Ayelet Israeli is an associate professor of business administration in the Marketing Unit. She teaches the E-Commerce course in the MBA elective curriculum, the Data Driven Marketing course in the Harvard Business Analytics Program, and in various Executive Education programs. She has previously taught the core Marketing course in the first year MBA required curriculum.
In her research, Professor Israeli studies omni-channel and e-commerce markets. In particular, she focuses on pricing and pricing policies, channel management, and online marketing. Her research has been published in Marketing Science andJournal of Marketing Research. Her dissertation won the 2014 INFORMS Society for Marketing Science Doctoral Dissertation Proposal Award, and she was named a finalist for the 2018 and the 2019 Frank M. Bass Award. Her work has been cited by The Wall Street Journal, The Atlantic, MSN Money, and Harvard Business Review.
Professor Israeli received her PhD in marketing from the Kellogg School of Management at Northwestern University. She holds an MBA from the Hebrew University of Jerusalem, where she also earned her MSc and BSc in computer science. In addition to her academic experience, Professor Israeli served as a lieutenant in the Intelligence Corps of the Israeli Defense Forces and worked as an engineer at Israel Aerospace Industries and at Intel Corporation in Israel.
When customers seek out online deals, it seems like a win for everybody: Brands, retailers, dealers, and distributors sell more goods, and buyers get a bargain. What's not to like? Here's the problem: Lured by rock-bottom online prices, customers often end up dealing with disreputable digital resellers that mistreat them, which can erode brand equity and undermine authorized sellers. For brands selling in the United States and Canada, there is a potent countermeasure: crafting and enforcing pricing policies that discourage anyone from advertising or selling a firm's products at an unauthorized discount. This article lays out the four steps essential to designing and enforcing policies with teeth that stay on the right side of antitrust law.
This paper investigates a manufacturer’s ability to influence compliance rates among its authorized online retailers by exploiting changes in the Minimum Advertised Price (MAP) policy and in dealer agreements. MAP is a pricing policy widely used by manufacturers to influence prices set by their downstream partners. A MAP policy imposes a lower bound on advertised prices, subjecting violating retailers to punishments such as termination of distribution agreements. Despite this threat, violations are common. I uncover two key elements to improve compliance: customization to the online environment and credible monitoring and punishments. I analyze the pricing, enforcement, and channel management policies of a manufacturer over several years. During this period, new channel policies take effect, providing a quasi-experiment. The new policies lead to substantially fewer violations. With improved compliance, channel prices increase by 2% without loss in volume. The reduction in violations is particularly stark among authorized retailers with lower sales volume, those that previously operated unapproved websites and those that have received violation notifications for the specific product before. Moreover, low service providers improve their service. At the same time, there is an increase in opportunistic behavior among top retailers, retailers that received notifications for other products, and for less popular products via deep discounting.
In this paper we investigate whether sellers treat consumers differently on the basis of how well informed consumers appear to be. We implement a large-scale field experiment in which callers request price quotes from automotive repair shops. We show that sellers alter their initial price quotes depending on whether consumers appear to be correctly informed, uninformed, or misinformed about market prices. We find that repair shops quote higher prices to callers who cite a higher benchmark price. We find that women are quoted higher prices than men when callers signal that they are uninformed about market prices. However, gender differences disappear when callers mention a benchmark price for the repair. Finally, we find that repair shops are more likely to offer a price concession if asked to do so by a woman than a man.
Manufacturers in many industries frequently use vertical price policies, such as minimum advertised price (MAP), to influence prices set by downstream retailers. Although manufacturers expect retail partners to comply with MAP policies, violations of MAP are common in practice. In this research, we document and explain both the extent and the depth of MAP policy violations. We also shed light on how retailers vary in their propensity to violate MAP policies and the depth by which they do so. Our inductive research approach documents managerial wisdom about MAP practices. We confront these insights from practice with a large empirical study that includes hundreds of products sold through hundreds of retailers. Consistent with managerial wisdom, we find that authorized retailers are more likely to comply with MAP than are unauthorized partners. By contrast to managerial wisdom, we find that authorized and unauthorized markets are largely separate, and that violations in the authorized channel have a small association with violations in the unauthorized channel. Last, we link our results to the literatures on agency theory, transaction cost analysis, and theories of price obfuscation.
Spreadsheet Supplement to Artea Teaching Note 521-041. This collection of exercises aims to teach students about 1)Targeting Policies; and 2)Algorithmic bias in marketing—implications, causes, and possible solutions. Part (A) focuses on A/B testing analysis and targeting. Parts (B),(C),(D) Introduce algorithmic bias. The exercises are designed such that the issues of algorithmic bias and discrimination would emerge inductively, “surprising” the students in the act of recommending a strategy that, inadvertently, is discriminating against customers who belong to minority groups. This is achieved via the combination of hands-on exercises, where students would make decisions based on data analyses and visualization, and in-class discussions, where students would defend their proposed strategies, discover the (discriminating) implications of those actions, and discuss possible solutions.
Ascarza, Eva, and Ayelet Israeli. "Spreadsheet Supplement to Artea Teaching Note." Harvard Business School Spreadsheet Supplement 521-705, September 2020.
View Details
Teaching Note for HBS No. 521-021,521-022,521-037,521-043. This collection of exercises aims to teach students about 1)Targeting Policies; and 2)Algorithmic bias in marketing—implications, causes, and possible solutions. Part (A) focuses on A/B testing analysis and targeting. Parts (B),(C),(D) Introduce algorithmic bias. The exercises are designed such that the issues of algorithmic bias and discrimination would emerge inductively, “surprising” the students in the act of recommending a strategy that, inadvertently, is discriminating against customers who belong to minority groups. This is achieved via the combination of hands-on exercises, where students would make decisions based on data analyses and visualization, and in-class discussions, where students would defend their proposed strategies, discover the (discriminating) implications of those actions, and discuss possible solutions.
Ascarza, Eva, and Ayelet Israeli. "Artea (A), (B), (C), and (D): Designing Targeting Strategies." Harvard Business School Teaching Note 521-041, September 2020.
View Details
This collection of exercises aims to teach students about 1)Targeting Policies; and 2)Algorithmic bias in marketing—implications, causes, and possible solutions. Part (A) focuses on A/B testing analysis and targeting. Parts (B),(C),(D) Introduce algorithmic bias. The exercises are designed such that the issues of algorithmic bias and discrimination would emerge inductively, “surprising” the students in the act of recommending a strategy that, inadvertently, is discriminating against customers who belong to minority groups. This is achieved via the combination of hands-on exercises, where students would make decisions based on data analyses and visualization, and in-class discussions, where students would defend their proposed strategies, discover the (discriminating) implications of those actions, and discuss possible solutions.
Spreadsheet Supplement to "Artea (B): Including Customer-level Demographic Data" and "Artea (C): Potential Discrimination through Algorithmic Targeting"
Ascarza, Eva, and Ayelet Israeli. "Spreadsheet Supplement to Artea (B) and (C)." Harvard Business School Spreadsheet Supplement 521-704, September 2020.
View Details
This collection of exercises aims to teach students about 1)Targeting Policies; and 2)Algorithmic bias in marketing—implications, causes, and possible solutions. Part (A) focuses on A/B testing analysis and targeting. Parts (B),(C),(D) Introduce algorithmic bias. The exercises are designed such that the issues of algorithmic bias and discrimination would emerge inductively, “surprising” the students in the act of recommending a strategy that, inadvertently, is discriminating against customers who belong to minority groups. This is achieved via the combination of hands-on exercises, where students would make decisions based on data analyses and visualization, and in-class discussions, where students would defend their proposed strategies, discover the (discriminating) implications of those actions, and discuss possible solutions.
This collection of exercises aims to teach students about 1)Targeting Policies; and 2)Algorithmic bias in marketing—implications, causes, and possible solutions. Part (A) focuses on A/B testing analysis and targeting. Parts (B),(C),(D) Introduce algorithmic bias. The exercises are designed such that the issues of algorithmic bias and discrimination would emerge inductively, “surprising” the students in the act of recommending a strategy that, inadvertently, is discriminating against customers who belong to minority groups. This is achieved via the combination of hands-on exercises, where students would make decisions based on data analyses and visualization, and in-class discussions, where students would defend their proposed strategies, discover the (discriminating) implications of those actions, and discuss possible solutions.
Ascarza, Eva, and Ayelet Israeli. Spreadsheet Supplement to "Artea: Designing Targeting Strategies". Harvard Business School Spreadsheet Supplement 521-703, September 2020.
View Details
This collection of exercises aims to teach students about 1)Targeting Policies; and 2)Algorithmic bias in marketing—implications, causes, and possible solutions. Part (A) focuses on A/B testing analysis and targeting. Parts (B),(C),(D) Introduce algorithmic bias. The exercises are designed such that the issues of algorithmic bias and discrimination would emerge inductively, “surprising” the students in the act of recommending a strategy that, inadvertently, is discriminating against customers who belong to minority groups. This is achieved via the combination of hands-on exercises, where students would make decisions based on data analyses and visualization, and in-class discussions, where students would defend their proposed strategies, discover the (discriminating) implications of those actions, and discuss possible solutions.
Teaching Note for HBS No. 521-020. This note focuses on algorithmic bias in marketing. First, it presents a variety of marketing examples in which algorithmic bias may occur. The examples are organized around the 4 P’s of marketing – promotion, price, place and product—characterizing the marketing decision that generates the bias and highlighting the consequences of such a bias. Then, it explains the potential causes of algorithmic bias and offers some solutions to mitigate or reduce this bias.
This note focuses on algorithmic bias in marketing. First, it presents a variety of marketing examples in which algorithmic bias may occur. The examples are organized around the 4 P’s of marketing – promotion, price, place and product—characterizing the marketing decision that generates the bias and highlighting the consequences of such a bias. Then, it explains the potential causes of algorithmic bias and offers some solutions to mitigate or reduce this bias.
Thingtesting, a brand discovery and testing digital community devoted to uncovering and exploring direct-to-consumer brands, had just received seed funding and was contemplating a second year of growth. The new year brought many challenges, as founder Jenny Gyllander had to decide how to scale her fledgling company. As the company matured, she had to decide what role she would play going forward: should she continue to operate as head curator and content developer of her own reviews or open up her platform to invite and crowdsource the reviews of her followers. Her value proposition positioned her at the nexus of several different audiences: millennial consumers eager to try new brands, venture capitalists eager to invest in them, and the companies and brands themselves, who were interested in seeing their new products succeed. What was the best path to monetizing the business? As Gyllander navigates the emerging landscape of influencer marketing, she needs to decide how to transform her online audience into a co-creating community and how to best wield her own influence to create value for her followers and other constituencies in a way that could fund a profitable business model.
Despite a heavy barrage of advertising, most consumers declare that their purchases are most influenced by the experiences, advice, and recommendations of others, and not by marketers. Interpersonal communication between and among consumers serves as a potent path for influence, and by 2022, marketing managers are expected to spend $15 billion on influencer marketing, a marketing technique in which companies partner with people with specialized knowledge, expertise, authority, social position, and/or personal relationships that enable them to have influence over others to co-produce marketing messages to promote their brands via offline and electronic word-of-mouth. This technical note provides an overview of influencer marketing and discusses the various roles that influencers play for marketers, including delivering marketing messages in a more authentic voice and setting than that offered by more traditional advertising media where messages are authored by firms. It categorizes different types of influencers, explains the social psychological processes that make interpersonal influence work, and provides guidelines on how to identify people with influence. It discusses the opportunities and challenges related to measuring the performance of influencers and assessing the costs and return-on-investment of influencer marketing programs. Finally, it highlights emerging risks in the contemporary influencer marketing landscape.
This case provides a post-mortem of the advertising technology (adtech) company Sizmek. Sizmek grew via multiple acquisitions, with the vision of becoming an integrated adtech company that could leverage AI to buy digital media, while creating and serving display and video content. At its peak, one of its acquisitions had an IPO valuation of almost 1 billion dollars. However, Sizmek struggled with an integration that was taking longer than expected, but also with misalignment of agency incentives, stricter data privacy regulations, and the rise of the walled gardens – Google, Facebook, and Amazon – that made the adtech ecosystem very competitive, with dynamics of winners-take-all markets. By 2019 Sizmek had to file Chapter 11, and was sold in pieces. Is there anything Sizmek could have done differently in their go-to-market strategy, product road-map, or overall partnerships to avoid such rapid collapse?
In mid-2019, Carlos Hank was deliberating over the results for Banorte Móvil—the mobile application for Banorte, Mexico’s most profitable and second-largest financial institution. Hank, who had been appointed as Banorte´s Chairman of the Board in January 2015, had overseen Banorte’s transformation (and multi-million-dollar investment) from a product and client volume-focused bank into a customer-centric, technology and data-driven organization with a radically new focus. Beyond investing in a new technology platform and deploying new digital channels, Banorte also invested in the development of internal capacities to convert data intelligence into profits. Spearheaded by José Antonio Murillo, an economist with a Ph.D. from Rice University, Banorte’s Analytics Business Unit (ABU) kicked off early on 2015, reporting directly to upper management and focused on increasing customer lifetime value. Over the course of its four-year trajectory, through data analytics and experimentation, which involved both experimenting with client incentives and tailored communication strategies through multiple channels, the ABU had successfully achieved higher product placement rates, particularly under Banorte Móvil’s platform. Yet, with 2.25 million active users, Banorte Móvil was still far from reaching its 4-million-user target by 2020. Adoption picked up, but Banorte Móvil was still losing many potential adopters along the mobile customer journey, particularly in its activation phase. Furthermore, 81% of app activity was for account balance or transaction views, with financial operations, such as card or service payments or acquisition of new bank products, accounting for a minority of the activity. If Banorte wanted to remain a top player in Mexico’s financial sector, it was clear to upper management that growth in mobile banking needed to be a priority. How could the bank successfully achieve its target? And, even if Banorte Móvil’s adoption numbers increased, would it be possible to get more value and engagement through this channel? As Banorte Móvil took off, Hank dwelled on the strategic decisions he had to make, especially considering what client segments to focus on for mobile adoption and use and how to effectively reach out to them. Furthermore, Hank knew they could not afford to neglect other banking channels, which still accounted for the bulk of Banorte’s operations.
When the mother-daughter founders of DivaCup set out with a mission to disrupt the menstrual care industry with an innovative product form, they initially struggled to gain legitimacy and convince retailers to carry their unique product. Fifteen years later, the product was available online and in more than 35,000 retail stores worldwide, and the company was struggling with omnichannel distribution challenges and frequent violations of their pricing policy due to intense competition among channel partners. International distribution and the prevalence of gray market sellers were causing large price dispersions across markets, which was harming the brand further. In October 2018, the entry of a formidable competitor, P&G (owner of menstrual care brands Tampax and Always), into the niche product segment causes the CEO and her team to consider the best plan to reposition and differentiate its product to broaden its appeal to users of other product forms, while maintaining its market leadership in the menstrual cup segment. This challenge is made more difficult due to prevalent miseducation about menstrual care and menstrual cups in particular.
DayTwo is a young Israeli startup that applies research on the gut microbiome and machine learning algorithms to deliver personalized nutritional recommendations to its users in order to minimize blood sugar spikes after meals. After a first year of trial rollout in Israel, CEO Lihi Segal and her team are devising a global go-to-market plan for the firm. The team is considering several target markets, ranging from people with diabetes to professional athletes, and distribution strategies including selling direct to consumers or through partnerships with healthcare professionals or insurance companies. Their choices are important because they will affect DayTwo's costs, pricing, positioning, distribution channels, marketing efforts, and product development.
Teaching Note for HBS No. 519-011. As its Series A extension round approaches, the founders of Hubble, a subscription-based, social-media fueled, direct-to-consumer (DTC) brand of contact lenses, are reflecting on the marketing strategies that have taken them to a valuation of $200 million and debating changes to them that will allow them to grow their business. Ensuring that their marketing dollars were being spent efficiently was critical to the data-driven management team and proving to be complicated as the company moved spending from digital marketing to offline media, which made attribution modeling more difficult. Decisions pertaining to product extensions, channel expansion beyond DTC e-commerce, and geographic expansion were also on the table to prove that Hubble's customer value proposition and operations could profitably scale.
Israeli, Ayelet. "Hubble Contact Lenses: Data Driven Direct-to-Consumer Marketing." Harvard Business School Teaching Note 519-056, January 2019.
View Details
As its Series A extension round approaches, the founders of Hubble, a subscription-based, social-media fueled, direct-to-consumer (DTC) brand of contact lenses, are reflecting on the marketing strategies that have taken them to a valuation of $200 million and debating changes to them that will allow them to grow their business. Ensuring that their marketing dollars were being spent efficiently was critical to the data-driven management team and proving to be complicated as the company moved spending from digital marketing to offline media, which made attribution modeling more difficult. Decisions pertaining to product extensions, channel expansion beyond DTC e-commerce, and geographic expansion were also on the table to prove that Hubble's customer value proposition and operations could profitably scale.
CEO Art Peck was eliminating his creative directors for The Gap, Old Navy, and Banana Republic brands and promoting a collective creative ecosystem fueled by the input of big data. Rather than relying on artistic vision, Peck wanted the company to use the mining of big data obtained from Google Analytics and the company's own sales and customer databases to select the next season's assortment. Peck was betting that intelligence fueled by big data could outperform a fashion industry creative director at predicting the future fashion trends and tastes of consumers.
Israeli, Ayelet, and Jill Avery. "Predicting Consumer Tastes with Big Data at Gap." Harvard Business School Teaching Note 518-053, November 2017.
View Details
CEO Art Peck was eliminating his creative directors for The Gap, Old Navy, and Banana Republic brands and promoting a collective creative ecosystem fueled by the input of big data. Rather than relying on artistic vision, Peck wanted the company to use the mining of big data obtained from Google Analytics and the company’s own sales and customer databases to select the next season’s assortment. Peck was betting that intelligence fueled by big data could outperform a fashion industry creative director at predicting the future fashion trends and tastes of consumers.
Israeli, Ayelet, and Robert J. Dolan. "Angie's List: Ratings Pioneer Turns 20." Harvard Business School Teaching Note 517-123, May 2017. (Revised January 2019.)
View Details
In 1995, before people “googled” or “yelped,” Angela Hicks (HBS, 2000) was establishing her Angie’s List as a pioneer in the accumulation and dissemination of consumer rating information. Hicks focused on the home repair and maintenance market and, as she put it, “particularly on high cost of failure situations where good information on potential service providers is correspondingly of high value.” Angie's List had a paid subscription model as it charged “members” for access to the information they collectively provided on service providers.
More recently, companies such as Yelp, TripAdvisor, and Google have started offering free access to their reviews while relying totally on site advertising and service provider fees for their revenues. In 2015, Angie's List collected close to $68 million in membership fees. In January 2016, with recent declines in the growth rate of member numbers, Angie’s List has to decide if it was time to drop the “paid subscription for all” model and introduce a free version of its service to its product line.
