Sales Impact Research Articles

Evaluating the E ffects of Having Friends on Sales with Observational Data

We use a two-phase approach to examine the causal sales impact of having friends with observational data. The first phase uses propensity score matching to address sample selection endogeneity. The second phase leverages principal stratification to address the issue that sales are zero to both inactive customers and active customers with no value, and that customer value is only well de ned for active customers. We apply our approach to data from a massive multiplayer online game, where the sale is measured by cash spending and customer value equals to the sale if a customer logs in. Our causal estimands are: (1) the average causal effects on the probability of logging in, and the probability and amount of cash spending over all customers; (2) the average causal effects on the probability and amount of cash spending for customers who would log in regardless of whether having friends or not. We conduct a likelihood-based analysis using mixtures of two-part models, discuss identification issue, investigate the plausibility of meaningful restrictions, and conduct some sensitivity analyses. Our results demonstrate that the two-phase approach can make more detailed causal inference and have better predictive performance than the one-phase approach that uses propensity score matching only.

Cause Marketing Effectiveness and the Moderating Role of Price Discounts

Can cause marketing (CM) be effective? If so, do price discounts moderate CM effectiveness? Despite the prevalence of linking product sales with donations to charity, field evidence of CM effectiveness is lacking. This is of particular concern for managers who wonder whether the findings of laboratory experiments extend to actual consumer purchases. Using large-scale randomized field experiments with more than 17,000 consumers, this research documents that CM can significantly increase consumer purchases. Notably, the answer to the second question is more complicated. Under the moderating role of price discounts, the impact of CM on sales purchases may follow an inverted U-shaped relationship—that is, strongest when price discounts are moderate rather than deep or absent. Follow-up lab experiments reveal that consumers' warm-glow good feelings from CM represent the underlying process. These findings provide novel insights into the boundary conditions and mechanisms of the sales impact of CM for researchers and managers alike.

Sales impact of displaying alcoholic and non-alcoholic beverages in end-of-aisle locations: An observational study

In-store product placement is perceived to be a factor underpinning impulsive food purchasing but empirical evidence is limited. In this study we present the first in-depth estimate of the effect of end-of-aisle display on sales, focussing on alcohol. Data on store layout and product-level sales during 2010–11 were obtained for one UK grocery store, comprising detailed information on shelf space, price, price promotion and weekly sales volume in three alcohol categories (beer, wine, spirits) and three non-alcohol categories (carbonated drinks, coffee, tea). Multiple regression techniques were used to estimate the effect of end-of-aisle display on sales, controlling for price, price promotion, and the number of display locations for each product. End-of-aisle display increased sales volumes in all three alcohol categories: by 23.2% (p = 0.005) for beer, 33.6% (p < 0.001) for wine, and 46.1% (p < 0.001) for spirits, and for three non-alcohol beverage categories: by 51.7% (p < 0.001) for carbonated drinks, 73.5% (p < 0.001) for coffee, and 113.8% (p < 0.001) for tea. The effect size was equivalent to a decrease in price of between 4% and 9% per volume for alcohol categories, and a decrease in price of between 22% and 62% per volume for non-alcohol categories. End-of-aisle displays appear to have a large impact on sales of alcohol and non-alcoholic beverages. Restricting the use of aisle ends for alcohol and other less healthy products might be a promising option to encourage healthier in-store purchases, without affecting availability or cost of products.

Hour-by-Hour Sales Impact of Mobile Advertising

Mobile technology offers a unique opportunity to reach consumers at the time and place they are most receptive to messages. Yet, marketers are uncertain about which hour of day mobile advertising is most effective. To answer this and identify the causal impact of advertising on sales, our study builds on two large-scale randomized field experiments with 19,200 mobile users. We identify that consumer purchase response varies on an hour-to-hour basis throughout the day and that these hourly effects interact with product type and framing. The first field experiment reveals that mobile ads for utilitarian products are most effective during morning hours, peaking between 10:00 am and noon. In contrast, ads for hedonic products are most effective during the afternoon, peaking between 2:00 and 4:00 pm. Our second field experiment replicates this finding with a single product but different ad copy framing manipulations. Ads with utilitarian framing are found to decrease the odds of consumer purchases on mobile devices by 80%, holding all else constant. Compared with naive industry approaches, optimally framed mobile ad copies can boost purchase rates by 71.8%. These findings not only support the hour-by-hour mobile ad effects, but also make a novel contribution by demonstrating how framing a product as either utilitarian or hedonic generates substantial differences in mobile ad effectiveness. Field experiments can eliminate the confounds of consumer heterogeneity and endogenous biases when identifying the causal but subtle advertising effects in real-world purchases. Marketers should optimize hours of the day, ad copy framing, and consumer segments to enhance mobile ad effectiveness in the field.

Usage Rates, Facebook Likes, and Online Piracy: Using Big Data to Manage Entertainment Products

The rapid digitization of the entertainment industry urges managers to find new ways to improve marketing effectiveness and combat online piracy. Smart devices and innovative online services generate brand-related tracking data that reflect consumers’ brand engagement (e.g., digital usage rates, Facebook Likes), yet managers are unaware of how to use these big data to improve brand performance. We therefore propose a model to assess how digitally tracked engagement behaviors, namely consumer brand usage (CBU, measured by digital usage rates) and consumer brand endorsement (CBE, measured by Facebook Likes), drive the impact of promotion and new product releases and affect the role of piracy. We apply our framework to the U.S. music industry, and develop web scrapers and API mining scripts to collect data for 569 music artists over 66 weeks. Both CBU and CBE enhance marketing effectiveness and positively mediate the sales impact of piracy, thereby offsetting piracy’s cannibalization effect by 55%. However, CBU exerts a systematically stronger influence in these relationships than CBE. Using simulations, we show how our findings can help managers to improve the allocation of their marketing efforts across brands and reduce the detrimental sales impact of piracy.

Quantifying the Dynamic Sales Impact of Location-Based Mobile Promotion Technologies

Location-based mobile promotion (LMP) is believed to trigger spontaneous and impulsive purchases. This field study quantifies the dynamic impact of LMP, using aggregated daily mobile movie promotions and purchase data from over 3 million realworld users of a large mobile service provider. The developed multivariate time-series models confirm that LMP has a significant contemporaneous impact on movie ticket sales. Bayesian vector autoregressive results also confirm that LMP has significant immediate (next day) and cumulative impacts on sales over the next 9 days. The immediate and cumulative impacts account for 66% of the total effects. These findings suggest that without considering the dynamic effects, practitioners and researchers may substantially under-estimate the value of LMP. Our analysis also reveals the presence of cross-channel effects between LMP and other forms of mobile related promotions.

Identifying Metrics That Matter: What Are the Real Key Performance Indicators (KPIs) That Drive Consumer Behavior?

Abstract Vector auto regression (VAR) is a form of econometric modeling that is receiving increased attention in marketing research applications. It is used to observe whether potentially relevant indicators have a real impact on sales or success factors. Compared with correlation, regression and conjoint techniques, VAR models are superior because they are able to show the impact of changes over time on the basis of real business data. The research shows how VAR models are applied in different marketing settings. VAR models can filter relevant metrics from a whole set of potentially relevant performance indicators and quantify the sales impact of each variable. They further observe lead and lag effects that cannot be tracked when measurement is conducted at a single point in time. Modeling can be performed on competitive brands as well. VAR models also make it possible to test whether the same success factors that drive a category also drive the sales of each of the brands in that category.

How to Design Mobile Advertising: A Field Experiment with Wireless Telecommunication Technology

Although mobile advertising seemingly offers practitioners tremendous potential given the ubiquitous nature of reaching subscribers anywhere, we know very little about it. Base on field experiment data, the developed zero-inflated Poisson model reveals that distance, promotion and product type could affect sales impact of advertising while using wireless telecommunication technology. When estimating the model, we use latent instrument variable (LIV) approach to rule out possible endogeneity problem. The empirical results indicate that distance, promotion, product type trigger sales respectively, while only promotion increases sales amount.

Optimizing Advertising Using Wireless Communication Technology: A Zero-Inflated Poisson Approach

Recently, more and more companies are using wireless location technology to improve advertising effectiveness. Although location-targeted advertising seemingly offers practitioners tremendous potential given the ubiquitous nature of mobile devices, we know very little about it. Base on field experiment data, the developed zero-inflated Poisson model reveals that distance, discount and CSR(Corporate Social Responsibility) could affect sales impact of advertising while using wireless communication technology. These results have implications for mobile marketing research and practice while using wireless communication technology.

Does Adding Inventory Increase Sales? Evidence of a Scarcity Effect in U.S. Automobile Dealerships

What is the relationship between inventory and sales? Clearly, inventory could increase sales: expanding inventory creates more choice (options, colors, etc.) and might signal a popular/desirable product. Or, inventory might encourage a consumer to continue her search (e.g., on the theory that she can return if nothing better is found), thereby decreasing sales (a scarcity effect). We seek to identify these effects in U.S. automobile sales. Our primary research challenge is the endogenous relationship between inventory and demand — e.g., dealers influence their inventory in anticipation of demand. Hence, our estimation relies on weather shocks at upstream production facilities to create exogenous variation in downstream dealership inventory. We find that the impact of adding a vehicle of a particular model to a dealer's lot depends on which cars the dealer already has. If the added vehicle expands the available set of sub-models (e.g., adding a four-door among a set that is exclusively two-door), then sales increase. But if the added vehicle is of the same sub-model as an existing vehicle, then sales actually decrease. Hence, expanding variety should be the first priority when adding inventory — adding inventory without expanding variety is actually detrimental. Based on this insight, given a fixed set of cars, vehicles should be allocated among a group of dealers so as to each dealer's variety. Our data indicate that the implementation of this could increase expected sales by about 2.5% without changing the total number of vehicles in the market, which vehicles are produced or the number of vehicles at each dealership. If the firm is willing to reduce aggregate inventory by a modest 2.9% (and no more than 10% at any one dealer), then the sales impact of the maximize variety, minimize duplication strategy doubles, to 5.0%.

Is the Multi-Platform Whole More Powerful Than Its Separate Parts?

<h3>ABSTRACT</h3> Cross-media campaigns are becoming a norm, yet there is a lack of knowledge on how they impact sales. This paper documents the sales response to cross-media campaigns and finds that, when online advertising is added to a television campaign, the extra reach achieved is primarily duplicated. Regularly a single television exposure stimulates sales among those exposed, with online advertising demonstrating a similar yet less consistent response. We do not find evidence of a synergy in sales impact, where the sum effect of exposure to both television and online is greater than the parts. We highlight challenges with such single-source research.

Services vs. Manufacturing – How Does Foreign and Domestic Sales Impact on Their R&amp;D?

While the distinction between manufacturing and services becomes increasingly blurred to some observers, we find, using a panel of Swedish firms, clear evidence that foreign sales (exports) are more important than domestic sales for stimulating R&D. This is particularly clear for manufacturing and this importance of foreign sales has increased over time, simultaneous to an opening up of the Swedish economy. Even though service industries have seen an increase in both R&D and trade over time, it is thus mainly manufacturing that has benefited from increased possibilities for absorptive capacity. This result suggests a clear dichotomy between manufacturing and services in terms of how they react to trade and how they turn towards the foreign market vs. the domestic market to find stimuli for innovation.

Measuring Dynamic Sales Impacts of LBA Using Wireless Communication Technology

Because of rapid development of wireless communication technology, there is an increasing adoption of mobile advertising, such as location based advertising (LBA). To what extent can LBA improve advertising effectiveness is an important topic in the field of wireless communication technology research. Most researches quantify long term impacts of advertisings by VAR (Vector Autoregressive) model. However, compared to VAR model, VECM (Vector Error Correction Model) is a better method in that it allows one to estimate both a long-term equilibrium relationship and a short-term dynamic error correction process. In this study, we employ VECM to explore LBA’s (Location Based Advertising) and PUA’s (Pop-up Advertising) sales impact in both short and long terms. The developed VECM reveals that LBA’s sales impact is about more than2 times as big as PUA’s in short dynamic term and nearly 6 times bigger than PUA’s in long equilibrium term. These findings add to advertising and VECM literatures. These results can give managers more confident to apply wireless communication technology to advertising.

Quantify sales impact of location-based advertising

Compared with the previous experimental or survey research studying attitudinal responses, the purpose of this study is to quantify sales impact of location–based advertising (LBA) with company archival data. The data are daily movie ticket sales via mobile phone applications, based on more than 20,000 movie fans from one of the world's biggest mobile companies. The developed dynamic structural equation model (DSEM) could estimate both contemporaneous and long–term sales impacts. Empirical findings reveal that: 1) in terms of magnitude, LBA's efficacy is stronger than that of pop–up advertising (PUA) both contemporaneously and accumulatively; 2) in terms of persistent time, while LBA only have contemporaneous impact, PUA's impact lasts for nine days. These findings would help managers to allocate resources more efficiently between advertising channels, and represent an early step of understanding LBA's effectiveness in generating sales.

Sales Impacts of Location Based Advertising Using Wireless Communication Technology

Using wireless communication technology, location-based advertising (LBA) is booming recently. Most LBA researches are cross-sectional analysis studying only contemporaneous impacts, which might seriously underestimate advertisings’ efficacy in the long run. Complementary to these researches, we employ a company archival dataset to explore LBA’s and PUA’s (Pop-up Advertising) sales impact in both contemporaneous and long run terms. The developed dynamic structural equation model reveals that (1)in terms of magnitude, LBA’s efficacy is stronger than that of PUA both contemporaneously and accumulatively; (2) in terms of persistent time, while LBA only have contemporaneous impact, PUA’ s impact lasts for 9 days. Theses empirical findings would help managers to allocate advertising budget efficiently while using wireless communication technology.

The Effects of Traditional and Social Earned Media on Sales: A Study of a Microlending Marketplace

Marketers distinguish three types of media: paid (e.g., advertising), owned (e.g., company website), and earned (e.g., publicity). The effects of paid media on sales have been extensively covered in the marketing literature. The effects of earned media, however, have received limited attention. The authors examine how two types of earned media, traditional (e.g., publicity and press mentions) and social (e.g., blog and online community posts), affect sales and activity in each other. They analyze 14 months of daily sales and media activity data from a microlending marketplace website using a multivariate autoregressive time-series model. They find that (1) both traditional and social earned media affect sales; (2) the per-event sales impact of traditional earned media activity is larger than for social earned media; (3) because of the greater frequency of social earned media activity, after adjusting for event frequency, social earned media's sales elasticity is significantly greater than traditional earned media's; and (4) social earned media appears to play an important role in driving traditional earned media activity.

The Effects of Traditional and Social Earned Media on Sales: A Study of a Microlending Marketplace

Marketers distinguish between three types of media: paid (e.g., advertising), owned (e.g., company website), and earned (e.g., publicity). The effects of paid media on sales have been extensively covered in the marketing literature. The effects of earned media, however, have received limited attention. This paper examines how two types of earned media, traditional (e.g., publicity and press mentions) and social (e.g., blog and online community posts), affect sales and activity in each other. Fourteen months of daily sales and media activity data from a microlending marketplace website are analyzed using a multivariate autoregressive time series model. The authors find that (i) both traditional and social earned media affect sales, (ii) the per-event sales impact of traditional earned media activity is larger than for social earned media, (iii) however, because of the greater frequency of social earned media activity, after adjusting for event frequency social earned media’s sales elasticity is significantly greater than traditional earned media’s, and (iv) social earned media appears to play an important role in driving traditional earned media activity.

The Value of Social Dynamics in Online Product Ratings Forums

Research has shown that consumer online product ratings reflect both the customers' experience with the product and the influence of others' ratings. In this article, the authors measure the impact of social dynamics in the ratings environment on both subsequent rating behavior and product sales. First, they model the arrival of product ratings and separate the effects of social influences from the underlying (or baseline) ratings behavior. Second, the authors model product sales as a function of posted product ratings while decomposing ratings into a baseline rating, the contribution of social influence, and idiosyncratic error. This enables them to quantify the sales impact of observed social dynamics. The authors consider both the direct effects on sales and the indirect effects that result from the influence of dynamics on future ratings (and thus future sales). The results show that although ratings behavior is significantly influenced by previously posted ratings and can directly improve sales, the effects are relatively short lived once indirect effects are considered.

Online Social Interactions: A Natural Experiment on Word of Mouth versus Observational Learning

Consumers’ purchase decisions can be influenced by others’ opinions, or word of mouth (WOM), and/or others’ actions, or observational learning (OL). Although information technologies are creating increasing opportunities for firms to facilitate and manage these two types of social interaction, to date, researchers have encountered difficulty in disentangling their competing effects and have provided limited insights into how these two social influences might differ from and interact with each other. Using a unique natural experimental setting resulting from information policy shifts at the online seller Amazon.com , the authors design three longitudinal, quasi-experimental field studies to examine three issues regarding the two types of social interaction: (1) their differential impact on product sales, (2) their lifetime effects, and (3) their interaction effects. An intriguing finding is that while negative WOM is more influential than positive WOM, positive OL information significantly increases sales, but negative OL information has no effect. This suggests that reporting consumer purchase statistics can help mass-market products without hurting niche products. The results also reveal that the sales impact of OL increases with WOM volume.

Online Social Interactions: A Natural Experiment on Word of Mouth Versus Observational Learning

Consumers’ purchase decisions can be influenced by others’ opinions, i.e., word-of-mouth (WOM), and/or others’ actions, i.e., observational learning. While information technologies are creating increasing opportunities for firms to facilitate/manage these two types of social interaction, researchers so far have encountered difficulty in disentangling their competing effects and have provided limited insights into how these two social influences may differ from and interact with each other. Based on a unique natural experimental setting resulting from information policy shifts at the online seller Amazon.com, we design three longitudinal, quasi-experimental field studies to examine three issues regarding the two types of social interaction: (1) their differential impact on product sales, (2) their lifetime effects, and (3) their interaction effects. An intriguing finding is that, while negative WOM is more influential than positive WOM, positive observational learning information significantly increases sales but negative observational learning information has no effect. This suggests that reporting consumer purchase statistics can help mass-market products without hurting niche products. Our results also reveal that the sales impact of observational learning increases with WOM volume.