Most companies aim for exceptional customer service, but too few are attentive to the subtle discrimination by frontline employees that can alienate customers, lead to lawsuits, or even cause lasting brand damage by going viral. This article presents research about the way bias occurs in the provision of core products and services ("exchanges"), the furnishing of help that exceeds the minimum required ("extras"), and the manner in which service is delivered ("etiquette"). By breaking customer service into these three dimensions, the authors offer a framework for identifying and addressing frontline bias in your own organization. They recommend talking to your customers, examining available data, and running experiments to get a better sense of what biases exist among your customer service workers. Armed with that information, you might try to mitigate prejudiced behavior by broadening employees' exposure to people of diverse backgrounds, giving them standard procedures to follow when they interact with customers, and encouraging a sense of responsibility to act fairly.
This field-based case follows Andrew Yang, who went from an unknown to a major competitor in the 2020 Democratic Party presidential race. Yang's initial rise to fame was largely owed to a group of young people who called themselves the Yang Gang. They made memes, GIFs, and music videos, had lengthy discussions about Yang's campaign platform on Reddit, showed up at rallies, and served as small donors. This success, however, was a double-edged sword. While the Yang Gang brought the candidate name recognition, the marketing strategy that worked for this group of superfans did not work for the more conventional voter base. The case asks students to evaluate the role of digital in Yang's brand strategy and establish strategic priorities for maximizing his chances of securing the nomination. Students must consider the importance of different voter segments to Yang's campaign and consider how effectively each media channel (e.g., television, YouTube, email) speaks to each of the voter segments. This case has been successfully taught in a second-year MBA elective on digital marketing. It was used at the end of the module about integrating-and going beyond-digital channels to grow a brand. The material encourages exploration of managing digital and traditional marketing tools among stakeholders with different objectives. This case can also be used to teach students about market segmentation, brand essence (using brand pyramids as a tool), and the concept of "crossing the chasm." As such, it could also be used in a broader introductory marketing course.
This set comprises six exercises: in each, a team of students must shop online for products to solve a particular problem. Problems range from outfitting a new apartment quickly to purchasing food for a festival to choosing car insurance. Students will experience a customer journey as they navigate shopping websites.
In this set of three exercises, students are cast in the roles of brand managers of one of three categories of consumer goods companies. They must choose a specific brand in their category and identify keywords to advertise that brand on Google search, in order to leverage or react to evolving consumer behavior. Students research trend tools and Google's keyword planner, explore and analyze consumer demand, identify customer segments, research product features, identify competitors, and finally choose paid search keywords.
This public-sourced case uses Facebook, the legendary social media platform, to unfold circumstances that allow an analysis of the firm's privacy risk around its marketing tools and use of collected consumer data. Although Facebook had made progress on providing users more transparency around how it operated, how policies were enforced, and how shared data has been collected since 2018, challenges persisted-most notably over how to deal with misinformation on Facebook's platform and what to do with efforts to regain public trust. Facebook needed access to user data in order to ensure its advertising revenue source remained profitable. The scandal around user data and Cambridge Analytica leaves the case open to exploring privacy policies, data use, and factors driving consumer concerns regarding their data. The case provides an overview of Facebook's platform and policies that can be used to discuss the responsibility future general managers have to consider privacy and transparency when using consumer data. Had Facebook crossed the line with users over data collection? And what, if anything, should be done about regulators' increasing interest in how the firm conducted business? Although this case was written and taught before the global COVID-19 pandemic struck, it provides an interesting contrast to consumers' attitudes toward trust and legitimacy issues before and after COVID-19.
This public-sourced case describes the evolution of Facebook monitoring users' activity around what potentially offensive or disturbing material could stay posted and what should be removed. Following intense media and regulatory (e.g., Congressional) scrutiny, the company revised policies and reorganized content-review teams. The case offers an opportunity to debate a key question the company faced: What should platform governance look like? As Facebook faced an exponential growth in the number of user-generated posts daily, posts that contained harmful, inappropriate content also increased, raising the need for Facebook to devise a set of guidelines that were not only applicable across many different types of posts and situations, but were also perceived as acceptable by Facebook users. The tradeoffs a platform business makes between user characteristics, intent, outcome, and norms can be explored as students put themselves in CEO Mark Zuckerberg's shoes to redesign Facebook's platform-governance policies. Students will also have a chance to examine how certain platform-governance policies can subsequently impact Facebook's future strategy: for instance, the more Facebook got involved in curating content for its users-thus serving as arbiter-the more potential to shift Facebook from a tech business to a news publisher increased, thereby subjecting the company to a different set of regulations.
Data gathered on the web has vastly enhanced the capabilities of marketers. With people regularly sharing personal details online and internet cookies tracking every click, companies can now gain unprecedented insight into individual consumers and target them with tailored ads. But when this practice feels invasive to people, it can prompt a strong backlash. Marketers today need to understand where to the draw the line. The good news is that psychologists already know a lot about what triggers privacy concerns off-line. These norms--and the authors' research--strongly suggest that firms steer clear of two ad-targeting techniques generally disliked by consumers: using information obtained on a third-party site rather than on the site on which an ad appears, which is akin to talking behind someone's back; and deducing information about people (such as a pregnancy) from analytics when they haven't declared it themselves. If marketers avoid those tactics, use data judiciously, focus on increasing trust and transparency, and offer people control over their personal data, their ads are much more likely to be accepted by consumers and help raise interest in engaging with a company and its products.