746 Jon M. Huntsman Hall
730 Walnut Street
University of Pennsylvania
Philadelphia, PA 19104
Ron Berman and Eleanor McDonnell Feit (Under Revision), Enhancing Power of Marketing Experiments Using Observational Data.
Abstract: Recent research has shown that randomized controlled trials designed to measure the average treatment effect of an advertisement (holdout experiments) are often under-powered and require unreasonably large samples to show that a campaign is profitable. We develop a method to increase the power of holdout experiments. If viewers can be ranked based on their responsiveness to advertising, the power to detect advertising response can be substantially improved by stratifying viewers into groups and using a post-stratified estimator of the advertising effect. We show that past observational data about consumers can be used to rank them by their ad responsiveness, even if past ad exposures were targeted. Thus firms with prior panel data on customer response to advertising (e.g. CRM data) can use this observational data to increase the power of holdout experiments. We apply the stratification approach to re-analyze several direct mail and email experiments for a retailer and an online ticket seller. In both settings, prior data on targeted marketing communications and transactions is available and we use this information to stratify customers. In several experiments, the stratified analysis improves the accuracy of the estimate of average ad response for the population by 10-20%.
Ron Berman and Zsolt Katona (Under Revision), The Impact of Curation Algorithms on Social Network Content Quality and Structure.
Abstract: Curation algorithms are selection and ranking algorithms that social media platforms use to improve user experience. This paper analyzes the impact of curation algorithms on the number of friends consumers connect to and the quality of content created by producers. The model takes into account both vertical and horizontal differentiation and analyzes three different types of algorithms. The results show that without algorithmic curation, the number of friends an individual has and the quality of content on the platform are strategic complements. Introducing algorithmic curation makes consumers less selective in their follower lists when content quality is low. In equilibrium, producers of content receive lower payoffs because they enter into a contest leading to a prisoner’s dilemma. The quality of content on the platform may increase if the marginal cost of producing this quality is high enough. Both of these effects may result theoretically in more diverse content consumption, but in equilibrium we find that a perfect filtering algorithm may reduce the horizontal distance of matched content resulting in a filter bubble. We identify an algorithm that focuses on filtering low quality items that results in higher quality of content as well as higher diversity under specific conditions.
Ron Berman, Colman Humphrey, Shiri Melumad, Robert Meyer (Under Revision), When Form Trumps Substance: A Dynamic Analysis of Microblogging During the 2016 U.S. Presidential Primary Debates.
Abstract: Microblogging sites such as Twitter play an important role in the transmission of news and opinion about economic, social and political events. There is, however, little guarantee that the content provides an unbiased account of the original events. In this paper we explore this issue by analyzing how the substantive and affective content of Tweets evolved throughout a series of important political events in 2015 and 2016: the Republican presidential primaries. We find that while the debates were in progress users Tweeted and retweeted a mix of policy-related and sensationalist topics, but that after the debates retweeting focused primarily on sensationalism. An analysis of content choice over time suggests that the persistence of such content was associated with a social diffusion process wherein sensationalist Tweets were most often initiated by “little voices” who had small followings and tended to post infrequently, then spread (through retweeting) by “shouters” who also had small followings but posted more actively, and finally were sustained after the debate by “leaders” with large followings who tended to Tweet selectively.
Ron Berman Case Study: United by Blue.
Description: United by Blue, an outdoor apparel brand, located in Philadelphia focused on ocean and waterway conservation. The company ran two crowdfunding campaigns; one in 2012 and the second in 2014 which surpassed its funding goal tenfold. The case describes the sequence of strategic choices the company must make in order to launch its third crowdfunding campaign for a highly innovative winter jacket. Through detailing the structure of crowdfunding platforms, campaign creations, and decisions needed to be taken by entrepreneurs, the case illustrates tradeoffs facing entrepreneurs and how crowdfunding can operate as both a marketing channel as well as a financing channel for firms. To request permission to use the case for teaching purposes, please contact the Wharton Course Materials Initiative
Ron Berman, Alexander Saldanha, Keshore Vunmaro Advertising Conversion Attribution (US Patent 8775248 B1).
Description: An advertising attribution system determines an attribution value for a set of advertising modalities associated with a conversion event. The modalities each provided an advertisement to a user who performed the conversion event. A conversion value associated with each of a plurality of modality subsets is determined representing the value to the advertiser of providing advertisements by the modalities in each modality subset. Based on the conversion value of each modality subset, a marginal value for each modality is determined for the set of modalities associated with the conversion event.
Ron Berman (Forthcoming), Beyond the Last Touch: Attribution in Online Advertising.
Abstract: Online advertisers often utilize multiple publishers to deliver ads to multi-homing consumers. These ads often generate externalities and their exposure is uncertain, which impacts advertising effectiveness across publishers. We analytically analyze the inefficiencies created by externalities and uncertainty when information is symmetric between advertisers and publishers, in contrast to most previous research that assumes information asymmetry. Although these inefficiencies cannot be resolved through publisher side actions, attribution methods that measure the campaign uncertainty can serve as an alternative solution to help advertisers adjust their strategies. Attribution creates a virtual competition between publishers, resulting in a team compensation problem. The equilibrium may potentially increase the aggressiveness of advertiser bidding leading to increased advertiser profits. The popular last-touch method is shown to over-incentivize ad exposures, often resulting in lowering advertiser profits. The Shapley value achieves an increase in profits compared to last-touch. Popular publishers and those that appear early in the conversion funnel benefit the most from advertisers using last-touch attribution. The increase in advertiser profits come at the expense of total publisher profits and often results in decreased ad allocation efficiency. We also find that the prices paid in the market will decrease when more sophisticated attribution methods are adopted.
Ron Berman and Zsolt Katona (2013), The Role of Search Engine Optimization in Search Marketing, Marketing Science.
Abstract: This paper examines the impact of search engine optimization (SEO) on the competition between advertisers for organic and sponsored search results. The results show that a positive level of search engine optimization may improve the search engine's ranking quality and thus the satisfaction of its visitors. In the absence of sponsored links, the organic ranking is improved by SEO if and only if the quality provided by a website is sufficiently positively correlated with its valuation for consumers. In the presence of sponsored links, the results are accentuated and hold regardless of the correlation. When sponsored links serve as a second chance to acquire clicks from the search engine, low-quality websites have a reduced incentive to invest in SEO, giving an advantage to their high-quality counterparts. As a result of the high expected quality on the organic side, consumers begin their search with an organic click. Although SEO can improve consumer welfare and the payoff of high-quality sites, we find that the search engine's revenues are typically lower when advertisers spend more on SEO and thus less on sponsored links. Modeling the impact of the minimum bid set by the search engine reveals an inverse U-shaped relationship between the minimum bid and search engine profits, suggesting an optimal minimum bid that is decreasing in the level of SEO activity.
MKTG 270 explores the digital marketing environment from both a consumer and business perspective. The course provides an overview of various online business models and delves into digital advertising and social media marketing techniques and technologies. A mixture of case studies, guest speakers and assignments, including one that uses real advertising data, translates theory into practice. It is recommended that students enrolling in the course be comfortable using Excel and are knowledgeable in applying regression analysis techniques. Students who would prefer a less technical course may wish to take MKTG 227, Digital Marketing and Electronic Commerce, a half cu course offered by the department.
MKTG 770 explores the digital marketing environment from both a consumer and business perspective. The course provides an overview of various online business models and delves into digital advertising and social media marketing techniques and technologies. A mixture of case studies, guest speakers and assignments, including one that uses real advertising data, translates theory into practice. It is recommended that students enrolling in the course be comfortable using Excel and are knowledgeable in applying regression analysis techniques. Students who would prefer a less technical course may wish to take MKTG 727, Digital Marketing and Electronic Commerce, a half cu course offered by the department.
This is a continuation of MKTG 954. This doctoral seminar reviews analytical models relevant to improving various aspects of marketing decisions such as new product launch, product line design, pricing strategy, advertising decisions, sales force organization and compensation, distribution channel design and promotion decisions. The primary focus will be on analytical models. The seminar will introduce the students to various types of analytical models used in research in marketing, including game theory models for competitive analysis, agency theory models for improving organization design and incentives within organizations, and optimization methods to improve decision making and resource allocation. The course will enable students to become familiar with applications of these techniques in the marketing literature and prepare the students to apply these and other analytical approaches to research problems that are of interest to the students.
Taught collectively by the faculty members from the Marketing Department, this course investigates advanced topics in marketing. It is organized in a way that allows students to 1) gain depth in important areas of research identified by faculty; 2) gain exposure to various faculty in marketing and their research values and styles; and 3) develop and advance their own research interests.
This course is taught collectively by the faculty members from the Marketing Department. It is designed to expose Doctoral students to the cutting-edge research in marketing models in order to help them to define and advance their research interests. This course will offer: in-depth discussions on some important topics in marketing by experts in respective areas; tools, and methodologies required for conducting research in those areas; broad exposure to our faculty members and their proven research styles.
Most marketers are doing something called 'p-hacking' in data analytics, potentially leading to wrong results, according to new Wharton research.Knowledge @ Wharton - 2018/08/16