Purchase Decision and Online Browsing Understanding Consumer Search Process And The Impact Of Service Provider Brands

INTRODUCTION

Consumers pre-purchase activation and path to conversion has intrigued academicians and marketers over the years. Of late the availability of clickstream data and new data sources that capture granular advertising data from traditional media (TV, Radio, Print, OOH, etc.) and digital media assets (display, paid and natural search, text links, and content networks) helps us explore the impact of marketing investments and quantify its impact on business performance and consumer decision making in more detail. The marketing funnel (see Figure A) is a key conceptual framework that is routinely used by practitioners to deconstruct the marketing activation and identify key issues.

Figure A: Marketing Funnel

CAPTURING ALL THESE effects along the marketing funnel and explaining the impact of interactions across these various stages is an area that requires seminal work and in this book, the author takes a few first steps using clickstream data to explore the consumer browsing and purchase behavior in airline ticket purchases using a joint framework. The author also extends this work into a multi-category framework to explain purchases in the travel category (air, hotel, and car rental). In the first essay we use a three-stage model to study (i) the choice of the first website visited (ii) the duration of browsing on travel websites before making a purchase and (iii) the choice of the website where consumers will make the purchase, and how a later stage choice is affected by decisions in previous stages. In the second essay of the dissertation we investigate online purchase behavior at the basket level and model the multi-category purchases in the travel product category. Any analysis using single-category data provides only a partial view of consumer behavior that ignores possible dependencies between consumer purchase outcomes across multiple products in the basket. This leads to a biased understanding of consumer purchase decisions as it pertains to basket purchases and transactions. We use a two-stage model to study (i) the propensity of consumers to purchase a combination of travel products as a basket and (ii) the choice of the website where consumers will make those purchases.

Significant insights are obtained that help us understand the consumer search process as well as the impact of service provider brands compared to travel portals when it comes to making purchases in the travel category. The scope of this book and the modeling efforts in this empirical research are limited by any limitations contained in the ComScore clickstream dataset from Wharton Research Data Services.

Amongst the 27 categories that the ComScore data set is comprised of Travel with 128.28 transactions (see Figure B) constitutes the highest median number of daily transactions during the study period in any category.

Figure B: Median number of daily transactions

Median number of daily transactions

Travel also constitutes nearly 50% of total online spending for households who bought travel in our dataset. ComScore also estimates the non-travel spending market in the US to be about $102bil in 2006 (online travel is a $70bn market in the US). Households also spend twice the time on travel websites where they can make a purchase compared to information-only websites. For these reasons we investigate the travel category in more depth in this dissertation.

An analysis of the travel browsing and purchase data provides a few useful insights that are relevant to understanding this dataset. We notice 68% of households (Figure C) make one or two travel purchases during the six months we study (Jul- Dec 2002). A closer look at the basket of travel products reveals single product purchases constitute 82% of all purchases with Air travel purchases being the most dominant at constituting almost half of all transactions (Figure D).

Figure C: Travel purchase transactions made by households