Abstract:
A robust finding in brand extension literature is that perceptions of high similarity or fit with the extension result in favorable extension evaluations. In this research, we examine two different types of fit – based on a category prototype or a category exemplar, and suggest that consumers’ implicit theories regarding the changeability or fixedness of human traits will impact the evaluations of extensions that vary on prototype or exemplar fit. In the first two studies, we find that while entity theorists evaluate extensions with high prototypical fit higher, incremental theorists assess extensions with high exemplar fit more favorably (study 1), regardless of whether consumers organize brand and product information around a brand category or a product category (study 2). In study 3, we further show that a failed extension with high prototypical fit is evaluated lower by entity theorists, while incremental theorists’ evaluations remain unaffected. We conclude with a discussion on future research and managerial implications of our findings. Copy of the paper is attached:

Abstract:
A rationed promotion occurs when a firm offers a limited quantity of product at a below market clearing price. Examples include airline supersaver fares and Black Friday discounts on consumer durables. We characterize the optimal pricing strategy for a monopoly firm with limited capacity facing generalized linear demand. Under a binding capacity constraint and a suitable convex demand function we show that it is optimal to offer a limited quantity at a price below the market clearing price, then sell the balance of inventory at a high reserve price. We consider duopoly competition in a similar environment and show that there is a unique dominant strategy Nash equilibrium in which discount prices are higher than optimal resulting in a loss of efficiency in market segmentation and corresponding decrease in total revenues. Nonetheless, by clearing excess inventory, promoting firms avoid direct competition for their most profitable customers and retain most of the revenues available to the monopolist. We extend these results to situations of asymmetric competition and horizontally differentiated products and find that in this context competition can lead to either higher or lower prices depending on the degree and nature of preference heterogeneity. Finally we relate rationed promotion to the second and third degree price discrimination strategies commonly used in yield management. For a monopolist seller we show that rationed promotions complement third degree price discrimination under appropriate conditions, but substitute for 2nd degree price discrimination when willingness to pay is only weakly correlated with preference for non-price attributes and demand exhibits the required convexity in price.

Abstract:
As “modern” retail expands into emerging markets with promises of better quality, variety and lower prices, they face competition from millions of “traditional” mom-and-pop stores. This paper addresses a basic customer selection question facing a modern retailer entering an emerging market: Should the retailer begin by targeting higher socio economic classes or value conscious lower socio economic classes? The conventional wisdom is that lower socio-economic classes are likely to resist new shopping practices; hence modern retail should target the rich initially. Economic value-based arguments suggest that while modern retail’s better quality and variety may attract higher socioeconomic classes, lower prices will be attractive to lower socioeconomic classes; hence modern retail should also target the lower socioeconomic classes. Further, given the lower density of modern retail, would it be patronized by the rich with better transportation or the poor with lower opportunity costs of time? We address these questions using survey and household panel data on FMCG purchases from India. Household level models of store format and expenditure share choice show a nonlinear relationship between modern store patronage and socio economic status. Both high and low socioeconomic households adopt more often and spend more on modern retail relative to medium socioeconomic households. The results based on observational panel data are corroborated with self-reported reasons: high socio-economic households buy from modern retail for higher quality and variety, low socioeconomic households buy due to low prices and lower opportunity cost of time. We conclude that given the substantially larger size of lower socio-economic households in emerging markets, it is critical for modern retailers to appeal to lower middle class customers even in the early stages of entry.

Abstract:
We develop a theoretical model of the pricing of a new durable good to determine the conditions under which the seller can utilize inferences about the buyer’s willingness to pay based not only on her decision to trade in the old good but also on its characteristics. We test the predictions of our theory using transaction data for new car purchases that includes information on the vehicles that new car buyers traded in. The results show that dealers infer a higher willingness to pay and charge higher prices to consumers who traded in a used vehicle than to those who did not. We also find that dealers charge even higher prices to those consumers who trade in used cars that were of the same make and model as the new one. We show that ignoring the information available in the type of trade-in can inflate the estimated value of the trade-in effect

Abstract:
Many judgments consumers make occur within a sequence of evaluations and therefore are likely to be influenced by context effects. Assimilation (contrast) refers to a positive (negative) relationship between the value people place on a target and the value they place on the context. Past work on assimilation and contrast in sequential evaluation presupposes only one or the other occurs. We propose that stimuli experienced within a sequence may be influenced simultaneously by assimilation and contrast to stimuli experienced earlier in the sequence, and we estimate a hierarchical Bayesian model that confirms these simultaneous effects within a unique, real-world dataset. We find individual evaluations (scores) are influenced by contrast effects to the scores of their immediate predecessors and to extreme scores within the sequence, while simultaneously influenced by assimilation to the first score. This work is the first to document that multiple assimilation and contrast effects simultaneously influence sequential hedonic evaluations.

Abstract:
The literature suggests that brand equity can be split into two parts - an attribute-based equity and a non-attribute based one that captures consumer preferences beyond the utility offered by individual attributes. In addition to measuring attribute-based equity, firms deploying portfolios of products within complex branding structures often seek to measure the presence, distribution and evolution of these potentially heterogeneous non-attribute based unique branding associations - labelled 'residual equity' – at each distinct layer of a product’s brand hierarchy. The authors develop and operationalize a robust and flexible Bayesian semiparametric model to first separate the attribute-based equity from latent residual equity, to jointly estimate this multi-level residual equity and to allow residual equity to exhibit statedependence using a random-walk prior. The model is empirically illustrated on syndicated US national beer sales data. The authors find significant, heterogeneous and temporally stable residual equity presence across the brand hierarchy and highlight some substantive implications arising therein.

Abstract:
Although firms spend substantial amounts of money to build and maintain brand equity, brand equity is fragile to product-harm crises such as product recalls. Product-harm crises can adversely affect consumer perceptions and preferences as well as weaken the effectiveness of marketing activities (Van Heerde et al. 2007). Despite the potentially devastating effects of these crises, most firms are inadequately informed and underprepared to handle them (Dawar and Pilluta 2000). How do product recalls affect brand equity? How do they influence the effectiveness of different types of advertising, such as product feature advertising, product model advertising and sponsorship advertising? How should firms allocate their advertising spending across different types? The answers to these questions will enable firms make better advertising allocation decisions and rebuild brand equity in response to product recalls. In this study, we develop a state space model of brand equity based on Kalman filtering to capture the dynamics related to spending on different types of advertising. We then integrate the Kalman filter with a random coefficient demand model based on Berry, Levinsohn, and Pakes (1995). The model allows us to capture the direct effects of product recall on brand equity as well as the indirect effects of recall on brand equity through the effectiveness of different advertising types. We estimate our model on a carefully compiled dataset on the automobile industry, comprising 35 auto companies that had a total of 1,206 recalls during 1997-2002. The dataset includes advertising data on five types, product feature advertising, recalled product model advertising, non-recalled product model advertising, promotional advertising, and sponsorship advertising. Using in a holdout sample, we perform counterfactuals to show how our proposed model and its parameter estimates can be used to improve the allocation of advertising budget across the different types.Our results reveal that product recall has no direct effect on brand equity. However, it has significant negative influence on the effectiveness of each advertising type, with sponsorship advertising suffering the most. The reallocated advertising budget based on our model is associated with higher sales and profits. In addition, our results have other important managerial implications. They suggest that, in response to product harm crises, firms should avoid advertising the recalled model, cut sponsorship advertising, and spend more on product feature advertising. Keywords: Advertising, Brand equity, Product harm crisis, Kalman filter, BLP approach.

Abstract:
We propose an approach to prioritize service improvements based on the twin objectives of higher customer value and lower cost. The approach involves (1) conducting qualitative customer studies to identify a list of possible service improvements, (2) conducting a quantitative, conjoint-like survey to determine values customers attach to each of the improvements, (3) collecting data on the costs of making the service improvements, and (4) putting the data in (2)–(3) together to prioritize improvements using a “bang for the buck” rule. The approach also allows for maximizing the likely increased service usage resulting from any subset of service improvements subject to a budget constraint. We illustrate the proposed approach in the context of improving passenger train service between a pair of cities in India. An adaptive self–explicated approach is used for obtaining customer values and cost estimates. The customer values so elicited display substantial face validity.

Abstract:
The spread of social media has meant that User-generated Content (UGC) has become an important form of communication. Most of the past research in social media has concentrated on analyzing message virality or the causes of why certain messages go viral. We posit that accounting for media virality, a phenomenon where messages are transferred beyond the original media they were carried in, has become imperative. In this paper we argue that media virality is a function of product characteristics and propose a framework for capturing this type of virality using just two dimensions, the inter media elasticity and inter media duration, together referred to as an entity’s Inter Media Reactivity (IMR). We illustrate the application of our concept using data on movie stars across several media. We calculate the IMR for each star, and demonstrate how media virality differs across each, and also analyze the starspecific characteristics that drive media virality. Subsequently, we relate the media virality of stars to the performance of their movies. Our research thus provides a theoretical contribution to the literature by exploring media virality while also providing several managerially relevant substantive insights about the motion picture industry. Key Words: Social media, advertising, motion picture industry

Abstract:
Marketing science can claim many successes in directing management decision making in practice. Segmentation and mapping tools, conjoint and discrete choice models, and econometric analyses of promotional scanner data represent a few of the area in which analytical techniques have influenced the planning and execution of marketing activities. However, when one looks at the level of penetration of these tools, I argue that the picture is considerably less rosey. This seminar firstly examines the level of impact on practice that marketing science has had from the perspectives of academics, consultants and managers (Roberts, Kayande, and Stremersch 20111) and then digs deeper into the characteristics of that work which has had the most impact, as measured by the finalists in the Gary Lilien Marketing Science Practice Prize (Lilien, Roberts, and Shankar 20112). 1 Roberts, John H., Ujwal Kayande, and Stefan Stremersch (2011) “From Academic Research to Marketing Practice: Exploring the Marketing Science Value Chain” Working Paper, The Australian National University. 2 Lilien, Gary L., John H. Roberts, and Venkatesh Shankar (2011) “Effective Marketing Science Applications: Insights from ISMS–MSI Practice Prize Finalist Papers and Projects” Marketing Science Institute Working Paper Series, Report No. 11-101

Abstract:
Although the short-run response of the financial markets to announcements of marketing investments has been well-established in the extant literature, less is known about how the basic information inherent in these announcements flows from the “black box” of the firm to external market participants. In response, we conceptualize that since marketing investments are determined by individuals with the best knowledge of the firm’s future environment, announcements of a marketing investment provide valuable information about the future productivity of that marketing investment. Our analysis yields two distinct predictions. First, market reaction to an investment decrease announcement will be less pronounced than market reaction to an investment increase announcement, a result that stands in sharp contrast to received theory. This is because the mean-shifting and uncertainty-reducing aspects of new information tend to mitigate each other for decrease announcements but reinforce each other for increase announcements. Second, the information content of marketing investment announcements will be intimately related to investors’ prior information; consequently, firms that announce similar marketing investments can provoke markedly different market reactions. Examining a sample of sales force size change announcements by 131 firms in the pharmaceutical industry, we find broad support for our analytical conceptualization.

Abstract:
Firms are increasingly facing consumers with a heightened focus on the social, employee, and environmental consequences of their marketing activities and stakeholders with a greater capacity to exert influence on firm and customer behavior. This trend points to a need for marketers to find competitive advantage and differentiation by evolving from a single minded customer focus towards an understanding of a broader set of stakeholder requirements. The emerging Stakeholder Marketing literature has articulated such a need and proposed an initial formulation of its implications for firm behavior and performance. However, consensus on the capabilities required by firms to adopt a stakeholder orientation has not been reached nor the consequences of such actions tested. This research draws on depth interviews of managers and a theory-in-use approach to introduce Constituents-Based Marketing (CBM) as the capability to develop product offerings that can address the needs of multiple constituencies at the same time. An empirical test across 44 countries uses instrumental variable regression and finds CBM capability to be positively associated with sales growth, employee engagement, and company trust. The analysis finds the CBM effects to be stronger in highly networked as well as competitive conditions. External stakeholder pressures serve as a motivator, while firm cross functional processes and market orientation provide the ability for a firm to develop CBM capabilities. The results are robust to estimation methods and endogeniety concerns. We present theoretical implications and managerial guidelines for the generation of market intelligence and the design of marketing organizations.

Abstract:
Many organizations (DMUs) are comprised of networked sub-DMUs or departments, i.e., the outputs of some departments serve as inputs to others and vice versa. For example, in media-platform firms like newspaper companies, the outputs of the editorial department are inputs to the advertising department and vice versa. Extant approaches for efficiency analysis either ignore the network effects or make parametric assumptions to enable statistical inference on the estimated efficiencies. Hence, we develop a new method that incorporates the network effects and furnishes nonparametric inference on efficiencies. Applying the proposed method to the US daily newspaper industry, we not only demonstrate that it outperforms existing approaches, but also that it yields new insights into department-level efficiencies. Keywords: Efficiency Analysis, Sliced Inverse Regression, Media-Platforms, Networked DMUs

Abstract:
Stores often follow a certain pricing strategy. They vary their prices because of several reasons – as a response to changing inventory levels, competitor pricing, change in costs on the supply side, nearing expiration date of perishable goods, manufacturer’s promotion or at times as a strategy to increase store foot-falls. This varying of prices is reflected as – EDLP (Every Day Low Price, where prices are kept at a constant level and Wal-Mart is one successful example) or Hi-Lo (where prices fluctuate due to frequent promotions and this is followed by well-known grocery chains like Dominick’s Finer Foods in Chicago). Stores also differ in the frequency by which they change prices. A pricing decision may be both short term and long term (Shaffer and Zhang, 2002). For instance, on a short-term, a retailer could use price discounts to compete against weaker brands, price discriminate among brand switchers or increase consumption (Bronnenberg, et al, 2006). On a long-term, the price change could reflect the changes in overall cost structure, demand or market structure. A retailer might also use discounts to drive category or store level profitability. Often the frequency of price change is also a result of the planning horizon set by a particular retailer (Bronnenberg, et al, 2006). Of late, researchers like Kadiyali, Sudhir and Rao (2001); Pauwells, et al, 2004; Bronnenberg, et al, 2006 have emphasized the need to study this frequency of price change. Pauwells, et al, 2004, for instance mention “as there is increasing evidence that the same relationship need not hold among two variables at different frequencies, various substantive marketing problems may warrant further investigation along that dimension”. From a customer’s perspective, she is able to explain some of this variability while not able to explain other. For instance, a customer is likely to explain why the price of milk is the highest close to the date of manufacture and why is it at a huge discount closer to the date of expiration. She is likely to have an explanation for seasonal discounts during festival times. However, other instances of price variations go unexplained. In this study we examine only unexplained variability (both depth and frequency of price change) From a customer’s perspective, unexplained volatility, both magnitude (depth) and frequency in price, is likely to be perceived as bad. It is likely to erode the ‘trust’ a customer has on the store and its offerings. A customer is likely to doubt the store’s quality, integrity or benevolence as a consequence of unexplained volatility. This is also likely to affect her decision to buy the store’s own brand. Some evidence of this effect can be found in the existing literature like Dhar and Hoch, 1995 who look at EDLP (low price variability) vs. Hi-Lo (high price variability) strategy of stores’ impact on store brand penetration. They find a significant relationship between the two. Using scanner panel data (from IRI) of two categories – diapers and deodorants, we also found a significant negative correlation between weekly price variation and store brand penetration. The literature on store brands penetration has focused on two aspects – understanding the cross category variation in store brand penetration through analysis of scanner panel data (Dhar and Hoch, 1995; Hoch and Banerjee, 1993; Sethuraman, 1992) and understanding the store brand customer using consumer-level data (Ailawadi and Neslin). When we say volatility is ‘generally’ is perceived by consumers as bad, we say this because certain consumers may not regard it as bad. For instance, Deal-prone consumers have been shown to value transaction utility rather than, or in addition to, the acquisition utility associated with buying on deal (i.e., buying on deal has psychological benefits irrespective of the financial consequences, Lichtenstein, Netemeyer, & Burton, 1990). Such customers are likely to welcome variability in pricing rather than look at it as an irritant. Customers are reluctant to buy store brands in product categories where the perceived risk is higher (Batra and Sinha 2000; Narasimhan and Wilcox 1998). Some of the product categories have well-established higher perceived risk (for example, cold and flu medicine) than others (for example, gift wrap). The effect of variability in pricing and frequency of price change on trust and subsequently store brand purchase is likely to be greater in product categories where trust is relatively more important. Based on the above, this study proposes to ask the following questions: Does store price volatility (magnitude, frequency, explained/ unexplained) effect store brand penetration? 1. Does ‘Trust’ play a mediating role between them? 2. Do customer characteristics (deal-proneness) play a moderating role in the relationship? 3. Does the trust associated with the product category also moderate the relationship?

Abstract:
In this paper, we investigate the causal effect of position in search engine advertising listings on outcomes such as click-through rates and sales orders. Since the position is typically determined through an auction, there are significant selection issues in measuring position effects. Correlational results are likely to be biased due to the selection in position induced by strategic bidding by advertisers. Additionally, experimentation is rendered difficult in this situation by competitors’ bidding behavior, which induces selection biases that cannot be eliminated by randomizing the bids for the focal advertiser. We show that a regression discontinuity approach is a feasible approach to measure causal effects in this important context, where other approaches to obtaining causal effects are typically infeasible. and apply it to a dataset of 23.7 million daily observations containing information on bids, search advertising results and linked outcomes. Our data set is unique in that it contains information not only for the firm but also its major competitors who are advertising in the same category. We find in our empirical application that causal position effects are significantly underestimated if the selection of position is ignored. The data set also contains information on two advertising targeting options offered by Google: Exact and Broad match and our causal estimates provide insights into the value of this type of semantic targeting. We find important differences in the effects of position for exact vs. broad match keywords, with exact match keywords showing strong position effects at the top most position, and broad match keywords having strong position effects only lower down.We are also able to study weekday and weekend effects and find that position effects are lower on the weekend than on weekdays.

Abstract:
We investigate the relative influence of human capital, distribution, opening week revenues and movie attributes on how long movies last in theaters. The focus of our investigation is the Italian market. We use a survival model with a parametric proportional hazards specification for our analysis. We also account for unobserved heterogeneity, endogeneity of opening week revenues, and a stochastic censoring mechanism, in our model and take a Bayesian approach for our analysis. Our results suggest that the quality and breadth of experience of actors, and the number of screens that a movie is initially released in, have a negative effect on longevity. Opening week revenues and attributes such as the quality of screenwriters, running time, some genres and some types of content ratings such as PG-14 , however, have a positive effect. Implications of the research and opportunities for future research are also discussed. Keywords: movies, playability, survival analysis, proportional hazards, stochastic censoring, Bayesian analysis.

Abstract:
We estimate the joint impact of frequency reward and customer tier components of a loyalty program on customer behavior and the corresponding sales. We provide an integrated analysis and measurement of a loyalty program incorporating customers’ purchase and cash-in decisions, points pressure and rewarded behavior effects, heterogeneity, and forward-looking behavior. We focus on measuring these effects for a hotel loyalty program. The results suggest interesting insights: for example, both components produce net gains in sales, but do not synergize, i.e., they are not complements in terms of incremental sales. We find strong evidence for points pressure, for both the customer tier and frequency reward components, using both model-based and model-free evidence. We find a two-segment solution revealing a “service-oriented” segment that highly values cash-ins for room upgrades and staying in “luxury” hotels, and a “price-oriented” segment that is more price sensitive and highly values the frequency reward aspects of the loyalty program. We conduct policy simulations that vary the reward structure and use the results to recommend adjustments in frequency reward and customer tier program requirements that improve firm revenues. We predict that compared with the existing program, an alternative design more stringent in awarding the frequency-based reward but more generous in awarding the tier-based reward increases the revenue by 8.99%.

Abstract:
We study competitive positioning and pricing strategies in conspicuous goods markets. For conspicuous products, each consumer values the product less as more consumers own it, thus exhibiting negative consumption externalities. We incorporate the effect of consumption externality for each consumer to be dependent on their location vis-à-vis the location of the firm. Specifically, we assume that a customer ‘closer’ to a firm may feel more let down if that firm caters to too many customers. Our work extends the existing literature by formally recognizing that consumers are heterogeneous in their sensitivity to product exclusivity. We find that for conspicuous goods, product differentiation is lower; a finding that explains otherwise counterintuitive empirical results in the literature. We also show that under some conditions, price competition can be higher in markets with negative consumption externalities. When firms are asymmetric with respect to consumption externality effects, we show that a firm which exerts higher externality tends to charge lower prices. Finally, firms may be better off if they reduce heterogeneity in their consumers’ sensitivity to consumption externality. Keywords: Negative Consumption Externalities, Conspicuous Goods, Pricing, Hotelling Models.

Abstract:
Mobile Marketing and Location-Based Services: Exploring Consumer Perceptions on Privacy, Utility, and Willingness-to-Pay I would like to present the results of two studies related to location-based marketing on mobile phones. Study 1 is under preparation to be submitted to ‘Information Systems Research’ and Study 2 at the AMA Summer Ed. Conference, 2011. I would like to obtain valuable feedback from your faculty. Study 1: Location-based services and applications (apps) on mobile devices are becoming increasingly popular because of the added value to personal productivity and entertainment. However, along with the added value comes a much greater (perceived) threat to personal privacy. Now, not only can technology track and use our personal information, but our location information as well. For example, some mobile applications visualize the real-time traffic congestion on a user’s commute to and from work. Because the latest smart phones are also equipped with an accelerometer which senses the user’s speed, the mobile application could sell its user’s speed and location data to the local police department in order to place speed cameras in the most strategic locations. On the other hand, the new open mobile platform has created a market place for many new developers and small businesses to enter the marketplace or move into the mobile arena. But if they want to succeed, they must be able to overcome the privacy concerns of mobile application consumers. Privacy has turned out to be the key concern of modern consumers in this 24/7 connected environment (Peltier et al, 2009). The purpose of this study is to empirically examine the significance of this increasingly relevant privacy dimension. Through two separate simulation experiments of over 800 consumers, we examine how the assurance of location information privacy (as well as mobile app quality and network size) influences users' perceptions of location privacy risk and the utility associated with the app which, in turn, affects their adoption intentions. The results indicate that location privacy assurance is of great concern and that assurance is particularly important when the app’s network size is low or if it’s quality cannot be verified. Study 2: Pay What You Want (PWYW) pricing has become a popular pricing strategy, and has attracted increasing interest from both academics and practitioners in the recent past (Kim et al, 2009). Yet, few studies have directly examined consumer behavior when sellers use this pricing across various products and distribution channels. In this study, we apply the theories of social norms, reference prices and privacy assurances to test various factors that affect consumers' willingness to purchase and pay (WTPP) for mobile applications in a PWYW condition. It is our conjecture that app designers can creatively activate social norms to positively influence mobile users’ intent to purchase an app and willingness-to-pay (WTP) than what they originally intended. The conceptual core of the paper is based on the rich stream of research pertaining to social norms and social influences (e.g., Cialdini et al. 2004). The research model and the hypotheses in this study are built on this platform and supported by multi-disciplinary theories on reference prices (Muzumdar et al. 2005; Dholakia and Simonson 2005), and network effects and privacy calculus (Culnan and Armstrong 1999; Laufer and Wolfe 1977). Two experiments with about 1200 consumers show that social information positively affects WTPP while reference prices negatively affect willingness to pay. In this, consumers tend to be greatly influenced by social information from local groups when compared to global groups. Further, privacy assurances significantly enhance consumers' willingness to pay.

Abstract:
Recognizing that initial public offerings (IPOs) represent the debut of private firms on the public stage, this study investigates how pre-IPO customer and competitor orientations (CCOs) affect IPO performance. Building on signaling theory, the authors propose that both orientations influence investors’ sentiments toward an IPO and that IPO-specific variables and facets of the organizational task and institutional environments moderate the influence of CCOs. The authors test the framework using data collected from computer-aided text analysis, expert coders, and secondary sources for 543 firms (IPOs) across 43 industries between 2000 and 2004. A Bayesian shrinkage model, which accounts for industry-specific effects and uses latent instrumental variables to account for CCO endogeneity in the IPO context, shows that these orientations positively influence IPO performance. As for the moderating role of IPO-specific variables and facets of the organizational task and institutional environments, the authors find that (1) underwriter reputation and venture funding positively moderate the effects of CCOs, (2) technological and market turbulence positively moderate and institutional complexity negatively moderates the effect of customer orientation, and (3) technological turbulence, competitive intensity, and institutional complexity positively moderate the effect of competitor orientation. The results also show that accounting for endogeneity using latent instrumental variables substantially improves the predictive validity of the model relative to alternate model specifications.

Abstract:
Marketing scientists often focus on the concept of “similarity” among brands to understand processes involved in brand switching, competition, brand loyalty, and market structure analysis. The spatial representation of “similarity” serves as a convenient method for describing, summarizing, and displaying such marketing data. Marketing researchers and managers have squeezed increasing levels of insights from such geometric configurations, also called perceptual maps. These maps have implications for definition of market segments and product positioning. This paper presents a new model of similarity, which is theoretically sound and lends itself to geometric representation of brands in a multidimensional metric space. We also provide an empirical test of the new model, and later develop a novel algorithm, which employs “similarity” type of marketing data to construct perceptual maps based on the proposed model. An empirical comparison of the results obtained from the new perceptual mapping technique versus a traditional technique previously used by marketing researchers is also presented.

Abstract:
Over the past 50 years, many studies have analyzed the response of sales and market share to advertising expenditures. Some other studies have examined this research and attempted to draw generalizations about whether advertising "works" or not. These meta-analyses have used a variety of sources of data including field experiments, market response models using aggregate time-series data, and single-source scanner panel data with information on household-level advertising exposure. The focus has generally been on the advertising-sales elasticities. However, none of the studies that have formed the meta-analysis databases have used the variable that managers really care about--profits--as the dependent variable. In this talk, I will review what we know about short-term advertising elasticities and include some new analyses of a proprietary database of over 270 brand case studies including data on the profitability of the media campaigns. I will compare the resulting advertising elasticities by different media (TV, online, magazine, outdoor, radio, newspapers) to prior findings.

Abstract:
In customer relationship management, firms manage customer value over the entire duration of the customer-firm relationship (i.e. customer lifecycle) through managing purchases, returns, and lifecycle duration. The extant literature has primarily focused on the amount ($) of purchases and returns in specific types of contexts and recognizes that investigations in other contexts are needed. In addition, the purchase and return quantity (units) decisions over customer lifecycle are not well understood in most contexts. Finally, firms commonly use explicit purchase commitments to manage customer behaviors but these have not been investigated empirically. This study investigates several issues concerning the quantities (units) of purchase and return, and purchase commitments, over entire customer lifecycle in an important contractual context that has not been investigated in depth before. The issues tested are: (1) Do customers purchase more quantity as their lifecycle progresses? (2) Do customers return more quantity as their lifecycle progresses? (3) What is the relationship of purchases and returns with customer defection? (4) Do more purchases imply more returns? (5) What is the relationship of returns with subsequent purchases and returns? And (6) What is the relationship between purchase commitments and customer behaviors of purchase, return, and defection? To answer the research questions, the paper jointly investigates purchase and return quantities, customer defection, and interpurchase time using data from a membership based direct marketing company where customer behaviors are stochastic. A comparison of the findings with those in the extant literature in other contexts reveals that many of the results are surprising and important. The paper also discusses the significance of the findings for customer management strategies of firms.

Abstract:
We study pricing of two competing manufacturers who choose uniform wholesale prices and sell through a retailer maximizing category profits. Departing from extant work we model brand demand as continuous in retail prices, not lumpy with a mass of consumers exhibiting discontinuous switching behavior. This demand model is based a model of brand choice by heterogeneous consumers. We find that retail pass-through depends on wholesale price difference between brands, not just a brand’s wholesale price. Moreover, the wholesale price difference must exceed a threshold determined by price sensitivity. We also show that if demand is sufficiently price sensitive, even if it is continuous in retail prices, equilibrium manufacturer pricing is in mixed strategies, which we interpret as trade promotions. This is in contrast to prior work that establishes the outcome of mixed strategies when demand is lumpy. We further show that in the absence of a strategic retailer that maximizes category profits, in our model there would be no trade promotions. Thus, we establish a connection between trade promotions and the presence of a strategic retailer. We also find that not unexpectedly, with low price sensitivity, no trade promotions occur, while under high price sensitivity, trade promotions occur with equilibrium in mixed strategies. The more interesting result is that price sensitivity affects both regular and promotion prices. Finally, competitive trade promotions in our model act as a check on raising regular prices, and so competition is about choosing regular prices, not just formulating a promotions strategy.

Abstract:
We conduct a thought experiment of what would happen if a marketer were to really care about his/her customers’ welfare (not merely because the customer brings revenue.) We examine a duopoly where one of the firms does not maximize profit, but instead maximizes customer surplus subject to a profit constraint. (Customer surplus for a firm is the sum of its customers’ individual customer surpluses, i.e., the dollar value the customer attaches to the product minus its price.) For the surplus-maximizing firm, profit is constrained to be at least X percent (where X% might be, say, 80-90%) of the profit it would have obtained under a profit maximization objective. The model assumes customer willingness to pay for quality is uniformly distributed and that customers follow a simple decision rule: when presented with two products of known quality and price, purchase one unit of the product which maximizes surplus, or if surplus is negative for both products, elect not to purchase any product. We further assume that both firms have the same marginal cost of production that is convex (quadratic) in quality. Competition between firms is modeled as a two-stage game, which is solvable by backward induction. In the first stage, one of the firms, whose identity is exogenously specified, moves first and decides its quality level fully anticipating the quality response of the second firm and the subsequent price competition. The second firm observes the first firm’s quality level and then decides its own quality level, anticipating the subsequent price competition. In the second stage, firms take qualities as given and choose prices simultaneously in accordance with a Nash equilibrium. Two possibilities are considered: (a) the first mover is the profit maximizing firm, and (b) the first mover is the customer-surplus maximizing firm. We compare the results to the corresponding base case of Moorthy (Marketing Science 1988) where both firms are profit maximizing. We find that firms can deliver significant additional value to their customers by forgoing small amounts of profit. The effectiveness of this strategy depends upon which firm is the first mover. In the case that the surplus-maximizing firm moves first, 5%-10% increases in customer surplus “cost” 6%-12% of potential profits. By contrast, when the surplus-maximizing firm moves second, we find that sacrificing 20% of profits is sufficient to more than triple the customer surplus it would have provided under a profit-maximizing objective. This is accomplished by the surplus-maximizing firm leapfrogging the first mover to become the higher quality producer.

Abstract:
With the growing globalization of business, international launch time window (the elapsed time between product launch in the home country and launch in the focal country) is becoming increasingly critical for the success of new products, in particular, short life cycle products, such as movies, books, music, and video games. We address important research questions related to this launch time window. What are the determinants of the launch time window? What are the relative effects of prelaunch advertising and word of mouth on the launch time window? What is the effect of launch time window on performance in foreign countries? We formulate an analytic model and develop predictions related to these questions. We empirically test these predictions and the influences of other variables using a unique dataset comprising 207 movies covering 78 countries during 2003-06. We develop and estimate a simultaneous system of equations, in which country revenues, launch time window, and prelaunch advertising are the dependent variables. Our results show that launch time window is positively associated with word of mouth but negatively related to prelaunch advertising efforts and foreign demand potential. Our findings offer new insights into the tradeoff between leveraging the word of mouth effect and investing in prelaunch advertising in determining the international launch window. Key words: International marketing strategy, new product marketing, market entry timing, econometrics, movies. This is a joint work with co-author, Reo Song, Assistant Professor, Kansas State University

Abstract:
In this paper, we examine the internal dynamics of a market through the various stages of its evolution. Using comprehensive auction data from about 300 art auctions across about 50 auction houses over a 13 year period, we examine the behavior of market participants and its impact on multiple outcomes of interest including the price formation process and the valuation of objects and brands. Our results show that the quality of market participation changes systematically over the course of market evolution. As a consequence, the antecedents or drivers of the price formation process changes significantly over time. Implications for auction houses and investors are discussed. This is joint work with Mayukh Dass of Texas Tech. and Srinivas Reddy of SMU.

Abstract:
Easy availability of information on a customer’s transactions with the firm and the pressure to establish financial returns from marketing investments has led to a dominance of models that directly connect marketing investments to sales at the customer level. Customer attitudes, on the other hand, have always been assumed to influence customer reactions to a firm’s marketing communications, but have rarely been included in models that determine customer value. We develop hierarchical dynamic models to empirically study whether inclusion of attitudes improves (a) estimates of marketing (sales) effectiveness in CLV models, and/or (b) predictions of CLV. If so, is the effect larger in retention or in spending? We develop models that enable us to fuse survey based customer attitude data with traditional CRM data. We assess whether inclusion of attitudes re-directs focus of marketing efforts, and whether the benefits outweigh the costs of collecting attitudinal information. For this study, we use monthly sales, sales calls, and survey based attitude information collected over three years from the same customers, of a multinational pharmaceutical firm. We develop hierarchical dynamic models that combine the sales and sales calls data that are available at regular time intervals, with customer attitudes that are only irregularly available, and carry out inference in the Bayesian framework. This is joint work with Rajkumar Venkatesan, University of Virginia, and Werner Reinartz, University of Cologne and INSEAD.

Abstract:
We model the impact of firm pricing decisions on different facets of the customer purchasing process in business-to-business (B2B) contexts and develop an integrated framework for inter-temporal targeted pricing to optimize long-term profitability for the firm. We model several inter-related customer decisions while accounting for customer heterogeneity, reference price effects, purchase dynamics, and endogeneity in pricing. The model is constructed to capture both the short- and long-term implications of the pricing policy. The complexity of inter-related joint customer decisions is modeled using hierarchical Bayesian copulas, which weave together different marginal distributions to form joint distributions. We capture dynamics in purchase behavior and the possible long-term impact of experienced prices using a non-homogenous hidden Markov model with multivariate interrelated state-dependent behaviors. In addition, we rely on the behavioral pricing literature in modeling the effect of price, using asymmetric reference price effects. Finally, endogeneity is accommodated using a Bayesian approach for control functions. We calibrate the model using longitudinal transaction data from a metals retailer. The results reveal several substantive insights about the short- and long-term impact of the firm's pricing decisions on each of the inter-related components of the customer’s purchasing behavior. For example, we find that the firm’s pricing decisions could have a long-term impact on its customers by shifting their preferences between a “vigilant” state - characterized by a cautious approach towards ordering and heightened price sensitivity, and a more “relaxed” state. Additionally, the proposed model exhibits superior predictive performance relative to several benchmark models.

Abstract:
This research looks at the type of information that is preferred by high vs. low ambivalence individuals. It is argued that individuals with high ambivalence will choose information that reduces their ambivalence while those who are low in ambivalence may choose information that helps them arrive at an accurate decision. Studies 1 and 2 show that ambivalent individuals rate consumer reviews as more informative and helpful if the reviews are congruent rather than incongruent in valence with their existing dominant reactions during an attitude formation stage. Low ambivalence individuals rate the consumer reviews opposite of the high ambivalence individuals. Further, among different types of attitude congruent reviews, reviews that invalidate one’s conflicting reactions are projected to be of greater information value than reviews that add to one’s dominant reactions (Study 3). Controlled for possible covariates, conflict reduction reviews are shown to be more effective in accentuating attitude-behavior correspondence among attitude ambivalent individuals (Study 4).

Abstract:
In the customer expectations arena, relatively little attention has been paid to the impact on expectations of variation in cultural variables unique to a country. Here we focus on one country, India, and a major cultural influence there—the extent of belief in karma. Prior research in the U.S. suggests that disconfirmation sensitivity lowers expectations. Here we examine whether belief in karma, and consequently having a long term orientation, counteracts the tendency to lower expectations in two studies which measure and prime respondents’ belief in karma. Results show that the extent of belief in karma, operating largely through its impact on long run orientation, does moderate (decrease) the effect of disconfirmation sensitivity on expectations. These findings suggest that it is important to tailor advertising messages by matching them with customer expectations and their cultural determinants. Keywords: Consumer expectations, Culture, Disconfirmation sensitivity, Belief in karma

Abstract:

Abstract:
Abstract: We document changes in price sensitivity over time for 18 different grocery categories, and show that some brands benefit more from promoting during a recession, whereas others would prefer to promote during a boom. Specifically, using the new IRI Marketing Dataset, in each category we use household-level purchase data to estimate price elasticities by quarter over 6 years. We use a random coefficients logit model of utility where we estimate a different price coefficient in every quarter. We find that price sensitivity varies substantially over time and that allowing for quarterly price variation improves the fit substantially in all 18 categories. This variation appears related to the business cycle although we see some differences across categories. The correlation between GDP (growth) and price elasticity is most negative for categories with low repurchase rates in all periods. We then simulate the consequences of a price cut at the trough and near the peak of the business cycle. We find considerable heterogeneity in whether recessions are opportunities. For recessions to be opportunities, three conditions appear necessary: (1) price sensitivity must be relatively high for the price decrease to have an effect, (2) it must increase in the recession and (3) consumers must exhibit considerable state dependence.

Abstract:
The strategic importance of Business-to-Business (B2B) relationships is well recognized, but their financial impact remains unclear. This study links social capital from three types of B2B-networks of young technology firms with their Initial Public Offering (IPO) value. We identify three matching types of absorptive capacity that facilitate the transformation of B2B social capital into IPO value. The test of the hypotheses with secondary data shows that social capital from horizontal alliance networks does not increase IPO value, while customer-relationship social capital does. Importantly, the interaction between each type of B2B social capital and matching absorptive capacity exhibits a significant and positive impact, demonstrating that absorptive capacity is needed to transform B2B social capital into financial value. The results are robust under alternative measures and modeling approaches. This study provides empirical evidence of B2B social capital’s financial value, as well as the contingency factors that are manageable through marketing activities. As one of the first studies in marketing-finance interface that focus on young firms, our findings provide novel insights to both managers and investors.

Abstract:
Using data on individual level sales force performance over a 3 year period at a Fortune 500 firm, we propose and estimate a dynamic structural model of sales force response to a bonusbased compensation plan. We combine Arcidiacono and Miller's new EM algorithm approach within a two step conditional choice probability (CCP) estimator to allow for sales force response heterogeneity, while preserving the computational advantage of the CCP estimators. Further, in contrast to typical dynamic choice applications estimated using real world data, where discount factors cannot be non-parametrically identified, our bonus-based compensation structure enables us to identify discount factors in a hyperbolic discounting model. We find evidence of present bias consistent with hyperbolic discounting. Further, bonuses enhance sales productivity. Overachievement commission rates help maintain the productivity of the high performance sales force and quarterly bonuses serve as pacers to keep the sales force on track to achieve their annual sales quotas.

Abstract:
We suggest a general consumer preference model to overcome the limitations of consumer choice models especially for experience products when significant information on non-quantifiable attributes is missing. We develop a three-step approach to integrate the nonquantifiable information with consumer preference similarity via the concept of virtual experts: (i) identifying virtual experts by Bayesian clustering of consumers’ preference data; (ii) selecting a virtual expert for each target customer, using the similarity of the customer to the virtual experts; and (iii) modeling the utility of a product in terms of product attributes and the matched expert’s opinions (ratings) and confidences (variances). We applied our approach to two types of consumer data: (1) online consumer ratings (stated preferences) data for Internet recommendation services and (2) offline consumer viewership (revealed preferences) data for movies. Our model’s predictions are superior to those of the previous efforts. Our approach to missing attribute information for preference models links two different methodological streams: collaborative filtering and consumer preference models. Therefore, our model is applicable for both Internet recommendation services and consumer choice studies and enables firms to take full advantage of all different types of information in retailers’ customer databases for predictive improvement.

Abstract:
Third-party product reviews (TPRs) have become ubiquitous and increasingly important in many industries. This study examines whether and how these reviews may influence the financial value of firms introducing new products. Based on an event study of the movie industry, we find that professional reviews exert significant and immediate impact on stock returns in the direction of their valence. The impact comes, however, from the valence of a review that is measured relative to other, earlier published, reviews, and not from the absolute valence of the review itself. Moreover, the impact exists only for reviews published prior to product introduction but disappears when actual sales information becomes available. We further find that advertising and word-of-mouth can help to increase the positive impact or buffer the negative impact of TPRs on firm value, regardless of the review valence. These results demonstrate that the stock market’s assessment of the sales potential of new products is a continuously updating process in which third-party reviews play a significant role.

Abstract:
The rapid growth in multichannel shopping is prompting managers and researchers to better understand cross-channel effects, that is, the effects of marketing efforts in one channel on purchases made through the other channels. A study by Jupiter Research predicts that by 2011, nearly half of all business transactions will be influenced by the Internet channel. In this regard, managers are interested in better understanding the effects of different channels on a consumer as she moves through the different stages (e.g., awareness, consideration, purchase intent, and purchase) in decision making. To effectively allocate resources across channels and advertising, they also need sound knowledge of the effects of different channels in conjunction with those of advertising. Despite the importance of cross-channel effects in the presence of advertising effects, there is scant research on this issue. In this paper, we examine channel (own and cross) effects and advertising effects on the outcomes in the different stages of consumer decision making. We develop a set of proportional response models to capture the hierarchy of consumer decision making. To model own and cross channel effects, we develop a simultaneous system of consumer response models. We estimate these models using time series and cross-sectional data from the auto insurance industry, which comprises the exclusive agent channel, the independent agent channel, the Web and the call center. Our results offer interesting insights. They show that cross-channel effects are significant. The exclusive agent channel and the Web are often synergistic. The exclusive agent and call center channels are complementary, while the independent agent and exclusive agent channels are competitive. Advertising has significant current effects on insurance quotes, qualified quotes and policies and significant lagged effects on brand awareness, insurance quotes, qualified quotes, and policies. Overall, advertising elasticity for insurance quotes is higher than the different channel elasticities, but this finding is reversed for insurance quotes within each channel. The findings offer important implications for marketers in planning their advertising and channel efforts.

Abstract:
As the online communities have become a commonplace to nd user-generated information on virtually any topic of interest on the Internet, user-generated content (UGC) sites have seen significant growth over the past several years. The main purposes of online users visiting UGC sites are to generate and access various types of media content. As a result, the behavioral event counts of generating and accessing UGC by online users serve as important metrics to judge the success of online communities. In this research, we develop a joint model to understand and describe the inherent behaviors and interactions of the members over time through the medium of UGC in online community. To this end, we propose a bivariate zero-inflated Poisson model to simultaneously model the daily counts of the two UGC-related activities that explicitly capture the critical features of member behaviors in online community. In particular, we consider interdependences of the repeatedly measured behavioral events within members, model the dependence of the current event counts on the past event counts, and explore the probable nonlinear effects of the time at the individual level. Furthermore, we incorporate interactions among members by constructing a set of individual-specific time-varying measures in an integrated modeling framework. In our empirical analysis, we use a comprehensive database from an online community provided by one of the largest Internet portal sites in Korea, and demonstrate that our general model captures the key behavioral aspects of the member behaviors in the UGC site. Some of our primary findings are as follows: (1) UGC-generating and accessing activities are signicantly correlated with each other within individual members; (2) The community managers play an important role in building and sustaining the community via content generation; (3) A member in the community is more likely to participate in the content-generation process as other members more actively respond to her content, and a member tends to access more UGC as others generate more UGC; (4) The elapsed time since a member's last visit has the nonlinear eects in the activeness and event rate of UGC-generating and consuming behaviors. As part of our substantive contribution, we highlight the model's ability to make accurate statements about the important metrics of the online community (e.g., the daily number of active members to the community).

Abstract:
We propose and empirically test a theory to explain the volume and type of consumer word of mouth for products. Our theory builds on Chitturi et al’s (2008) theoretical framework and includes innovativeness as a stimulant of word of mouth. The theory leads to four sets of hypotheses. The first two are regarding the effects of utilitarian and hedonic attributes and the third is related to the role of innovativeness. Specifically, the hypotheses related to the role of utilitarian attributes are that satisfying consumers on such attributes does not generate positive word of mouth while not pleasing them on those attributes results in negative word of mouth. The hypotheses related to hedonic attributes state that pleasing consumers on those attributes leads to positive word of mouth but that not doing so does not stimulate negative word of mouth. The third set of hypotheses are that innovativeness increases the volume of word of mouth but that it could increase both positive and negative word of mouth simultaneously. The fourth set of hypotheses is related to the effects of ignoring the role of innovativeness in analyzing consumer word of mouth. We expect this to result in two types of effects: inflated estimates of the effects of utilitarian and hedonic attributes on word of mouth and incorrect signs for the effects. We test the theory empirically on consumer word of mouth over a seven year period for 279 models of automobiles across 36 brands. All four sets of hypotheses are supported by our empirical results and suggest that utilitarian and hedonic attributes play very different roles in consumer word of mouth. Specifically, utilitarian attributes are the primary determinants of negative word of mouth in that, if products are not satisfactory on those attributes, consumers are likely to express their unhappiness to other consumers. Satisfying consumers on these attributes, however, does not necessarily lead to positive word of mouth. Hedonic attributes, on the other hand, are primarily responsible for positive word of mouth in that, if they are pleased with the performance of products on those attributes, consumers are likely to discuss the positive aspects of the products with other consumers. Not pleasing consumers on these attributes, on the other hand, is unlikely to stimulate negative word of mouth. Product innovativeness plays a role that is different from that of product attributes. Specifically, innovativeness can lead to both negative and positive word of mouth. Additionally, because of this dual effect, omitting innovativeness from an analysis of the role of product attributes in consumer word of mouth can lead to inflated and incorrect estimates of their effects. We also present the managerial implications of our findings.

Abstract:
Matching contributions are increasingly being used by charities and gift campaigns in attracting donations from potential donors. Extant research has been ambiguous about the effect of matching contribiutions on compliance. We show that matching contributions increase compliance and that this effect is moderated by having a threshold for the matching contributions. We examine the mediating role of donation impact or dollar efficacy and show that it explains best the role of matching contributions on compliance.

Abstract:
This research investigates how consumers learn to use a product feature and the impact that the learning process has on consumer acceptance of new products. Skill learning has been assumed to follow a power law trend, with the largest gains occurring early in the product’s usage followed by monotonically decreasing performance improvements over trials. This paper proposes an alternative process in which skill acquisition of individual consumers exhibits a discrete step function which corresponds with a moment of insight. The learning process also has consequences such as performance, perceived ease of use and usage intentions which are explored in a series of lab experiments.