Mostrar el registro sencillo del ítem

dc.rights.licensehttps://creativecommons.org/licenses/by-sa/2.5/ar/es_AR
dc.contributor.advisorGallino, Santiagoes_Ar
dc.contributor.authorFernández, Leoneles_AR
dc.date.accessioned2023-01-09T14:27:38Z
dc.date.available2023-01-09T14:27:38Z
dc.date.issued2020
dc.identifier.urihttps://repositorio.utdt.edu/handle/20.500.13098/11573
dc.description.abstractThe Digital landscape has evolved vastly since the early 2000s in terms of analytical tools and tracking software. With the Rise of 4G to 5G, smartphones have become the norm when surfing through the web. New problems arise in terms of measuring business performance like Cross-Channel and Multi-Channel Attribution. Companies are selling more products and services on their Websites and marketplaces than ever before. Brands must become digital natives and translate all of their offline business into the internet. When Brands invest in multiple marketing channels and those channels mix up in the Customer Journey, new measurement problems arise. Based on the current standard methodology on web analytics, companies track their conversions (signups, subscriptions, orders) and assign each channel’s attribution using simple heuristics. In other words, simple decision models. It has been vastly studied that single-touch attribution does not perform well under complex business scenarios like those observed nowadays. Attribution modeling has been a hot topic in the last decade due to the rise of Machine Learning and data mining. Nowadays, there are two current trends. The problem can be analyzed from a Machine Learning standpoint, understanding that it looks like a Classification problem with a Binary Outcome (0/1). On the other hand, Shapley Values and Game theory also adapt efficiently to the question, where every player gets credit for contributing to conversions. Given that there are different state-of-the-art models which perform better than others and that multiple papers are trying to improve robustness, predictive accuracy, interpretability, this thesis will focus primarily on applications and interpretability of the model. Most of today’s Marketing Managers and teams find it extremely hard to use and apply these types of models due to the complexity of the topic and black-box models, which have little to no interpretability. The idea is to encourage more companies into the MTA landscape to test their models and optimize them specifically for their industry in this work. Additionally, to my knowledge, there is no research on Markov Chains applied to Subscription Business Models that are substantially different from E-Commerce Customer Journeys.es_AR
dc.description.sponsorshipPor motivos relacionados con los derechos de autor este documento solo puede ser consultado en la Biblioteca Di Tella. Para reservar una cita podés ponerte en contacto con serviciosbiblio@utdt.edu. Si sos el autor de esta tesis y querés autorizar su publicación en este repositorio, podés ponerte en contacto con repositorio@utdt.edu.es_AR
dc.format.extent67 p.es_AR
dc.format.mediumapplication/pdfes_AR
dc.languageenges_AR
dc.rightsinfo:eu-repo/semantics/restrictedAccesses_AR
dc.subjectDigital Analyticses_AR
dc.subjectAnálisis de datoses_AR
dc.subjectPredicción tecnológicaes_AR
dc.subjectPrevisiones de ventases_AR
dc.titleApplications of Multi-Touch Attribution Modellinges_AR
dc.typeinfo:eu-repo/semantics/masterThesises_AR
thesis.degree.nameMaster in Management + Analyticsen
thesis.degree.grantorUniversidad Torcuato Di Tellaes_Ar
thesis.degree.grantorEscuela de Negocioses_Ar
dc.subject.keywordMulti-Touch Attributiones_AR
dc.subject.keywordLogistic regressiones_AR
dc.subject.keywordMarkov chaines_AR
dc.subject.keywordGoogle Analyticses_AR
dc.subject.keywordAdformes_AR
dc.subject.keywordClick Streames_AR
dc.subject.keywordData-Driven Modellinges_AR
dc.type.versioninfo:eu-repo/semantics/acceptedVersiones_AR


Ficheros en el ítem

Thumbnail

Este ítem aparece en la(s) siguiente(s) colección(ones)

Mostrar el registro sencillo del ítem