The impact of automation-driven performance max campaign, smart bidding strattegis and tenscore third-party marketing automation tool on conversion, Roas and other KPIs in Google ADS PPC management

MDDM Mas e ’s deg ee P og am in Da a-D i en Ma ke ing THE IMPACT OF AUTOMATION-DRIVEN PERFORMANCE MAX CAMPAIGN, SMART BIDDING STRATEGIES, AND TENSCORE THIRD PARTY MARKETING AUTOMATION TOOL ON CONVERSION, ROAS, AND OTHER KPIs IN GOOGLE ADS PPC MANAGEMENT OKEKE MOSES Disse a ion p esen ed as a pa ial equi emen o ob aining he Mas e Deg ee P og am in Da a-D i en Ma ke ing NOVA In o ma ion Managemen School Ins i u o Supe io de Es a ís ica e Ges ão de In o mação Uni e sidade No a de Lisboa NOVA In o ma ion Managemen School Ins i u o Supe io de Es a ís ica e Ges ão de In o mação Uni e sidade No a de Lisboa THE IMPACT OF AUTOMATION-DRIVEN PERFORMANCE MAX CAMPAIGN, SMART BIDDING STRATEGIES, AND TENSCORE THIRD-PARTY MARKETING AUTOMATION TOOL ON CONVERSION, ROAS, AND OTHER KPIs IN GOOGLE ADS PPC MANAGEMENT by Okeke Moses Mas e Thesis p esen ed as a pa ial equi emen o ob aining he Mas e ’s deg ee in Da a-D i en Ma ke ing, wi h a specializa ion in digi al ma ke ing and analy ics. Supe iso : Ma lon Dalmo o July, 2023 STATEMENT OF INTEGRITY I he eby decla e ha ing conduc ed his academic wo k wi h in eg i y. I con i m ha I ha e no used plagia ism o any o m o undue use o in o ma ion o alsi ica ion o esul s along he p ocess leading o i s elabo a ion. I u he decla e ha I ha e ully acknowledge he Rules o Conduc and Code o Hono om he NOVA In o ma ion Managemen School. Okeke Moses Lisbon, July 10 h, 2023 i DEDICATION This hesis is dedica ed o my belo ed daugh e - T easu e Chisom Okeke- Moses o he lo e and belie in me. This p ojec will se e as a mo i a ional piece ha will p opel he o a ain a highe g ound in any ca ee o he choice. ACKNOWLEDGEMENTS I wan o exp ess my deep g a i ude o my wi e, M s. Sa ah N Okeke-Moses, o he suppo and unde s anding h oughou he pe iod o his p ojec . She is a blend o s eng h and a pe soni ica ion o op imism who sha es a co e alue o ha ing a uni ed on o he p og ess o ou amily. I also wan o app ecia e my child en-T easu e, Dominion, Zinnachimdinma and Chimamanda who endu ed my absence du ing he pe iod o his p ojec . Secondly, I wan o hank my supe iso , P o esso Ma lon Dalmo o, o cons an ly being a ailable o a end o my ques ions whene e I beckoned on him. I wan o app ecia e his guidance, knowledge, and he expe iences he sha ed wi h me o ensu e ha his p ojec becomes a eali y. i ABSTRACT Google ads PPC ad e ising is cons an ly e ol ing and becoming mo e dynamic by inco po a ing new ools o au oma ize ma ke ing campaigns. S aying cu en wi h he la es ends can be challenging as all digi al ma ke ing channels a e in oducing new echnological ends o help PPC ad e ise s mee hei campaign goals. This s udy explo es how a new au oma ed campaign impac s key pe o mance indica o s, especially he ROAS and he numbe o con e sions o e-comme ce ad e ising on Google ads. The s udy ound ha he pe o mance max campaign has a mo e posi i e impac on ge ing a g ea e numbe o con e sions han he s anda d shopping campaign. Rega ding he impac on ROAS, his s udy concludes ha au oma ed bidding s a egy ( a ge ROAS) is a key ac o ha in luences he ROAS o a campaign and no he ype o campaign in ol ed. The s udy p oposes u he echnological de elopmen o upg ading Google Ads s anda d shopping campaigns such ha PPC manage s can use sma bidding s a egies on s anda d shopping campaigns du ing he campaign se up wi hou wai ing o ge ce ain con e sions be o e swi ching om manual o au oma ed bidding s a egies. The s udy also ound ha Tensco e, as a hi d-pa y MA, helps educe he ime in ol ed in es uc u ing keywo ds and ad g oups in sea ch campaigns. KEYWORDS Google Ads, PPC, Ma ke ing Au oma ion, Pe o mance Max campaign, ROAS, Con e sions, Sma bidding s a egy, Shopping campaign, Thi d-Pa y Ma ke ing Au oma ion. ii INDEX 1 INTRODUCTION---------------------------------------------------------------------------------------------- 1 1.1 BACKGROUND 1 1.2 THIRD-PARTY MA OVERVIEW 3 1.3 RESEARCH QUESTION 4 1.4 KEY PERFORMANCE INDICATORS (KPI) 5 1.5 RESEARCH OBJECTIVES 5 2 LITERATURE REVIEW ---------------------------------------------------------------------------------------- 8 2.1 IMPRESSION METRIC 16 2.2 THE CLICKS AND CLICK-THROUGH RATE --------------------------------------------------------------- 17 2.3 CONVERSIONS And CONVERSION RATES -------------------------------------------------------------- 18 2.4 ROAS 19 2.5 Google Ads Au oma ed Bidding (Sma bidding s a egies) --------------------------------------- 20 2.6 MAXIMIZE ROAS BIDDING STRATEGY (Ta ge ROAS Sma Bidding S a egy) ----------------- 22 2.7 CHALLENGES Wi h MA 23 2.8 GOOGLE PERFORMANCE MAX CAMPAIGN ----------------------------------------------------------- 24 2.9 THIRD-PARTY MA 26 2.10 TENSCORE AS A THIRD-PARTY MA 29 2.11 PROJECT MODEL AND HYPOTHESIS 31 2.12 PROJECT MODEL EXPLAINED 33 3 METHODOLOGY -------------------------------------------------------------------------------------------- 35 3.1 RESEARCH METHOD 35 3.2 METHODOLOGY APPROACH AND CONCEPT FROM PILOT STUDY -------------------------------- 35 3.3 THE EXPERIMENT STRUCTURE 36 3.4 CONVERSION AND ROAS TRACKING 37 3.5 GOOGLE ANALYTIC INTEGRATION 37 3.6 Pmax CAMPAIGN VS STANDARD SHOPPING CAMPAIGN ------------------------------------------ 37 3.7 PMAX VS S anda d Sea ch Campaign 40 3.8 INTEGRATING TENSCORE WITH GOOGLE ADS PLATFORM 40 3.9 EXPERIMENT DURATION AND KPI SAMPLE SIZES --------------------------------------------------- 41 3.10 MATHEMATICAL MODEL FOR EXPECTED RESULTS VALIDATION --------------------------------- 42 3.11 INTEGRATING RELIABILITY IN THE EXPERIMENTATION PROCESS -------------------------------- 43 iii 4 RESULTS ------------------------------------------------------------------------------------------------------ 45 4.1 CONTEXT DESCRIPTIONS 45 4.2 RESEARCH RESULTS OF ALL KPI FOR TACTICAL GEAR E-COMMERCE CAMPAIGNS ----------- 45 4.3 CONVERSION METRICS RESULTS FROM THE TACTICAL GEAR E-COMMERCE CAMPAIGNS EXPERIMENTS 45 4.4 ROAS METRICS RESULTS FROM THE TACTICAL GEAR E-COMMERCE CAMPAIGNS EXPERIMENTS 46 4.5 CLICKS AND CLICK RATES METRICS RESULTS FROM THE SPLIT CAMPAIGNS ------------------- 48 4.6 RESULTS OF ALL KPI FOR LEADS CAMPAIGN ---------------------------------------------------------- 50 5 DISCUSSIONS ------------------------------------------------------------------------------------------------ 52 5.1 SUMMARY 52 5.2 EVALUATING THE IMPACT OF GOOGLE ADS PMAX CAMPAIGN ON ROAS ---------------------- 52 5.3 EVALUATING THE IMPACT OF GOOGLE ADS PMAX CAMPAIGN ON CONVERSION COMPARED TO STANDARD SHOPPING CAMPAIGN --------------------------------------------------------- 53 5.4 EVALUATING THE IMPACT OF GOOGLE ADS SMART BIDDING STRATEGY (TARGET ROAS) ON ROAS 54 5.5 Valida ing Hypo hesis-H3 56 5.6 RESEARCH MODEL AND HYPOTHESIS ADOPTION --------------------------------------------------- 56 5.7 CONTRIBUTIONS 57 6 CONCLUSION ------------------------------------------------------------------------------------------------ 59 6.1 RESEARCH LIMITATION 59 6.2 INDICATIONS FOR FURTHER STUDIES 60 BIBLIOGRAPHY ------------------------------------------------------------------------------------------------------ 62 APPENDIX ------------------------------------------------------------------------------------------------------------- 66 ix LIST OF FIGURES Figu e 2.1 Pmax campaign use su ey epo adap ed om Sea ch engine land ............................ 26 Figu e 2.2 A causal Model diag am o sea ch ads a a que y le el. .................................................. 32 Figu e 2.3 Hypo hesis 1 model diag am showing he ela ionship be ween Pmax, Con e sions & SSC. 32 Figu e 2.4 Hypo hesis 2 model diag am showing he ela ionship be ween Pmax, ROAS& SSC ............................................................................................................................................................ 32 Figu e 2.5 Hypo hesis 3 model diag am showing he ela ionship be ween sma bidding, ROAS & manual bidding. .................................................................................................................................. 33 Figu e 2.6 Model diag am illus a ing Google Ads’ Pmax Campaign, S anda d Shopping/Sea ch campaign, sma bidding, manual bidding, and hei impac on Con e sions and ROAS .................. 33 Figu e 3.1 50% SPLIT: Con ol & T ea men G oups .......................................................................... 39 Figu e 3.2 Pmax and SSC as T ea men & Con ol G oups Respec i ely............................................ 39 Figu e 3.3 Tensco e in e ace showing campaign s uc u e sco e on Google Ads ............................ 40 Figu e 4.1 Cha showing he Numbe o con e sion esul s om Pmax & SSC ................................. 46 Figu e 4.2 Con e sion T ends om Pmax and SSC Expe imen om Ma 20-June 15 ...................... 46 Figu e 4.3 Con e sion Ra es om Pmax and SSC Campaign Expe imen s ........................................ 46 Figu e 4.4 ROAS Sha e be ween PMAX & SSC .................................................................................... 47 Figu e 4.5 ROAS T ends Pmax and ROAS T ends S anda d Shopping ............................................. 47 Figu e 4.6 Con e sion Values and Ad Spend om Pmax & SSC -spli expe imen ............................ 48 Figu e 4.7 cha showing he numbe o con e sions be ween Pmax & SSC Campaign ................... 48 Figu e 4.8 Clicks & Click Ra es Pmax Vs. SSC Campaign Resul s ......................................................... 49 Figu e 4.9 Imp essions om Pmax and SSC wi h De ice D ill Down a e 90 days ............................ 50 Figu e 4.10 Click end om Pmax Vs. SSC Spli Expe ience a e a pe iod o 90 days...................... 50 Figu e 5.1 ROAS alue and end om manual bidding s a egy in SSC ............................................ 55 Figu e 5.2 ROAS alue and end om a ge ROAS sma bidding s a egy in SSC .......................... 55 Figu e 5.3 SSC ROAS T ends Showing a Rise in ROAS a e swi ching o Sma Bidding om May 9 h, 2023 ............................................................................................................................................ 56 16 he digi al ad e ising. Depending on he goal he ad e ise wan s o achie e wi h he in eg a ion o he MA such as b and awa eness, sales, o leads gene a ions a e calcula ed by hese me ics. KPI hence o h e e ed a e ime se ies da a because hey consis o disc e e ime da a me ics whose alues change wi h ime (Tappé & Mülle , 2022). Hence, his s udy is aimed a in es iga ing he e ec o he impac o hese au oma ed asks on key pe o mance index o Google ads campaigns. Acco ding o Allazo (2020), key pe o mance me ics a e moni o ed o e alua e he e iciency o ce ain campaigns. The equi emen o he use o MA is o see posi i e ou comes on he common pe o mance me ics. This s udy will in es iga e he impac o bo h in-house and hi d-pa y MA on he ollowing pe o mance me ics: imp essions, click- h ough a e (hence o h e e ed o as CTR), con e sions, e u n on ad spend (hence o h e e ed as ROAS). 1.5 RESEARCH OBJECTIVES F om he beginning o he ise o he In e ne o online ma ke ing, PPC ad e ising has been an ac i e segmen in main ma ke ing ac i i ies, me ging ou sides (Allazo , 2020). The ou sides o PPC in eg a e he PPC channels (Google Ads, Bings Ads, Yahoo, Baidu, e c.), he PPC ne wo ks (display and sea ch ne wo k), he PPC manage s known as ad e ise s and he in e ne use s who a e he a ge s o PPC ad e isings. The need o MA is o in eg a e hese ou media o achie e enhanced esul s o all ou lis ed ac o s. Fo he ad e ise s, i is how o achie e hei ma ke ing objec i es such as con e sions, e u n on ad spend (ROAS he ea e ) and o he key pe o mance me ics. The use s will be conce ned abou ha ing good expe iences while sea ching o he p oduc s and se ices o hei in e es . Hence he esea ch objec i es o his s udy a e ou lined as lis ed below: a) To e alua e he impac o he implemen a ion Google ads au oma ed campaign wi h ML ea u e known as “Pe o mance Max Campaign”) on ROAS and Con e sions plus o he pe o mance me ics in B2C PPC managemen ; b) To s udy he impac o in eg a ing Tensco e hi d-pa y MA wi h Google Ads PPC on he ease o achie ing epe i i e asks in Google Ads PPC managemen ; c) To e alua e he impac o Google ads’ sma bidding s a egy – “maximize con e sion alue wi h a ge ROAS” on ROAS in B2C Campaign managemen ; 17 This esea ch will exploi hese pas esea ch gaps. I will be based on he echnological equi emen o s ay upda ed wi h he ending skills needed o d i e eme ging ends in digi al ma ke ing au oma ion. I was emba ked wi h he aim o ha e a eal- ime app oach o he implemen a ions and in eg a ion o hi d-pa y MA ools in Google Ads PPC managemen wi h a iew o using he da a de i ed om he s udy o make in o med da a-d i en decisions o PPC manage s who would like o in es iga e u he on he need o in eg a e hese au oma ed ea u es in hei ma ke ing s a egies. I will also p o ide academic insigh o u he esea ch. Ma ke ing s a egis s can bene i om academic ma ke ing’s hough leade ship o lea n how o ans o m hese challenges in o s a egic oppo uni ies o compe i i e ad an age (Plangge e al., 2022). I is in ended o p o ide p ima y insigh s d awn om he p ac ical in eg a ion o Tensco e as a hi d-pa y MA. I will alida e i s ask- elie ing abili y o c ea ing Google ads PPC campaigns and he claim o imp o ed quali y sco e. Fu he mo e, he s udy will di e deep o es ablish he impac o implemen ing he Google ads au oma ed campaign known as Pe o mance Max campaign on ROAS in B2C agains he manual s anda d shopping campaign in he Google Ads PPC pla o m. 18 2 LITERATURE REVIEW A s udy on he p e ailing ends in digi al ma ke ing es ablished ha he use o a i icial in elligence and machine lea ning o d i e digi al ma ke ing au oma ion is he cu en p e ailing ends in digi al ma ke ing (Ko ane e al., 2019). Ma ke ing au oma ion ools ha e become an essen ial pa o mode n digi al ma ke ing s a egies. These ools help businesses au oma e epe i i e asks, such as lead gene a ion, email campaigns, social media managemen and PPC campaign managemen o imp o e hei ma ke ing e iciency and pe o mance (Hammoud e al., 2019). Thi d-pa y ma ke ing au oma ion ools in eg a e wi h a ious pla o ms, including Google Ads, o enhance hei capabili ies and a ge pe o mance me ics. The in eg a ion o hi d-pa y ma ke ing au oma ion ools wi h Google Ads can signi ican ly impac con e sions and e u n on ad spend (ROAS) o businesses. This li e a u e e iew aims o explo e he impac o In-buil MA ea u es o he Google ads Pe o mance Max campaign, Google Ads ROAS sma bidding s a egy and he impac o in eg a ing Tensco e hi d-pa y ma ke ing au oma ion ools wi h Google Ads wi h a iew o s udying hei impac on con e sions, ROAS and o he pe o mance me ics wi h a en ion o ease o doing epe i i e asks highligh ing ele an esea ch and indings. This li e a u e e iew will explo e he cu en esea ch on he impac o in eg a ing hi d-pa y ma ke ing au oma ion ools wi h Google Ads on a ge pe o mance me ics, wi h a ocus on con e sions and ROAS me ics in B2C. I will del e in o echnological inno a ion in AI d i en au oma ion p ocesses in Google Ads PPC managemen . The wo d ‘Au oma ion” as ega ds o digi al space has been iden i ied as a p ocess o elimina e manual inpu s. I wo ks wi h da a o execu e epe i i e asks using a p e-de ined se o ules (Anicca Webina s, 2023). Hence, Ma ke ing au oma ion e e s o he use o echnology o s eamline and au oma e epe i i e ma ke ing asks. I is aimed a helping businesses imp o e he e iciency and e ec i eness o hei ma ke ing e o s and enables ma ke e s o ocus on highe -le el s a egies (Singh e al., 2018). I is he u iliza ion o ma ke ing echnology o au oma e p ocesses in digi al ma ke ing ope a ions, such as he alloca ion o campaign budge , campaign c ea ion, and gene a ion o enhanced pe o mance epo s (Biegel, 2009). 19 Ano he de ini ion o MA by pas s udies e e s o i as he use o so wa e and echnology o au oma e epe i i e ma ke ing asks such as email ma ke ing, social media managemen , lead gene a ion, and cus ome segmen a ion (Rosenbe g & Cze niawski, 2016). This sea ch made a s ong e e ence o MA as a “so wa e’ as many pas s udies ha e linked he phenomenon as such; his is u he suppo ed by ano he s udy which de ines MA as a so wa e pla o m ha is buil o p o ide indi idual and pe sonalized con en based on speci ic o de s placed by use s (Nilsson & Tsakmaki, 2019). MA has a la ge scope o capabili ies, such as main aining cu en and p ospec i e cus ome da abases, moni o ing and analyzing cus ome beha io s on he websi e o on c oss de ice in e ace and au oma ion o PPC campaign managemen (Świeczak, 2019). In his esea ch, i is impo an o emphasize ha MA is no only es ic ed o email ma ke ing ope a ion bu should no be con used wi h CRM as bo h a e no he same as many a e made o belie e (Hammoud e al., 2019). A s udy by Mu phy (2018) on he eadiness o SMEs o ma ke ing au oma ion ound ha 43% o esponden s deployed MA on PPC ad e ising. The main goal o ma ke ing au oma ion is o inc ease he e iciency and e ec i eness o ma ke ing e o s, esul ing in highe con e sion a es, inc eased e enue, and imp o ed cus ome engagemen (Dwye , Tanne , & Leich weis, 2017). Ma ke ing au oma ion is also de ined as so wa e ha au oma es basic ma ke ing ope a ions (Desai, 2019). I is a so wa e solu ion ha uses an in e ne acili y in he o m o so wa e as a se ice (hence o h e e ed o as (SAAS) o au oma e ma ke ing p ocesses (Hamalainem, 2020). The o igin o he ph ase ‘ma ke ing au oma ion’ can be aced o a pape p esen ed by John DC Li le a he 2001 Choice Symposium (Heimbach e al., 2015). Acco ding o Li le (2001), MA e e s o he use o in e ne -based au oma ed ma ke ing decision suppo . Li le (2001) o mula ed he i e amewo ks o ma ke ing au oma ion o include: 1. da a inpu collec ed by p e ious in e ac ion o use s (Seme ádo á & Weinlich, 2020), 2. eal- ime decision ules, 3. upda e o decision ules, 4. eedback o si e managemen and, 5—s a egy choice. Fu he a gumen s o make hese amewo ks possible sugges ed ha ad e ise s mus ha ness hei cus ome s' digi al oo p in s and ac o hem in o hei con e sion unnels (Heimbach e al., 2015; Li le, 2001). Today, These digi al oo p in s a e e e ed o as cus ome ’s da a, and he use o i g a i a es owa d da a-d i en decision ma ke ing. This makes he con ex o MA o cen e on he au oma ion o epe i i e ma ke ing asks and he au oma ion 20 o da a-d i en decisions and s a egies. Acco ding o Seme ádo á & Weinlich (2020), ma ke ing au oma ion has been gaining a en ion om ma ke ing s a egis s and academics’ ci cle, and i s p ojec ed ha he MA indus y will see a ise in in es men o up o $25billion by he end o 2023 (Mu phy, 2018). Ma ke ing au oma ion has se e al bene i s o businesses. Fi s , i allows companies o sa e ime and esou ces by au oma ing epe i i e asks, which ees up ma ke e s o ocus on mo e s a egic ac i i ies. Second, i helps companies imp o e hei lead-gene a ion e o s by acking cus ome s' beha io s p o iding a ge ed messaging and pe sonalized con en o po en ial cus ome s (Co sa o e al., 2021). The wo aspec s o au oma ion o epe i i e ask and acking o consume beha io s leads o e icien ma ke ing s a egies ha esul s in g owing sales and p o i abili y (Co sa o e al., 2021). Thi d, i allows companies o nu u e leads o e ime h ough au oma ed wo k lows, esul ing in highe con e sion a es and inc eased cus ome loyal y (Nilsson & Tsakmaki, 2019). The leads sco ing unc ion aspec o he MA in ol es au oma ic assigning o poin s g ades o websi e isi o s depending on he a ious pages isi ed such as any aluable lead unnels (Seme ádo á & Weinlich, 2020). Ma ke ing au oma ion acco ding o a s udy has been p o en o a ec buye s' decision- making p ocess h ough b and awa eness (Nilsson & Tsakmaki, 2019). Finally, ma ke ing au oma ion p o ides aluable insigh s in o cus ome beha io and p e e ences, enabling companies o make da a- d i en decisions and imp o e hei o e all ma ke ing s a egy (Nai & Gup a, 2020). Ma ke ing au oma ion and i s implemen a ion ha e limi a ions, especially among middle- and small-scale businesses. These challenges ange om cus ome da a collec ion and he expe ise needed o implemen and in eg a e he MA ((Seme ádo á & Weinlich, 2020). While ma ke ing au oma ion has many bene i s, he e a e also se e al challenges ha companies may ace when implemen ing i . One o he bigges challenges is ensu ing he echnology is p ope ly in eg a ed wi h exis ing sys ems and p ocesses, which can equi e signi ican ime and esou ces (Chen, Chen, & Hsiao, 2019). In addi ion, companies mus ensu e ha hei da a is accu a e and up o da e o a ge e ec i ely and segmen cus ome s. Ano he challenge is a oiding o e -au oma ion, which can lead o impe sonal o i ele an messaging and esul in lowe engagemen a es (Rosenbe g & 21 Cze niawski, 2016). Companies should ollow se e al bes p ac ices o ensu e ha ma ke ing au oma ion e o s a e success ul. Fi s , hey should s a by clea ly de ining hei ma ke ing goals and objec i es, and hen selec he app op ia e au oma ion ools and pla o ms o achie e hem (Nilsson & Tsakmaki, 2019). Second, companies should ocus on p o iding pe sonalized and ele an con en o cus ome s based on hei p e e ences and beha io , a he han simply sending gene ic messaging (Kim, Lee, & Yoon, 2018). Thi d, companies should egula ly analyze hei da a and me ics o op imize hei ma ke ing e o s and make da a-d i en decisions (Nai & Gup a, 2020). Ma ke ing au oma ion has become an in eg al pa o digi al ma ke ing and i ’s widely used o s eamline and op imize a ious ma ke ing ac i i ies. I has been seen as a ma ke ing ideology o deli e pe sonalized ma ke ing and inc ease con e sion a es (Rae, 2016). One o he a eas whe e MA has made a signi ican impac is PPC managemen . Acco ding Anicca Webina (2023). MA is used in PPC o s eamline p ocesses such as accoun se up and managemen bu equi es some le el o human in e en ions o se up. PPC (pay-pe -click) digi al ad e ising has become inc easingly popula in ecen yea s as a way o businesses o each and a ge speci ic audiences online. PPC ad e ising allows businesses o c ea e and display ads on di e en pla o ms, such as sea ch engines, social media, and websi es, and only pay publishe s only when a use clicks on hem (Desai, 2019). S udies ha e shown ha PPC ad e ising can e ec i ely inc ease websi e a ic, gene a e leads, and boos sales (Mu phy, 2018; Smi h, 2019). One o he p ima y jus i ica ions o using PPC in digi al ma ke ing is i s abili y o deli e highly a ge ed a ic o a websi e. This is achie ed using keywo ds and ad placemen a ge ing, which allow ad e ise s o a ge speci ic demog aphics and in e es s. Acco ding o a s udy by Google, PPC isi o s a e 50% mo e likely o make a pu chase han o ganic isi o s. PPC me ics a e easie o measu e han SEO, which makes i easie o ack ROI and budge alloca ion (Val e Me e , 2023). Howe e , businesses need o ha e a well-planned and execu ed s a egy o see he desi ed esul s (Plangge e al., 2022). A s udy by Szymanski & Lipinski (2018) examined he ac o s ha in luence he e ec i eness o PPC ads. The esea che s ound ha ad ele ance, ad posi ion, and ad o ma we e he 22 mos signi ican ac o s in de e mining he success o a PPC campaign. Addi ionally, he s udy ound ha a ge ed keywo ds and ad copy also played impo an oles in he success o a campaign. Pay-pe - click (PPC) ad e ising has become a popula ma ke ing channel o businesses o all sizes due o i s a ge ed and measu able na u e. In a s udy by Ru z e al. (2009), he au ho s examined he impac o ad copy and landing page design on PPC campaign pe o mance. The esea che s ound ha a combina ion o compelling ad copy and an e ec i e landing page could signi ican ly imp o e he con e sion a e o a PPC campaign. The s udy also sugges ed ha es ing di e en ad copy and landing page designs can help ad e ise s iden i y he mos e ec i e combina ions. Ano he ac o ha a ec s PPC ad e ising is he digi al audience a ge ing which can be ha nessed wi h digi al analy ics (Kabi aj & Joghee, 2023). The au ho s examined he impac o audience a ge ing on PPC campaign pe o mance. The esea che s ound ha a ge ing speci ic audiences can imp o e campaign pe o mance by inc easing ele ance and engagemen . The s udy also sugges ed ha using da a om p e ious campaigns o iden i y high-pe o ming audience segmen s can help ad e ise s achie e g ea e success in u u e campaigns. E ec i e PPC managemen is c ucial o achie ing success in digi al ad e ising. The s udies e iewed in his li e a u e e iew highligh he impo ance o ac o s such as ad ele ance, ad copy, landing page design, machine lea ning, and audience a ge ing in imp o ing campaign pe o mance. Ad e ise s should con inuously moni o and op imize hei PPC campaigns o achie e he bes esul s. All hese ac o s when in eg a ed can lead o an enhanced PPC pe o mance in ela ion o a ge KPIs. Howe e , managing a success ul PPC campaign can be ime-consuming and complex. To help ma ke e s op imize hei PPC campaigns and achie e hei desi ed e u n on ad spend (ROAS), a ious ools and s a egies ha e been de eloped, including ma ke ing au oma ion and a ious bidding s a egies such as ROAS bidding. A s udy by Miklosik e al. (2019) es ima ed ha 2.5 quin illion o da a a e being gene a ed on a daily basis o e he in e ne and he esea ch explo ed he use o machine lea ning echniques in PPC managemen as an au oma ed p edic ion ool in digi al ad e ising. The au ho s ound ha machine lea ning algo i hms can be used o op imize bidding s a egies and ad a ge ing, esul ing in imp o ed campaign pe o mance. The s udy also highligh ed he impo ance o inco po a ing his o ical campaign da a in o machine lea ning models o op imal pe o mance, whe e 23 bo h inbuil and hi d-pa y MA a e ound o be e y e en ul. Thus, machine lea ning is a key unc ion ha empowe s PPC campaign op imiza ion h ough au oma ion (Hall, 2022). Thus, in he esea ch conclusion made by Miklosik e al. (2019), machine lea ning can be used o p edic digi al ad e ising by ex ac ing la ge amoun s o da a o da a-d i en decisions in PPC ma ke ing. The machine lea ning p ocess used in ma ke ing au oma ion can ei he be passi e o ac i e lea ning p ocesses whe e passi e au oma ed lea ning in ol es he use o pas campaign da a o clicks eams. In con as , ac i e lea ning in ol es di ec ly asking ques ions o en ich he ma ke ing au oma ion da a (Jä inen & Taiminen, 2016). To complemen hese in e nal ma ke ing au oma ion sys em ac i i ies, hi d-pa y ma ke ing au oma ion is made a ailable by hi d-pa y endo s. Thi d-pa y ma ke ing au oma ion ools a e designed o help ad e ise s op imize and au oma e hei PPC campaigns. Acco ding o a s udy by Wilson (2019), hi d-pa y ma ke ing au oma ion has a signi ican impac on PPC managemen as i helps ad e ise s s eamline and op imize hei campaigns, leading o inc eased ROAS and imp o ed campaign pe o mance. Ad anced hi d-pa y MA u ilizes cloud ma ke ing au oma ion so wa e hos ed by hi d-pa y companies wi h he ad an ages o sa ing ad e ising he cos o hos ing hei own se e , da a in eg a ion, and sys em main enance as he ad e ise s will only in eg a e hese al eady made sys ems in o hei digi al ma ke ing channels (Smi h, 2016). The s udy also ound ha hi d-pa y ma ke ing au oma ion ools p o ide ad e ise s wi h ad anced ea u es, such as bid op imiza ion and audience a ge ing, which help inc ease hei ads' ele ance. Many s udies ha e in es iga ed he impac o in eg a ing hi d-pa y au oma ion ools wi h PPC managemen and ha e made some no able esea ch conclusions. Fo ins ance, Hall (2022) p o ides insigh in o au oma ion ools in au oma ed PPC, including hi d-pa y ma ke ing au oma ion ools. The au ho concludes ha hi d-pa y au oma ion ools can imp o e PPC pe o mance by educing manual asks and imp o ing ad a ge ing. O he esea che s in es iga ed he impac o ma ke ing au oma ion on PPC ad e ising and concluded ha hi d-pa y au oma ion ools can imp o e PPC pe o mance by educing cos s and imp o ing a ge ing (Smi h, 2016). This au ho also examines he impac o PPC ad e ising on business g ow h and concludes ha hi d-pa y ma ke ing au oma ion ools can imp o e PPC pe o mance by educing cos s, imp o ing a ge ing, and inc easing con e sion a es. Hall (2022) 24 examines he impac o ma ke ing au oma ion on PPC ad e ising and concludes ha au oma ion ools can imp o e PPC pe o mance by educing manual asks, imp o ing ad a ge ing, and inc easing con e sion a es by op imizing campaigns using his o ical con e sion da a. A u he s udy suppo ed he indings ha au oma ion ools can imp o e PPC pe o mance as a iable ma ke ing channel wi h he abili y o enhance o ganiza ional s a us and boos hei equi y alues h ough s eng hening he cus ome ela ionship by using ecip oci y in communica ions, which is a undamen al and e ec i e p ocess o cus ome acquisi ion and e en ion mechanism (Świeczak, 2019). I is e iden ha in Google ads PPC channels, cus ome acquisi ion is boos ed using e a ge ing campaigns while le e aging he cus ome audience segmen a ions o deli e loyal y p og am h ough au oma ed a ge ing. One o he mos common ypes o PPC is Google Ads, which allows ad e ise s o bid o esul s anywhe e on he Google sea ch engine (Desai, 2019). Google Ads PPC is one o he mos popula ad e ising pla o ms, wi h 92.96% o all sea ch engine ad e ising ma ke sha e wo ldwide (S a is a, 2023). Google Ads gene a es an a e age o $8 o e e y $1 spen on ad e ising (Google, 2021). Fu he mo e, Google Ads is esponsible o 28% o all websi e isi s om paid sea ch ad e ising (Sma phone Use Beha io Repo , 2021). Many s udies ha e shown he e ec i eness o Google Ads PPC as a leading PPC channel. Fo example, a s udy by Wo dS eam ound ha Google Ads PPC has an a e age click- h ough a e (CTR) o 3.17% o sea ch ads and 0.46% o display Ads, wi h abou 82% o Ad spend alloca ed o Google ads, acco ding o hei epo pulled om 18,000 ad e ise s on Localiq (McCo mick, 2023). Addi ionally, a s udy ound ha 64.6% o people click on Google Ads when hey a e looking o buy i ems online, wi h paid sea ches bea ing ou o ganics in he a io o 2;1 o sea ch keywo ds wi h comme cial in en s in he Uni ed S a es (Kim, 2022). Resea ch by Allazo (2020) s a ed ha cus ome s om Google sea ch engine ad e ising ha e mo e ansac ions han o he s om ela ed channels. Many e e ences in academic and indus y publica ions also suppo he e ec i eness o Google Ads PPC as a leading PPC channel. Fo example, a s udy by Cho and Lee (2018) ound ha Google Ads PPC is one o he mos e ec i e o ms o online ad e ising. Addi ionally, a s udy by To ge son (2022) ound ha Google Ads PPC is a cos -e ec i e ad e ising me hod o small businesses. One o he p ima y 25 bene i s o Google Ads PPC is i s e ec i eness in d i ing websi e a ic and con e sions. Ano he s udy by Wo dS eam ound ha businesses using Google Ads PPC had a con e sion a e o 3.75%, signi ican ly highe han he indus y a e age o 2.35% (Wo dS eam 2018). Acco ding o Hanapin Ma ke ing, Google Ads PPC gene a ed an a e age o $2 in e enue o e e y $1 spen on ad e ising (Hanapin Ma ke ing, 2018). These indings sugges ha Google Ads PPC is a highly e ec i e o m o digi al ad e ising, deli e ing a s ong e u n on in es men (ROI) o businesses. Based on hese dis inguishing ac o s o Google Ads PPC on in o ma ion ex ac ed om he pas esea ch abo e, his li e a u e e iew explo es he e ec i eness o Google Ads PPC as a leading PPC channel wi h he aim o in es iga ing he impac o i s inhe en o in-house au oma ion ea u es in “Google ads pe o mance max campaign,” i s sma bidding s a egy and in eg a ion wi h Tensco e as a hi d- pa y au oma ion ool on Con e sions, ROAS and o he key pe o mance indica o s (he ea e e e ed as KPI). Se e al s udies ha e highligh ed he bene i s o ma ke ing au oma ion in Google Ads PPC managemen o B2C businesses. Acco ding o a s udy by Geh e al. (2018), ma ke ing au oma ion can help businesses sa e ime, inc ease e iciency, and imp o e cus ome engagemen . The s udy ound ha businesses ha used ma ke ing au oma ion in hei Google Ads PPC campaigns could gene a e highe click- h ough a es (CTR), educe bounce a es, and inc ease con e sion a es. Ano he s udy by Digi al Ma ke ing Ins i u e (2022) ound ha ma ke ing au oma ion can help businesses pe sonalize hei messages o cus ome s, leading o highe engagemen and con e sion a es. The s udy ound ha 83% o ma ke e s who used ma ke ing au oma ion we e able o inc ease hei con e sion a es h ough leads nu u ing. This s udy c i ically e iews pas li e a u e ega ding he e ec o MA on some pe o mance indica o s on Google Ads PPC managemen . 2.1 IMPRESSION METRIC An imp ession is said o be made when an ad eques by a use is comple ed and displayed (Shandilya e al., 2023). The i s KPI mus exis be o e all o he KPIs can be measu ed in digi al ad e ising because wi hou imp essions, he e won’ be a click o con e sions. An imp ession is coun ed when an ad is se ed as pa o sea ch engine esul s in sea ch engine ma ke ing (hence o h e e ed o as SEM) and his is based on he numbe o imes he keywo ds a e sea ched which is 32 has been p o en o ha e esul ed in a 15% inc ease in ROI compa ed o manual bidding (Baldwin e al., 2020). Ano he s udy ound ha using Maximize ROAS bidding s a egy esul ed in a 12 % inc ease in ROAS compa ed o manual bidding (Baldwin e al., 2020). I is impo an o no e. This esea ch p oposes a hypo hesis ha ha pe o mance max campaign, a Google ads AI’s d i en au oma ed campaign has mo e posi i e impac o numbe o con e sions han s anda d shopping/sea ch campaigns in B2C. H3: Google Ads’ sma bidding -maximize con e sion alue wi h a ge ROAS s a egy has mo e signi ican posi i e impac on ROAS han manual bidding s a egy. 2.7 CHALLENGES Wi h MA Despi e he bene i s o ma ke ing au oma ion in Google Ads PPC managemen , implemen ing i can be challenging o B2C businesses. One o he main challenges is he lack o expe ise and esou ces. Acco ding o a s udy by Aspa wa e al. (2020), many B2C businesses do no ha e he necessa y expe ise o esou ces o implemen ma ke ing au oma ion in hei Google Ads PPC campaigns. The s udy ound ha businesses ha lacked he necessa y esou ces and expe ise we e mo e likely o expe ience challenges such as poo da a quali y, low engagemen a es, and educed ROI. Ano he challenge o implemen ing ma ke ing au oma ion is he need o ongoing op imiza ion. Acco ding o a s udy by Kim e al. (2021), businesses ha used ma ke ing au oma ion in hei Google Ads PPC campaigns had o con inuously op imize hei campaigns o ensu e ha hey we e e ec i e. The s udy ound ha businesses ha did no op imize hei campaigns we e mo e likely o expe ience educed engagemen a es and lowe con e sion a es. Bes P ac ices o B2C Businesses: To o e come he challenges o implemen ing ma ke ing au oma ion in Google Ads PPC managemen , B2C businesses can ollow bes p ac ices. Acco ding o a s udy by S ini asan e al. (2019), some o he bes p ac ices o B2C businesses include: a) de ining clea objec i es and goals o he Google Ads PPC campaign; b) segmen ing he a ge audience based on 33 demog aphics, beha io , and in e es s; c) de eloping pe sonalized messages ha esona e wi h he a ge audience. Using a a ie y o ad o ma s such as ex ads, display ads, and ideo ads and implemen ing con e sion acking o measu e he e ec i eness o he campaign and con inuously op imizing he campaign based on he pe o mance da a. As digi al ad e ising con inues o e ol e, Google Ads PPC will likely see some changes. One po en ial de elopmen is he use o a i icial in elligence (AI) and machine language (ML) o op imize ad a ge ing and bidding s a egies. A s udy by Kenshoo ound ha using AI and ML in Google PPC campaigns esul ed in a 30% inc ease in con e sion and a 33% dec ease in cos pe acquisi ion (CPA) (Kenshoo, 2019). The Pe o mance Max campaign is one o he Google Ads campaigns wi h ull in-house ma ke ing au oma ion ea u es powe ed by ML algo i hm. 2.8 GOOGLE PERFORMANCE MAX CAMPAIGN Google Ads Pe o mance Max campaign is a new global-based campaign ha allows ad e ise s o place ads au oma ically on all Google ads in en o y by s ee ing he campaign inpu s h ough ma ke ing au oma ion. I uses sma bidding, a ibu ion echnology, machine lea ning, and au o-gene a ed asse s o enhance campaign audience signals (Google, n.d.). Pe o mance Max campaign shows ads on all Google ne wo k anging om display, sea ch, YouTube, shopping, and disco e y. The main di e ence be ween o he ypes o Google ads campaigns and pe o mance is based on he ac ha pe o mance ads will au oma e he in o ma ion been p o ided by he ad e ise (Bishop, 2021). Pe o mance Max campaign is designed o help businesses achie e hei desi ed ROAS by using ad anced machine lea ning algo i hm o au oma ically adjus bids and budge in eal- ime based on he pe o mance o each ads (Flo es, n.d.). A ecen su ey by sea ch engine land e ealed ha 67% o ad e ise s who pa icipa ed in he su ey a e cu en ly using pe o mance max campaign. 62% o he esponden demons a ed hei us a ion wi h he keywo ds and sea ch e ms epo ing insigh as he e is no keywo d epo ing as sea ches a e based on au oma ed a ge ing (Ad hena, 2023). The pe o mance max campaign has only wo bidding s a egies o selec om, including maximizing con e sion alues and con e sions (Sande s, 2022). The e o e, he ad e ise has he op ion o using a ge cos pe con e sion (TCPA) o 34 se a equi ed amoun he wan s o pay o a gi en con e sion o a ge cos o ROAS o se bid o he maximize con e sion alues o a gi en con e sion alues. Acco ding o Google (Google, n.d.), he bene i s o a pe o mance campaign a e unlocking new cus ome s ac oss he channels, d i ing be e esul s ac oss ad e ise s’ goals, simpli ying campaign managemen , and ge ing mo e anspa ency insigh s. Kim (2021) p o ides an o e iew o he ea u es o Google Ads pe o mance and how hey can bene i ma ke e s. The au ho highligh s he pla o m’s machine-lea ning capabili ies ha op imize ad placemen , bidding, and c ea i e elemen s o d i e be e pe o mance. Ano he s udy ha explo ed how Google Ads Pe o mance Max can help ma ke e s imp o e hei e u n on in es men (ROI) es ablished ha Pe o mance Max can au oma e Ad c ea ion and a ge ing, educing he ime and e o s equi ed o PPC managemen (Amanda, 2022). Pe o mance campaign has g ea is p ojec ed o ha e g ea impac on PPC managemen elying on how he Pe o mance Max campaign au oma ion ea u es can educe he need o manual op imiza ion and allowing ma ke e s o ocus on s a egy and c ea i e. The Pe o mance Max Campaign u ilizes he powe o au oma ion o iden i y he bes op imiza ion s a egy in B2C (Amanda, 2022). I is o no e ha pe o mance max campaign does no imply 100% au oma ed asse s in e ms o campaign se up as he e a e le el o con ol o be e ec ed by he PPC manage such as budge alloca ion, ad scheduling, loca ion a ge ing, and asse se up (Bishop, 2021). Pe o mance Max Campaign can also educe he lea ning cu e by au oma ing campaign se up and op imiza ion, h e eb y e du c in g h e le a ni n g c u e a nd he i me e q ui ed o P PC m an ag e me n b e c aus e G o o g le a d s w i ll a u o m a e h e a s se s h e PP C m an ag e p o i de s (Bishop, 2021). Ma ke ing au oma ion can ei he be used as an in-house ool o a hi d-pa y ool in PPC managemen . Acco ding o a epo by Ma ke s and Ma ke s, he global ma ke ing au oma ion so wa e is expec ed o g ow om $3.3 billion in 2019 o $6.4 billion by 2024, wi h a p ojec ion o $9.5 billion by 2027, his g ow h is being d i en by inc easing demand o ma ke ing au oma ion ools, pa icula ly he hi d-pa y ma ke ing au oma ion ools (Ma ke And Ma ke s, n.d.). 35 Figu e 2.1 Pmax campaign use su ey epo adap ed om Sea ch engine land. 2.9 THIRD-PARTY MA Ma ke ing channels such as Google Ads, Me a o Business and o he s ha e hei own in e nal o inhouse-ma ke ing au oma ion sys em, while o he s a e being con igu ed using hi d-pa y companies’ ools. Thi d-pa y MA ools help o complemen in-house au oma ion ools. Thi d-pa y ma ke ing au oma ion e e s o he use o au oma ion ools de eloped by hi d-pa y companies as opposed o being de eloped in-house by he business i sel . These ools a e designed o help businesses imp o e he e iciency and e ec i eness o hei ma ke ing e o s. They can include ea u es such as email au oma ion, lead nu u ing, and PPC campaign managemen (Pa el, 2022). Acco ding o Me a (2023), hi d-pa y au oma ion se ices a e hose o e ed by ex e nal endo s ha allow independen so wa e in eg a ion o hei so wa e as a se ice wi hin a wo kplace, whe eby his so wa e can hen be ins alled by any cus ome o deli e alue au oma ion and p o ide so wa e and ools o business o au oma e hei ma ke ing ope a ions. Thi d-pa y MA se ices ha e been shown o p o ide nume ous bene i s o businesses including inc eased e iciency, cos sa ings and imp o ed cus ome engagemen . Acco ding o s udy by Ma ke o, businesses ha use hi d-pa y MA so wa e can see 14.5% inc ease in sales p oduc i i y and a 12.2% educ ion in ma ke ing o e head cos s (Ma ke o, 2017). One o he main bene i s o using hi d-pa y ma ke ing au oma ion ools o PPC managemen is imp o ed e iciency. Se e al s udies ha e ound ha hese ools can help businesses sa e ime and educe he esou ces equi ed o manage PPC campaigns. Fo example, a s udy by Kenshoo ound ha businesses ha 36 used hei ma ke ing au oma ion ool could sa e up o 30% o hei ime on PPC managemen asks (Kenshoo, 2020). Ano he s udy by Wo dS eam ound ha businesses ha used hei ma ke ing au oma ion ool we e able o sa e up o 20 hou s pe week on PPC managemen asks (Wo dS eam, 2020). Ano he s udy by Abe deen G oup ound ha businesses using ma ke ing au oma ion so wa e expe ienced a 451% inc ease in quali ied leads and a 53% highe con e sion a e han hose who did no use au oma ion ools (Abe deen G oup, 2014). Thi d-pa y ma ke ing au oma ion ools can also help businesses imp o e hei PPC campaigns' e u n on in es men (ROI). Se e al s udies ha e ound ha businesses ha used hese ools achie ed highe con e sion a es, lowe cos -pe -click (CPC), and highe click- h ough a es (CTR). Fo example, a s udy by AdEsp esso ound ha businesses ha used hei ma ke ing au oma ion ool we e able o achie e a 62% lowe CPC and a 42% highe CTR han businesses ha did no use hei ool (AdEsp esso, 2020). Ano he s udy by Kenshoo ound ha businesses ha used hei ma ke ing au oma ion ool we e able o achie e a 9% highe con e sion a e and a 30% lowe CPA han businesses ha did no use hei ool (Kenshoo, 2020). Thi d-pa y ma ke ing au oma ion can ha e a signi ican impac on businesses, especially in e ms o ma ke ing e iciency, cus ome engagemen , and e enue g ow h. Acco ding o a s udy by Sales o ce, businesses ha use ma ke ing au oma ion so wa e can see a 34% inc ease in lead gene a ion, a 42% inc ease in quali ied leads, and a 25% inc ease in e enue g ow h (Sales o ce, 2017). While hi d-pa y ma ke ing au oma ion can p o ide many bene i s, he e a e also some disad an ages. One o he main challenges wi h using hi d-pa y au oma ion ools is ha hey may no be ully in eg a ed wi hin a business's exis ing sys ems, which can lead o da a inconsis encies and e o s. Addi ionally, hi d-pa y au oma ion ools may be cos ly, especially o small businesses, and may equi e a signi ican in es men in aining and suppo o be used e ec i ely). Howe e , i is essen ial o no e ha he e ec i eness o hi d-pa y au oma ion ools depends on he quali y o he so wa e, he business's indus y, and he business's ma ke ing goals. Thi d-pa y ma ke ing au oma ion se ices o e many bene i s o businesses, including inc eased e iciency, cos sa ings, and imp o ed cus ome engagemen . 37 Howe e , he e a e also some challenges and conside a ions o conside when using hi d- pa y au oma ion ools. Businesses should ca e ully e alua e hese ools' e ec i eness and in eg a ion wi h exis ing sys ems be o e in es ing in ma ke ing au oma ion so wa e. One o he mos signi ican challenges o hi d-pa y ma ke ing au oma ion in PPC managemen is he lack o campaign con ol. As Balak ishnan and V ooman (2016) iden i ied, when ou sou cing PPC managemen o hi d-pa y p o ide s, he e is a isk o losing con ol o e he campaign's budge , messaging, and o e all s a egy. Addi ionally, hi d-pa y au oma ion ools may no op imize campaigns as e icien ly as in-house eams. These challenges can lead o subop imal esul s and educe he e u n on in es men . Ano he challenge o hi d-pa y ma ke ing au oma ion in PPC managemen is he isk o inc eased cos s. Acco ding o S ewa d (2019), hi d-pa y ools can be expensi e and may equi e addi ional s a o manage he campaigns e ec i ely. Fu he mo e, hi d-pa y ools may no always in eg a e well wi h exis ing ma ke ing sys ems, leading o addi ional expenses. The lack o anspa ency and isibili y is also a challenge o hi d-pa y ma ke ing au oma ion in PPC managemen . As Zei z (2021) iden i ied, hi d-pa y ools may no p o ide de ailed pe o mance epo s, making i challenging o unde s and he campaigns' e ec i eness ully. This can hinde he op imiza ion p ocess and lead o subop imal esul s. Finally, da a p i acy and secu i y a e signi ican conce ns wi h hi d-pa y ma ke ing au oma ion in PPC managemen . Wi z e al. (2016) no ed ha ou sou cing PPC managemen o hi d-pa y p o ide s can pose a isk o sensi i e cus ome da a. The e o e, o ganiza ions mus ensu e ha hi d-pa y p o ide s comply wi h da a p i acy egula ions and ha e app op ia e secu i y measu es. 2.10 TENSCORE AS A THIRD-PARTY MA One o he essen ial ac o s o conside when selec ing a hi d-pa y ma ke ing au oma ion ool o PPC managemen is he ea u e se , and businesses should look o ools ha p o ide a comp ehensi e se o ea u es, including bid managemen , campaign au oma ion, and epo ing (Kim e al., 2019). This ac is alida ed because obus MA ea u es can help businesses s eamline 38 hei PPC managemen asks, educe cos s, and imp o e campaign pe o mance (Kim e al, 2019). Ano he c i ical ac o o conside when selec ing a ma ke ing au oma ion ool o PPC managemen is i s in eg a ion capabili ies. As Lu e al. (2021) no ed, businesses need o choose ools ha can in eg a e wi h hei exis ing ad e ising pla o ms, such as Google Ads o Facebook Ads. This is because in eg a ion capabili ies can help businesses s eamline hei wo k low, educe e o s, and imp o e he accu acy o hei da a. The ease o use o a ma ke ing au oma ion ool is ano he c i ical ac o ha businesses need o conside . As no ed by Hsu e al. (2021), businesses should look o ools ha a e easy o use and equi e minimal aining. This is because ease o use can help businesses o imp o e hei e iciency, educe e o s, and inc ease use adop ion. The cos o a ma ke ing au oma ion ool is ano he essen ial ac o o conside . Kim e al. (2019) no ed ha businesses should look o ools ha p o ide a cos -e ec i e solu ion ha i s hei budge . Howe e , i is essen ial o balance cos agains he ool's ea u e se , as choosing a cheape ool wi h ewe ea u es may esul in educed campaign pe o mance and inc eased cos s in he long un. Finally, cus ome suppo is ano he c i ical ac o o conside when selec ing a ma ke ing au oma ion ool o PPC managemen . Acco ding o Lu e al. (2021), businesses should choose eliable cus ome suppo ools, including phone and email suppo , online documen a ion, and a knowledge base. This is because eliable cus ome suppo can help businesses esol e issues quickly, educe down ime, and imp o e hei o e all expe ience wi h he ool. Thi d-pa y ma ke ing au oma ion ools such as Tensco e claim o imp o e ad quali y sco es wi h ela i e p icing as low as $25/mon h o one Google Ads accoun . I also has a good cus ome suppo mechanism and easily in eg a es wi h Google Ads wi hou he need o much expe ise inpu ; based on hese poin s, his li e a u e e iew explo es he impac o Tensco e as a hi d-pa y ma ke ing au oma ion ool on Google Ad quali y sco e and he a ious aspec o Tensco e as a ma ke ing au oma ion ool, including i s ea u es, bene i s, and limi a ions. Tensco e is a hi d-pa y ma ke ing au oma ion ool ha allows businesses o au oma e hei ma ke ing ac i i ies. I is a comp ehensi e ma ke ing au oma ion ool ha p o ides businesses wi h 39 a ious ea u es o enhance hei ma ke ing ac i i ies. I s key ea u es include lead nu u ing, lead gene a ion, email ma ke ing, social media ma ke ing, and cus ome segmen a ion (Tensco e, n.d). The ool also p o ides businesses wi h de ailed analy ics and epo ing capabili ies o help hem measu e he e ec i eness o hei ma ke ing campaigns. Tensco e o e s se e al bene i s o PPC ad e ise s using he ool o ma ke ing. Fi s ly, he ool helps PPC manage s sa e ime and esou ces by au oma ing hei ma ke ing ope a ions, allowing hem o ocus on o he aspec s o hei campaigns. Secondly, Tensco e helps PPC manage s imp o e hei lead gene a ion and lead nu u ing esul ing in inc eased sales and e enue o B2C, Thi dly, Tensco e p o ides Google Ads PPC pla o ms wi h de ailed analy ics and epo ing capabili ies, enabling hem o measu e he e ec i eness o hei ma ke ing campaigns and make da a-d i en decisions. I is de ined as a SaaS ool ha specializes in helping ad e ise s op imize hei me ics and egula e click cos on hei PPC campaigns (Zayas, 2021). The ool pe o ms simple bu powe ul unc ions o moni o ing he indi idual quali y sco es o Google ads campaigns once i is linked o he Google ads in e ace (D elle , 2011). The ool has been desc ibed as a low-cos PPC solu ion ha helps imp o e ad e quali y sco e and help educe was e o ad-spend while imp o ing anking on he sea ch engine esul s, he eby boos ing ROI especially in SMES (Digi al school o ma ke ing, 2020). Acco ding o Bena d (2022), Tensco e se es as a b and moni o ing PPC ool ha o e s ma ke e s ecommenda ions ega ding op imizing hei campaigns. The ool has been ound e y use ul and can help PPC ad e ise s o keep ack o hei campaign pe o mance and make he bes decisions on how o dis ibu e hei campaign budge o mo e p o i able keywo ds he eby educing was e (Bena d, 2022). While Tensco e o e s se e al bene i s o businesses, i also has limi a ions. Fi s ly, he ool can be expensi e o small businesses making i di icul o hem o a o d. Secondly, Tensco e equi e businesses o ha e signi ican amoun o da a o be e ec i e, which can be a challenge o small businesses. Thi dly, Tensco e’s use in e ace can be o e whelming o some use s, making i di icul o hem o na iga e he ool wi hou any in-house expe ise (Ma ke ing au oma ion inside , 2019). 2.11 PROJECT MODEL AND HYPOTHESIS 40 This e iew explo e ROAS as he dependen a iable de ined ma hema ically acco ding o Se h (2023) as ROAS = [( AD REVENUE / AD COST) x 100] . This means ha achie ing mo e Con e sion alues wi h less Ad spend will esul o highe ROAS. ROAS depend on many ac o s in PPC ad e ising and p ospec i e online ma ke ing s a egy is e alua ed based on i s inc emen al e u n on ad spend (Chen & Au, 2022). E alua ing ROAS in digi al ad e ising has been a undamen al p oblem in ma ke ing (Chen e al., 2018). In he Google ads pla o m, he me ic o ROAS is gi en as con e sion alue di ided by he cos o ad e ising. Many pas s udies ha e ied o ind he connec ion be ween ROAS and he exis ing PPC s a egy. Chen and Au (2022) de ined iROAS as inc emen al e u n on ad spend in ela ion o esponse o he PPC s a egy. Many s udies ha e ca ied ou esea ch o de e mine he impac o inc ease on Ad budge on ROAS as a dependen a iable (Chan & Pe y, 2017). O he s udies ocus on e alua ing he impac o he Media Mix Model on ROAS. Medial Mix Model in ol es he use o obse a ional da a such as p ice, sales, Ad spend and economic ac o s o Fo ecas he impac on ROAS (Chen e al., 2018) bu no much has been dedica ed on expe imen ing o p ac ical alida ion o using MA as a casual e ec on ROAS. This e iew iden i ies his esea ch gap ha will be exploi ed and expanded in o a model by in es iga ing he impac applica ion o ROAS sma bidding s a egy and Google Ads au oma ed pe o mance Max campaign as a PPC s a egy on ROAS and use i o de e mine he le el o inc emen al ROAS hence o h e e ed o as iROAS. This li e a u e will e iew and modi y he implici assump ion model de eloped by Chen e al. (2018) whe e Y s and o Sales alue wi h a casual pa h a ec ed by O and P, whe e O and P e e o O ganic esul s and Paid esul s in sea ch engines espec i ely, A s and o auc ion and Q s and o use ’s sea ch que y in he sea ch engine (see ig 2.2 below). 41 Figu e 2.2 A causal Model diag am o sea ch ad a a que y le el. Adap ed om Chen e al. (2018). Modi ying his model, IF Y= Y Sales alue om Google ads PPC ad e ising, O= S ep esen ing S anda d shopping campaign s a egy, A = A ep esen ing Ad spend, Q = G ep esen ing Google ads PPC channels and P = P ep esen ing Pe o mance Max campaign (au oma ed campaign), hen his e iew eplica es he abo e implici model as shown below in ig 2.6. This s udy p oposed h ee hypo heses as ep esen ed below: H1: Pe o mance max campaign has mo e posi i e impac on con e sions han s anda d shopping/sea ch campaign Figu e 2.3 Hypo hesis 1 model diag am showing ela ionship be ween Pmax, Con e sions & SSC H2 Pe o mance max campaign has mo e signi ican impac on ROAS han s anda d shopping/sea ch campaign Figu e 2.4 Hypo hesis 2 model diag am showing ela ionship be ween Pmax, ROAS& SSC H3: Google Ads’ sma bidding -maximize con e sion alue wi h a ge ROAS s a egy has mo e signi ican posi i e impac on ROAS han manual bidding s a egy. Figu e 2.5 Hypo hesis 3 model diag am showing ela ionship be ween sma bidding, ROAS & manual bidding. The h ee hypo hesis p oposes a ela ionship be ween pe o mance max campaign, s anda d shopping/sea ch campaign, sma bidding s a egy ( a ge ROAS), manual bidding s a egy and hei impac on con e sions and ROAS in Google ads PPC channels, using hese 3 hypo heses wi h he model adap ed om Chen e al. (2018) which es ablished a ela ionship on ROAS and cos o ad spend, a holis ic model is hen p oposed o demons a e he h ee hypo hesis 48 Figu e 3.1 50% SPLIT: Con ol & T ea men G oups Figu e 3.2 Pmax and SSC as T ea men & Con ol G oups Respec i ely Figu e 3.3 Tensco e in e ace showing campaign s uc u e sco e on Google ads 3.7 PMAX VS S anda d Sea ch Campaign The di e ence be ween he s anda d shopping and sea ch campaigns is ha shopping campaigns a e o p oduc ma ke ing while sea ch ads a e o bo h p oduc ma ke ing and se ice- 49 based. To inc ease he eliabili y o he da a used in he p ojec analysis, he au ho es ed he Pmax campaign o di e en B2C online businesses. Hence, a spli expe imen was c ea ed using Pmax as a ea men g oup and sea ch ads as a con ol g oup, bu his ime, he online business being a ge ed is a se ice based whe e leads gene a ion is he goal o he campaign ins ead o pu chases/sales. The dependen a iable is CONVERSION, while expe imen is in es iga ing he causal e ec s o Pmax and Sea ch ads as independen a iables on con e sions The expe imen was spli in o 50%, indica ing equal a ic and budge alloca ion as was done when ROAS is he dependen a iable. Fig 3.3 and Fig 3.4 illus a e as explained. 3.8 INTEGRATING TENSCORE WITH GOOGLE ADS PLATFORM The Tensco e so wa e was in eg a ed in o he Google Ads in e ace by linking he speci ic Google ads IDs by he au ho as he adminis a o o he PPC in e ace. A e connec ing he Tensco e, he exis ing campaigns o be used as con ol g oups we e selec ed and es uc u ed using he Tensco e es uc u ing in e ace. The Tensco e so wa e immedia ely sugges ed au oma ed changes o be made o he exis ing campaign o imp o ed pe o mance s a ing om he campaign Ad g ouping s uc u es. A new se ies o ad g oups was c ea ed om he exis ing campaign and au oma ically published in he Google Ads in e ace. All he newly c ea ed ad g oups a e di e en ia ed om he exis ing Ad g oups wi h a TS P e- ix o no e hose ad g oups being c ea ed using he Tensco e ool. A e he new ad g oups we e c ea ed, a spli campaign was se up as an A/B es expe imen as seen in Fig 3.3. The con ol g oup campaign con ains Ad g oups manually c ea ed wi hou he Tensco e so wa e, while he ea men campaign con ains only l he ad g oups au oma ically c ea ed using he Tensco e hi d-pa y ools wi h a TS p e ix. All o he pa ame e s in he campaign a e made o be he same as he spli g oups equally ecei ed 50% o a ic and budge sha e du ing he pe iods o he expe imen a ions. The p ocess is epea ed in bo h se ice-based businesses and ecomme ce-based campaigns in B2C online businesses. 50 3.9 EXPERIMENT DURATION AND KPI SAMPLE SIZES The o al du a ion o he expe imen span om Ma ch 20 h, 2023, o June 15 h, 2023, which in ol ed app oxima ely 90 days. Ideally, he expe imen was se o e mina e on June 20 h bu he da a o he esul s was ex ac ed un il June 15 h. The numbe o clicks ep esen s he subjec sizes in he expe imen as i shows a numbe o people who saw he expe imen and clicked on i . O he impo an da a which consis o he sample sizes a e he imp essions and CTR, which indica ed he a e a which people clicked on he ads du ing he expe imen pe iods. These sample sizes a e below o bo h he E-comme ce pla o m and leads campaign. Table 3.1 Sample size o he expe imen 3.10 MATHEMATICAL MODEL FOR EXPECTED RESULTS VALIDATION Bea ing in mind he hypo hesis o his s udy is u he lis ed below as ollows: a) H1 - Pe o mance max campaign has a mo e posi i e impac on con e sions han he s anda d shopping/sea ch campaign; b) H2 - Pe o mance max campaign has a mo e signi ican impac on ROAS han he s anda d shopping/sea ch campaign; c) H3 - Google Ads’ sma bidding -maximize con e sion alue wi h a ge ROAS s a egy has a mo e signi ican posi i e impac on ROAS han manual bidding s a egy. 51 3.10.1 Fo Hypo hesis 1 – Impac On Con e sion The change in he numbe o con e sions is deno ed by ∆CN whe e: • CN is he numbe o con e sions. • PMAXNC is he numbe o con e sions om Pe o mance Max campaign. • SCCNC is he numbe o con e sions eco ded by s anda d sea ch/shopping campaigns. 𝑓𝑜𝑟 𝑡ℎ𝑖𝑠 𝑠𝑡𝑢𝑑𝑦, ∆CN which is change in he numbe o con e sions which is he di e ence in he numbe o con e sions be ween he ea men g oup and he con ol g oup in he expe imen s which a e Pmax and SSC espec i ely. Hence, Hypo hesis 1 is TRUE IF PMAXNC > SCCNC which implies ha ∆CN >0, whe e ∆CN = PMAXNC – SCCNC 3.10.2 Fo Hypo hesis 2 -Impac on ROAS Fo his s udy, he change in ROAS is deno ed by ∆R meaning he di e ence be ween ROASPMAX And ROASSSC. Whe e ROASPMAX is he e u n on ROAS om Pmax campaign and ROASSSC. is he e u n on ad spend om s anda d shopping/sea ch campaigns. The e o e, ∆R = ROASPMAX - ROASSSC. IF ROASPMAX > ROASSSC. hen ∆R will be posi i e which implies ha hypo hesis- H2 is sus ained o be TRUE else Hypo hesis 1 is FALSE. The e o e, H2 is TRUE only when ∆R = ROASPMAX - ROASSSC leads o inc emen al ROAS (iROAS) as de ined by (Chen e al, 2018) 3.10.3 Fo hypo hesis 3 -Impac o sma bidding s a egy ( a ge ROAS) on ROAS To de e mine he impac o sma bidding s a egy on ROAS, his s udy conside ed he change in ROAS o s anda d shopping campaign be o e and a e he implemen a ion sma bidding s a egy. The e o e, ∆R is he change ROAS be ween manual and sma bidding calcula ed o e equal pe iod o ime which is 45 days each o he o al o 90 days o he spli expe imen . ROASMD ep esen he ROAS om manual bidding while ROASSM he e o e iROAS which is inc emen in ROAS esul ing om i ROASSTB -ROASMD 3.11 INTEGRATING RELIABILITY IN THE EXPERIMENTATION PROCESS Reliabili y as a ac o o quan i a i e esea ch measu emen de ined as an indica o o he 52 s abili y o he measu ed alues ob ained in epea ed measu emen s unde he same ci cums ances using he same measu ing ins umen (Maslaka & Su ucu, 2020). To ensu e he eliabili y in he me ics being analyzed in his p ojec , he me hodology been adap ed ensu ed ha he ollowing measu es we e aken, iz. The expe imen was ca ied ou in 3 h ee di e en B2C businesses which included women appa els online s o e in USA, ac ical gea online s o e based in Canada, Cus om ashion online s o e based in Aus alia, Colonics he apy online se ices based in Aus alia and Luck smi h se ices based in USA. The spli in pa ame e s used in he expe imen whe e sha ed equally in he a io o 50% so ha bo h he con ol g oups and ea men g oups ecei ed he same a ics and he same budge alloca ion wi hin he ime o he expe imen s. The a ge loca ions o bo h he con ol g oups and he ea men g oups a e made o be he same o a oid any undue ad an age acc uing o ei he he con ol o ea men g oup. Bo h he con ol and ea men g oups a e made o ha e he same ad scheduling, his allows all he ads o un a he same pe iods o he day. The landing pages o bo h he con ol and ea men g oups a e same, his allows o uni o m exposu e o he same useabili y a io in ela ion o landing page expe ience. Fo he ecomme ce p oduc s campaigns, bo h he ea men and con ol g oups a e connec ed o he same me chan cen e accoun meaning ha same p oduc s a e being ad e ised ac oss he wo g oups. 53 4. RESULTS The heo e ical concep o his p ojec unde sco es he e ec i eness o MA on he ollowing concep s iz is con e sions, ROAS, and o he PPC pe o mance me ics in Google ads ad e ising channels. The hi d-pa y au oma ion which is ano he leading concep unde sco es he easy o ca ying ou epe i i e asks agains he human manipula ions. The esul s o his s udy a e explo ed and displayed wi h a en ion o connec ing he esul s o hese he si ical concep s. 4.1 RESEARCH RESULTS OF ALL KPI FOR TACTICAL GEAR E-COMMERCE CAMPAIGNS The esul s below show all majo KPI measu ed agains he wo main campaigns unde s udy in he spli expe imen o he B2C ac ical gea E-comme ce pla o m. The wo ypes o campaigns a e Pmax and SSC Table 4.1 All Ta ge KPI om Pmax and s anda d shopping campaigns a e 90 o spli expe imen 4.2 CONVERSION METRICS RESULTS FROM THE TACTICAL GEAR E-COMMERCE CAMPAIGNS EXPERIMENTS The s udy compa es he impac o ull in en o y au oma ed pe o mance campaign agains s anda d campaign in a ac ical gea supply indus y and he esul s o pe iods o 90 days. F om Table 1, i shows ha Pmax eco ded a o al o 55.3 con e sions while he S anda d campaigns eco ded only 14 con e sions he pi-cha and ba cha in ig 4.1below illus a e % p opo ion which shows ha ou o he o al numbe s o con e sions eco ded du ing he pe iods, Pmax eco ded 79.8% o he con e sions while he SSC eco ded 28.2% o he con e sions. 54 Figu e 4.1 Cha showing Numbe o con e sion esul s om Pmax & SSC The con e sion ends as shown in ig 4.2 below shows ha Pmax s a ed eco ding con e sion om Ma ch 21 which is he second day o he expe imen while he S anda d shopping campaign s a ed eco ding con e sions on Ma ch 29 which is 9 days a e he expe imen s a ed se ing ads. Figu e 4.2 Con e sion T ends om Pmax and SSC Expe imen om Ma 20-June 15 Figu e 4.3 Con e sion Ra es om Pmax and SSC Campaign Expe imen s 55 4.3 ROAS METRICS RESULTS FROM THE TACTICAL GEAR E-COMMERCE CAMPAIGNS F om he Table 1.0 abo e, a su p ise me ics shows ha SSC eco ded a highe ROAS agains he Pmax campaigns despi e eco ding he highes numbe o con e sions wi h a alue o 4.57 and 4.01 espec i ely. The igu e in Fig4.4 shows he ROAS p opo ions o each o he campaigns ep esen ed in bo h pi-cha and ba cha simul aneously. F om Fig 4.4, he ROAS. Ano he in e es ing me ics is he ROAS ends in Fig 4.5 which shows ha S anda d sea ch campaigns eco ded e y low ROAS un il May 9 h.when he bid s a egy was changed om manual bidding o au oma ed bidding -Ta ge ROAS Figu e 4.4 ROAS Sha e be ween PMAX & SSC Figu e 4.5 ROAS T ends Pmax and ROAS T ends S anda d Shopping 56 Figu e 4.6 Con e sion Values and Ad Spend om Pmax & SSC -spl expe imen 4.4 CLICKS AND CLICK RATES METRICS RESULTS FROM THE SPLIT CAMPAIGNS The click me ics is an impo an me ics ha measu e he willingness o people o in e ac wi h ads by clicking on i a e imp essions a e es ablished (Gi aldo-Rome o e al., 2021) Acco ding he Pi- cha in Fig4.7, Pmax ecei ed 71.4% o he o al clicks du ing he pe iods o he campaign expe imen while SSC ecei ed only 28.4% o he clicks. The 71.4% ep esen a o al o 1,842 clicks while he 28.5% ecei ed om SSC ep esen a o al o 739 clicks as shown in Fig 4.7 below. Figu e 4.7 cha showing numbe o con e sions be ween Pmax & SSC Campaign Ano he impo an esul s he e is he click a e which is he a e a which an ad is being clicked compa ed o he numbe o imp essions being made (Osmundson, 2022) FIG4.8 indica ed ha Au oma ed 57 ML d i en campaign PMAX has a highe CTR compa ed o S anda d shopping campaigns despi e showcasing he same ypes o p oduc s iews du ing he ime o he campaign. Figu e 4.8 Clicks & Click Ra es Pmax Vs SSC Campaign Resul s Fig 4.9 indica ed an in e es ing phenomenon whe e he s anda d shopping/sea ch campaign eco ded highe imp ession han he au oma ed campaigns. Ou o a o al o 116,092 imp essions eco ding du ing he pe iods o he expe imen , SSC campaign go 65,023 while Pmax ecei ed 51,069 imp essions as illus a ed in he Fig 4.90 below. A d ill down o he imp essions shows ha SSC ecei ed mo e imp essions on mobile de ice wi h 52,611 imp essions while Pmax ecei ed 35,587 imp essions on mobile de ices. The Pmax had mo e imp ession on compu e de ice han SSC wi h 14.658 while SSC go 11,362 imp essions as seen in ig 4.9 below. 64 Figu e 5.3 SSC ROAS T ends Showing a Rise in ROAS a e swi ching o Sma Bidding om May 9 h, 2023. F om he esul s om he expe imen as shown in Fig 5.1 and 5.2 abo e iROAS = ROASSTB - ROASMD ROASSTB = 8.23 and ROASMD = 0.86. Hence iROAS which is he inc emen in ROAS is 8-23-0.86 = 7.37 The e o e, ROAS om sma bidding s a egy eco ded iROAS o 7.37 when compa ed o manual bidding which p o ed H3 o be TRUE. 5.4 RESEARCH MODEL AND HYPOTHESIS ADOPTION Ou o h ee hypo heses o he s udy, wo has been p o ed o be accep ed while only one is p o ed o be alse as discussed abo e, his is summa ized below along wi h he model. 65 Table 5.2 Hypo hesis adop ion 5.5 PRATICAL CONTRIBUTIONS 5.5.1 INSIGHT TO PPC MANAGERS AND PPC SPECIALIST Wha his implies o PPC manage s and specialis s is ha combining SSC and Pmax is a e y g ea way o ha ness a highe ROAS. Ano he impo an s a egic ou come o he s udy is he need o s a wi h Pmax campaign and ha e i gene a e many numbe s o con e sions be o e c ea ing shopping campaigns. In his way, i will educe ROAS down ime since i will ake he s anda d shopping campaign longe pe iod be o e eco ding a signi ican ROAS. Google ads uses accoun his o y o de e mine a e age a ge cos pe acquisi ion and a e age a ge ROAS. Since Google ads equi e a minimum o 15 con e sions and a ce ain ROAS his o y be o e one can se up sma 66 bidding s a egy on s anda d shopping campaigns, i hen become impe a i e o use Pmax campaign o gain accele a ed ac ion on accoun his o y o ROAS o enable be e and s a egic campaign op imiza ion du ing he ime SSC is unning on manual bidding. 5.5.2 MANAGERIAL PLANNING AND CONTROL This s udy p oposes a manage ial decision and budge a y alloca ion in e ms o campaign planning in b2c ecomme ce ad e ising. I p oposes alloca ing mo e budge o pe o mance max campaign a he i s wo mon hs o he campaign be o e di e si ying o s anda d shopping campaign. This will enable maximum ROAS since he SSC will ha e exi manual bidding be o e ge ing he equi ed ROAS. 5.5.3 OPPORTUNITY FOR PRODUCT FEATURE UPGRADE The esul o his s udy shows ha s anda d shopping campaign has a highe ROAS when ea ed wi h sma bidding s a egy. The p esen echnical de elopmen ea u es o he s anda d shopping campaign only allow i o un on manual bidding s a egy un il a ce ain numbe o con e sions has been egis e ed be o e i can swi ch om manual bidding o sma bidding. Upg ading he SSC o un on sma bidding s a egy du ing he campaign se up s age wi hou he need o ge a ce ain numbe o con e sions be o e swi ching om manual o sma bidding s a egy will p esen mo e oppo uni y o ad e ise s o deli e mo e ROI by boos ing he ROAS igh om he beginning o he campaign. 67 6 CONCLUSION This s udy del ed deeply in o he impac o Ma ke ing au oma ion on Google Ads PPC managemen and he a endan e ec o Tensco e-Thi d-pa y au oma ion ool. The s udy was able o es ablish ha ha use o ma ke ing au oma ion is a leading end in PPC ad e ising ha ha nesses he la ge chunk o da a in ol ed in Google ads PPC and s eamlined hem in mo e op imized o m o be e pe o mance. The s udy was able o es ablish ha he Pe o mance Max campaign can educe ime used in c ea ing and se ing up campaigns as i blended he ads asse s ac oss all he Google ads in en o ies. The ask and cumbe someness o c ea ing di e en Google ads asse s ha will appea on all in en o ies o Google such as YouTube, display ne wo k, The s udy ound ou ha Google ads PPC ad e ising is becoming oo complex o human managemen and he e o e es ablished ha one o he bes au oma ed ea u es agains manual maipula ion is he use o Google ads au oma ed bidding s a egy o gain highe e u n on ad spend. ROAS The s udy concludes ha sma bidding s a egies wi h a ge ROAS has g ea posi i e impac on e u n on ad spen . I was e y e iden in he s udy ha a ansi ion om manual bidding in s anda d shopping campaign will exi a s anda d shopping campaign om low ROAS o a highe and posi i e e u n on ad spend. This e ela ion is in con ac o many o he s udies ha had p edic ed ha Pmax campaign has mo e po en iali y o gene a ing a highe ROAS han s anda d campaign. The campaign bulk ad g ouping s uc u e and how o o ganize hem in o e icien hemes ha will blend in o Tensco e as a hi d-pa y au oma ion in his s udy p o ed o be e y e ec i e by c ea ing an in e ace be ween Google ads campaigns and PPC managemen especially in es uc u ing a campaign. 6.1 RESEARCH LIMITATION This esea ch is limi ed by some ac o s which includes he olume and scope o he PPC me ics in ol ed. These limi a ions ange om he pe iod om which he da a we e ex apola ed o he amoun o campaign budge and he ype o se ices been a ge ed which in his case a e la gely B2C indus ies. Da a in PPC managemen a e g ea ly impac ed o e long pe iod o ime and he size 68 o campaign budge unde in es iga ion. Many big sized companies wi h e y la ge amoun o campaign budge may ha e a ied obse a ions in analysis wi h he au oma ion ool and campaign ea u es been used. Fu he s udies should endea o o explo e bigge campaign da a by ex ac ing me ics om la ge campaign budge om mul i-na ional companies and agencies. Also ha ing access o da a anging o e 12mon hs o ac i e campaign pe iod will b ing a mo e ull-de ailed s udy ha will be mo e alid in making da a d i en decisions. The hi d-pa y ools used in he p ojec a e s anda d subsc ip ion-based model which ha e limi ed unc ionali y compa ed o ull-op ion subsc ip ions ha ha e all he hi d-pa y au oma ion ea u es. The esea ch only explo es he use o Google ads PPC as he digi al ma ke ing channels, he e may be alid hypo hesis ha he esul s and conclusion eached in his esea ch may no be applied in o he digi al ma ke ing channels like paid socials as his was no conside ed du ing he esea ch p ocess and applica ion. 6.2 SUGESTIONS FOR FURTHER STUDIES A ibu ion model plays a e y impo an ole in deciding he con e sion me ics as i de ines he di e en le els o ouch poin s be o e gi ing con e sion c edi s o a gi en campaigns The di e en ypes o a ibu ions p esen in Google Ads PPC as a he ime o his s udy a e da a-d i en, las clicks and posi ioned based a ibu ions models (Google Ads Gene al Resou ces, 2023). This s udy uses las click model, which implies ha only KPI gene a ed h ough las clicks a e coun ed. I is es ima ed ha his may a ec he pe o mance o a gi en campaign. Resea ch on a ibu ion model has shown ha many e u ning use s who did no comple e a pu chase on hei i s isi a e being coun ed as a di ec isi o and he alues o hei ansac ions a e no e en ually been c edi ed o he campaigns ha ini ially b ough hem o he websi e (Ru z e al., 2009). The e o e, his esea ch ecommends subjec ing u u e expe imen s in o a ious a ibu ion models as possible o ensu e ha each campaigns a e gi en he c edi s i dese ed as di e en ypes o campaigns may pe o m bes in a pa icula a ibu ion model han o he s. Gene a i e AI should be in eg a ed in he nex s udy o e alua e how his will a ec he a ge KPIs. Google indica ed du ing he 2023 Google Ma ke ing Li e (GML) held in he mon h o May, 69 2023 ha i would be launching au oma ed asse s c ea ion ha will use a gene a i e AI o w i e headlines and ex ads using cus ome s sea ch que ies. The implica ion o his is ha he ads ha will be seen by use s will be w i en by AI in eal ime and in consonan s wi h hei sea ches on he sea ch engines. Du ing he ime o his esea ch, his ea u e has no been o icially launched. Fu u e s udies should in eg a e his gene a i e au oma ion echniques wi h a iew o inding hei impac s in PPC campaign managemen as agains he s a ic campaign asse s c ea ion. In eg a ing google p oduc s udio Tha allows au oma ed c ea ion o p oduc s images and linking hem in o he me chan cen e should also be esea ched upon as hese a e he p ojec ed ma ke ing au oma ion ea u es ha will be seen a he end o 2023 wi h he aim o deepening au oma ion in Google ads c ea ion and managemen . Fu he s udies should endea o o in eg a e i s pa y da a in hei Google ads PPC channels while ca ying ou u he s udies in ela ed opics. This sugges ion is made based on he expec ed impac ha i s pa y da a a e assumed o ha e on e ining AI d i en campaigns op imiza ion including s anda d campaigns ool. In eg a ing i s pa y da a may ha e he likelihood o impac ing on he esul s and conclusion. 70 BIBLIOGRAPHY Ad hena. (2023, May 18). You comp ehensi e guide o Pe o mance Max campaigns. Sea ch Engine Land. Ex ac ed o m: h ps://sea chengineland.com/you -comp ehensi e-guide- o-pe o mance- max-campaigns- 417277 Allazo , K. (2020). Role o Au oma ion in PPC Managemen : Google Ads Sma Bidding S a egies and Oppo uni ies o hi d-pa y solu ions. Budapes : Co inus Uni e si y o Budapes . Amanda AI. (2022, Augus 24). The nex gene a ion in au oma ion: Pe o mance Max. Ex ac ed om: h ps://amandaai.com/ esou ces/blog/p oduc / he-nex -gene a ion-in-au oma ion- pe o mance- max/ Anicca Webina s. (2023). PPC au oma ion & AI: Ha nessing i s Powe s. A ailable a : h ps://anicca.co.uk/blog/ppc-au oma ion-ai-ha nessing-i s-powe s/ Ayanso, A., & Ka imi, A. (2015). The mode a ing e ec s o keywo d compe i ion on he de e minan s o ad posi ion in sponso ed sea ch ad e ising. Decision Suppo Sys ems, 70, 42–59. Bena d, P. K. (2022, Augus 8). The 10 Bes PPC Tools o Maximum ROI in 2022. DATASLAYER Da a Science o Ma ke e s. Ex ac ed om: h ps://da aslaye .ai/es/blog/ he-10-bes -ppc- ools- o - maximum- oi-in- 2022. Biegel, B. (2009). The cu en iew and ou look o he u u e o ma ke ing au oma ion. In Jou nal o Di ec , Da a and Digi al Ma ke ing P ac ice, 10(3), 201–213. Bishop, A. (2021, June 7). E e y hing We Know Abou Google’s Pe o mance Max Campaigns. Sea ch Engine Jou nal. Ex ac ed om: h ps://www.sea chenginejou nal.com/google-pe o mance- max/409736/#close Chan, D., & Pe y, M. (2017). Challenges And Oppo uni ies In Media Mix Modeling. A ailable a h ps:// esea ch.google.com/pubs/a chi e/45998.pd . Chen, A., & Au, T. C. (2022). Robus causal in e ence o inc emen al e u n on ad spend Wi h andomized pai ed geo expe imen s. Annals o Applied S a is ics, 16(1), 1–20. Chen, A., Chan, D., Pe y, M., Jin, Y., Sun, Y., Wang, Y., & Koehle , J. (2018). Bias Co ec ion Fo Paid Sea ch In Media Mix Modeling. Ex ac ed om: h p://a xi .o g/abs/1807.03292 71 Chen, Y., & He, C. (2006). Paid Placemen : Ad e ising and Sea ch on he In e ne , NET Ins i u e Wo king Pape #06-02. Ex ac ed om: h p://www.NETins .o g, CMA Compe i ion & Ma ke s Au ho i y. (2020). Online pla o ms and digi al ad e ising Ma ke s udy. Re ie ed om h ps://www.go .uk/cma-cases/online-pla o ms-and-digi al-ad e ising-ma ke - s udy Cohen, L., Manion, L., & Mo ison, K. (1980). Resea ch Me hods in Educa ion. London: Rou ledge. Co sa o, D., Maggioni, I., & Oli ie i, M. (2021). Sales and ma ke ing au oma ion in he pos -Co id- 19 scena io: alue d i e s in B2B ela ionships. I alian Jou nal o Ma ke ing, (4), 371–392. Deligiannis, A., A gy iou, C., & Kou esis, D. (2020). Building a Cloud-based Reg ession Model o P edic Click- h ough Ra e in Business Messaging Campaigns. In e na ional Jou nal o Modeling and Op imiza ion, 26–31. Desai, V. (2019). Digi al Ma ke ing: A Re iew. In e na ional Jou nal o T end in Scien i ic Resea ch and De elopmen (IJTSRD), 5(5), 196-200. Digi al Ma ke ing Ins i u e. (2022, Decembe 22). The Ul ima e Guide o Ma ke ing Au oma ion. Digi al Ma ke ing Ins i u e. Ex ac ed om: h ps://digi alma ke ingins i u e.com/blog/ he-ul ima e- guide- o- ma ke ing- au oma ion D elle , J. (2011, Feb ua y 18). Quali y Sco e T acking Tool: An In e iew Wi h TenSco es Founde Ch is Thunde . Ex ac ed om: h ps://sea chengineland.com/quali y-sco e- acking- ool-an-in e iew- wi h- ensco es- ounde -ch is- hunde -64744 EDWARDS, S. (2023). A Guide o Google Ads Bidding S a egies in 2023. Ex ac ed om: h ps://g ow hmindedma ke ing.com/blog/google-ads-bidding-s a egies/ Feedonomics. (2023, Feb ua y 9). Google’s Pe o mance Max Campaigns A e He e. A e You Ready? Flo es, V. (n.d.). Wha a e google’s au oma ed pe o mance max campaigns? Ex ac ed om: h ps://www.adlucen .com/ esou ces/blog/wha -a e-googles-au oma ed-pe o mance-max- campaigns/ Gi aldo-Rome o, Y. I., Pé ez-De-los-cobos-agüe o, C., Muñoz-Lei a, F., Higue as-Cas illo, E., & Liébana- Cabanillas, F. (2021). In luence o egula o y i heo y on pe suasion om google ads: An eye acking s udy. Jou nal o Theo e ical and Applied Elec onic Comme ce Resea ch, 16(5), 1165– 72 1185. Google. (n.d.). Abou Pe o mance Max campaigns. Google Ads Help. Re ie ed om h ps://suppo .google.com/google- ads/answe /10724817?hl=en#:~: ex =Pe o mance%20Max%20is%20a%20new,in en o y%20 o m%20a%20single%20campaign. Google Ads Gene al Resou ces. (2023). Da a-d i en a ibu ion me hodology. Re ie ed om: h ps://suppo .google.com/google-ads/answe /6394265?hl=en Google Ads Help. (2023). Abou au oma ed bidding. Re ie ed om: h ps://suppo .google.com/google- ads/answe /2979071?hl=en Google Ma ke ing Li e. (2023). Measu e You Ma ke ing Wi h Con idence. Google Ads YouTube Channels. Gue ini, M., S appa a a, C., & S ock, O. (2010). E alua ion Me ics o Pe suasi e NLP wi h Google AdWo ds. Hall, J. (2022, Sep embe 14). PPC Au oma ion: Tips & Tools Fo Be e Paid Ad Resul s. Ex ac ed om: h ps://lunio.ai/blog/paid-sea ch/ppc-au oma ion- ips- ools/ Hamalainem, R. (2020). The an eceden o ma ke ing au oma ion success in sys em implemen a ion p ocess – case innish manu ac u ing company. Jy äskylä Uni e si y School o Business and Economics. Hammoud, A. G., Taw ik, F. H., & Fa hallah, A. M. (2019). Implemen a ion o Ma ke ing Au oma ion Sys em in Business Ma ke ing. Heimbach, I., Kos y a, D. S., & Hinz, O. (2015). Ma ke ing Au oma ion. In Business and In o ma ion Sys ems Enginee ing, 57(2), 129–133. Janku ė-Ca maciu. I. (2020, Ap il 4). Why Imp ession Ma ke ing Is Impo an ? Re ie ed om: h ps://wha ag aph.com/blog/a icles/imp essions-ma ke ing Jä inen, J., & Taiminen, H. (2016). Ha nessing ma ke ing au oma ion o B2B con en ma ke ing. Indus ial Ma ke ing Managemen , 54, 164–175. Jun ila, A. (2022). Google Sea ch Engine MRK Guide. Bachelo Thesis, Sa onia Uni e si y o Applied Sciences, Kuopio, Finland. Kabi aj, S., & Joghee, S. (2023). Imp o ing Ma ke ing Pe o mance: How Business Analy ics con ibu e o Digi al Ma ke ing. In e na ional Jou nal on Technology, Inno a ion, and Managemen , 3(1). Kim, L. (2022, Decembe 4). The Wa on ‘F ee’ Clicks: Think Nobody Clicks on Google Ads? Think Again! 73 Wo dS eam. Ex ac ed om: h ps://www.wo ds eam.com/blog/ws/2012/07/17/google- ad e ising Ko ane, I., Zno ina, D., & Hushko, S. (2019). Assessmen o ends in he applica ion o digi al Ma ke ing. Scien i ic Jou nal o Polonia Uni e si y, 33(2), 28–35. Laubens ein, C. (2022, Decembe 4). Wha ’s a Good Click-Th ough Ra e (CTR) o Google Ads? Wo dS eam. Ex ac ed om: h ps://www.wo ds eam.com/blog/ws/2010/04/26/good-click- h ough- a e. Li, J., Wang, Z. L., Zhao, H., G a ina, R., Fo ino, G., Jiang, Y., & Tang, K. (2017). Ne wo ked human mo ion cap u e sys em based on qua e nion na iga ion. BodyNe s In e na ional Con e ence on Body A ea Ne wo ks. Re ie ed om h ps://doi.o g/10.1145/0000000.0000000 Lincoln, J. (2022, Ma ch 30). How o calcula e oas: unde s anding e u n on ad spend. Igni e Visibili y. Re ie ed om h ps://igni e isibili y.com/how- o-calcula e- oas/ Li le, J. (2001). Ma ke ing Au oma ion On The In e ne . In 5 h In i a ional Choice Symposium (pp. 1-5). Ma ke And Ma ke s. (n.d.). Ma ke ing au oma ion ma ke by componen s (so wa e and se ices)...Global o ecas o 2027. Re ie ed om h ps://www.ma ke sandma ke s.com/Ma ke -Repo s/ma ke ing-au oma ion-so wa e-ma ke - 155627928.h ml McCo mick, K. (2023, Ma ch 2). Wha Is Digi al Ad e ising? Types, Bene i s & Examples (+P o Tips!). Re ie ed om h ps://www.wo ds eam.com/blog/ws/2023/02/24/digi al-ad e ising Me a. (2023). Thi d Pa y Apps Re ie ed om . h ps://de elope s. acebook.com/docs/wo kplace/ hi d-pa y-apps/in oduc ion/ Miklosik, A., Kuch a, M., E ans, N., & Zak, S. (2019). Towa ds he Adop ion o Machine Lea ning- Based Analy ical Tools in Digi al Ma ke ing. IEEE Access, 7, 85705–85718. Mu phy, D. (2018). Sil e bulle o mills one? A e iew o success ac o s o implemen a ion o ma ke ing au oma ion. In Cogen Business and Managemen , 5(1), 1–10). Nai , K., & Gup a, R. (2020). Applica ion o AI echnology in mode n digi al ma ke ing en i onmen . Wo ld Jou nal o En ep eneu ship, Managemen and Sus ainable De elopmen , 17(3), 318–328.