":"Organization","@id":"https:\/\/igloonet.cz\/blog\/#organization","name":"igloonet","url":"https:\/\/igloonet.cz\/blog\/","logo":{"@type":"ImageObject","inLanguage":"cs","@id":"https:\/\/igloonet.cz\/blog\/#\/schema\/logo\/image\/","url":"https:\/\/igloonet.cz\/blog\/wp-content\/uploads\/2015\/07\/RGB_LVPC_igloonet.png","contentUrl":"https:\/\/igloonet.cz\/blog\/wp-content\/uploads\/2015\/07\/RGB_LVPC_igloonet.png","width":650,"height":465,"caption":"igloonet"},"image":{"@id":"https:\/\/igloonet.cz\/blog\/#\/schema\/logo\/image\/"},"sameAs":["https:\/\/www.facebook.com\/igloonet","https:\/\/x.com\/igloonet","https:\/\/instagram.com\/igloonet\/","https:\/\/www.linkedin.com\/company\/igloonet"]},{"@type":"Person","@id":"https:\/\/igloonet.cz\/blog\/#\/schema\/person\/a1aada43c3b360cc63f8e2c01809d429","name":"Adam \u0160ilhan","image":{"@type":"ImageObject","inLanguage":"cs","@id":"https:\/\/igloonet.cz\/blog\/#\/schema\/person\/image\/","url":"https:\/\/secure.gravatar.com\/avatar\/aa03ce5f493a168de15b09b9835eb6f92c752b564ac1f18cf0ff09c3863bcba2?s=96&d=retro&r=g","contentUrl":"https:\/\/secure.gravatar.com\/avatar\/aa03ce5f493a168de15b09b9835eb6f92c752b564ac1f18cf0ff09c3863bcba2?s=96&d=retro&r=g","caption":"Adam \u0160ilhan"},"description":"Vedu igloo t\u00fdm analytik\u016f a s klienty \u0159e\u0161\u00edm tu z\u00e1bavnou \u010d\u00e1st na pomez\u00ed dat a marketingu. V posledn\u00ed dob\u011b m\u011b chytnul freediving, tak m\u011b ob\u010das m\u016f\u017eete vid\u011bt topit.","sameAs":["https:\/\/igloonet.cz\/adam","https:\/\/www.linkedin.com\/in\/adamsilhan\/","https:\/\/x.com\/NEmarketak"],"url":"https:\/\/igloonet.cz\/blog\/author\/adam\/"}]}},"amp_enabled":true,"_links":{"self":[{"href":"https:\/\/igloonet.cz\/blog\/wp-json\/wp\/v2\/posts\/8179","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/igloonet.cz\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/igloonet.cz\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/igloonet.cz\/blog\/wp-json\/wp\/v2\/users\/7"}],"replies":[{"embeddable":true,"href":"https:\/\/igloonet.cz\/blog\/wp-json\/wp\/v2\/comments?post=8179"}],"version-history":[{"count":12,"href":"https:\/\/igloonet.cz\/blog\/wp-json\/wp\/v2\/posts\/8179\/revisions"}],"predecessor-version":[{"id":8231,"href":"https:\/\/igloonet.cz\/blog\/wp-json\/wp\/v2\/posts\/8179\/revisions\/8231"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/igloonet.cz\/blog\/wp-json\/wp\/v2\/media\/8197"}],"wp:attachment":[{"href":"https:\/\/igloonet.cz\/blog\/wp-json\/wp\/v2\/media?parent=8179"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/igloonet.cz\/blog\/wp-json\/wp\/v2\/categories?post=8179"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/igloonet.cz\/blog\/wp-json\/wp\/v2\/tags?post=8179"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}