A Non-parametric Approach to the Multi-channel Attribution Problem

Yadagiri, Meghanath Macha; Saini, Shiv Kumar; Sinha, Ritwik

doi:10.1007/978-3-319-26190-4_23

Cited by 12 publications

(9 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The academic study of this ecosystem has focused a lot on the relationship between the advertiser and the publisher [5]. Additionally, the relationship between the advertiser and the audience member has been studied; in particular, the question of estimating the value of individual channels and ads have been studied [15,14,19]. But, unlike in OSNs, little work has been published in the study of the relationship between the publisher and the audience members.…”

Section: Related Workmentioning

confidence: 99%

Audience Prism

Kumari

Dhamnani

Bhatnagar

et al. 2016

Proceedings of the 3rd IKDD Conference on Data Science, 2016

Self Cite

View full text Add to dashboard Cite

The largest Media and Entertainment (M&E) web portals today cater to more than 100 Million unique visitors every month. In Customer Relationship Management, customer segmentation plays an important role, with the goal of targeting different products for different segments. Marketers segment their customers based on customer attributes. In the non-subscription based media business, the customer is analogous to the visitor, the product to the content, and a purchase to consumption. Knowing which segment an audience member belongs to, enables better engagement. In this work, we address the problems: 1) How can we segment audience members of an M&E web property based on their media consumption interests? 2) When a new visitor arrives, how can we classify them into one of the above defined segments (without having to wait for consumption history)? We apply our proposed solution to a real world data-set and show that we can achieve coherent clusters and can predict cluster membership with a high level of accuracy. We also build a tool that the editors can find valuable towards understanding their audience.

show abstract

Section: Related Workmentioning

confidence: 99%

Audience Prism

Kumari

Dhamnani

Bhatnagar

et al. 2016

Proceedings of the 3rd IKDD Conference on Data Science, 2016

Self Cite

View full text Add to dashboard Cite

show abstract

“…Zantedeschi, Feit, and Bradlow (2015) proposed a hierarchical Bayesian model for measuring multi-channel advertising response. Yadagiri, Saini, and Sinha (2015) developed non-parametric and semi-parametric approaches to estimate the value contributed from different marketing channels. Methodologically, we adopt and extend the work developed by Yadagiri et al (2015), whereby we develop a parametric approach to calculate the value function from each stage of the search process.…”

Section: Literature Reviewmentioning

confidence: 99%

Multi-Click Attribution in Sponsored Search Advertising: An Empirical Study in Hospitality Industry

Anderson

Cheng

2017

Cornell Hospitality Quarterly

View full text Add to dashboard Cite

Sponsored search advertising has become a dominant form of advertising for many firms in the hospitality vertical, with Priceline and Expedia each spending in excess of US$2 billion in online advertising in 2015. Given the competition in online advertising, it has become essential for advertisers to know how effectively to allocate financial resources to keywords. Central to budget allocation for keywords is an attribution of revenue (from converted ads) to the keywords generating consumer interest. Conventional wisdom suggests several ways to attribute revenues in the sponsored search advertising domain (e.g., last-click, first & last-click, or evenly distributed approach). We develop a multi-click attribution methodology using a unique multi-advertiser data set, which includes full advertiser and consumer-level click and purchase information. We add to the literature by developing a two-stage multiclick attribution methodology with a specific focus on sponsored search advertising in the hospitality industry with which we develop a parametric approach to calculate the value function from each stage of the estimation process. Given our multi-advertiser data set, we are able to illustrate the inefficiency of single-click attribution approaches, which undervalue assist clicks while overvaluing converted clicks.

show abstract

“…There have been several studies that offered data-driven approaches to the attribution to overcome the weaknesses of standard heuristic models. Yadagiri et al (2015) and Nissar and Yeung (2015) use Shapley value in their non-parametric approach to attribution as a game theory-based model. In his thesis, Rentola (2014) used two models: binary logistic regression to classify customers to converters and nonconverters (purchasers/non-purchasers), as well as a logistic regression model with bootstrap aggregation.…”

Section: Introductionmentioning

confidence: 99%

Impact of direct traffic effect on online sales

Kakalejčík

Bucko

Danko

2020

JRIM

View full text Add to dashboard Cite

Purpose This study aims to analyze the impact of newly created brand awareness on customer’s buying behavior in online environment. Design/methodology/approach The authors analyzed more than 280,000 online customer journeys from four e-commerce stores based in Slovakia. Within the results of the interaction analysis of individual customer journeys, the authors determined three criteria based on the level of theoretical brand awareness. The purpose was to determine their occurrence in real-world data. Findings It was found that each of the specified criteria accounts for the significant share of the company’s revenues. Based on these criteria and the level of their occurrence, the authors introduced the term direct traffic effect. Research limitations/implications Because of the available Web analytics tools, the data might be imprecise because of data collection issues. There is also ambiguity in the interpretation of the customer journey. Practical implications The company can build awareness among prospective customers by offering them a positive customer experience during the first interactions online. Data proved that customer will not only repeatedly visit the website from the direct traffic source but also his customer journey will end with the purchase of the company’s products. Originality/value This paper fulfills the need for further research on the impact of multi-channel marketing on brand awareness and consumer behavior, respectively.

show abstract

A Non-parametric Approach to the Multi-channel Attribution Problem

Cited by 12 publications

References 9 publications

Audience Prism

Audience Prism

Multi-Click Attribution in Sponsored Search Advertising: An Empirical Study in Hospitality Industry

Impact of direct traffic effect on online sales

Contact Info

Product

Resources

About