個案總覽

This case explores the industry-wide lack of employee diversity in the technology sector, and Yelp's decision to take a leadership position in identifying strategies to increase diversity. The goal of the case is to provide an opportunity for students to develop a framework for understanding the factors that might lead to a less diverse workforce, and for evaluating approaches to increase diversity. The case opens in 2014, when Yelp hired Rachel Williams into a newly created position - the Head of Diversity and Inclusion. Rachel was tasked with developing strategies to ensure that Yelp was attracting and retaining a diverse, productive workforce, and creating a welcoming environment for all employees. The case puts students in Rachel's footsteps upon being hired, and tasks them to propose a set of changes for Yelp to make to increase the diversity of the workforce.

Paktor is a popular mobile-based online dating app from Singapore, where a user can swipe right or left on a profile to indicate her interest in a potential match. The case is designed to explore issues related to pricing, market design, and launch strategies in the context of online marketplaces. Students are asked to evaluate Paktor's existing design features and pricing, and formulate recommendations on design choices, pricing, and global expansion.

Online marketplaces such as eBay, Uber, and Airbnb have the potential to reduce racial, gender, and other forms of bias that affect the off-line world. And in the early days of internet commerce, the relative anonymity of transactions did make it harder for participants to discriminate. But as listings began to include photos, names, and other means of identification, bias emerged in areas ranging from labor markets to credit applications to housing--sometimes made worse by a lack of regulation, the absence of in-person interactions, and the use of automation and big data. How can companies reverse the tide? The key lies in more-intentional platform design, say the authors, who offer a framework for creating a thriving marketplace while minimizing the risk of discrimination. For starters, they say, companies must track and report on potential problems and carefully test choices that may influence the extent of discrimination. And they should thoroughly examine four design decisions, asking themselves: (1) Are we providing too much information? In many cases, the simplest, most effective change a platform can make is to withhold information such as race and gender until after a transaction has been agreed to. (2) Could we further automate the process? Features such as "instant book," allowing a buyer to sign up for a rental, say, without the seller's prior approval, can reduce discrimination while increasing convenience. (3) Can we make discrimination policies more top-of-mind? Presenting them during the actual transaction process, rather than burying them in fine print, makes them less likely to be broken. (4) Should we make our algorithms discrimination-aware? To ensure fairness, designers need to track how race or gender affects the user experience and set explicit objectives.

Spreadsheet supplement for case 916038.

Advertising Experiments at RestaurantGrades is an exercise in which students are asked to analyze and make a recommendation on the basis of simulated experimental data. The setting is a hypothetical restaurant review company called RestaurantGrades (RG), which shows advertisements above the organic search results when someone searches on the page. RG is trying to understand whether the current advertising package it sells to restaurants is effective at driving traffic-and whether an alternative advertising package might be better. To do this, RG has run an experiment with two treatment groups and a control group-with random assignment done at the restaurant level. Restaurants in the control group are provided with no advertising; restaurants in the first treatment group are given RG's current advertising package, and restaurants in the second treatment group are given an alternative advertising package. Students are provided with the resulting data-which includes an indicator for which arm a business was in, as well as outcomes at the monthly level for each business-and asked to assess which advertising package is more effective. The exercise provides students with an opportunity for experiential learning around the use of experiments to guide business decisions.

Algorithms tend to be myopic. They focus on the data at hand--which often pertains just to short-term outcomes. Algorithms are powerful predictive tools, but they can run amok when not applied properly. Consider what often happens with social media sites. Today many use algorithms to decide which ads and links to show users. But when these algorithms focus too narrowly on maximizing click-throughs, sites quickly become choked with low-quality content. While clicks rise, customer satisfaction plummets. The glitches, say the authors, are not in the algorithms but in the way we interact with them. Managers need to recognize their two major limitations: First, they're completely literal; algorithms do exactly what they're told and disregard every other consideration. While a human would have understood that the sites' designers wanted to maximize quality as measured by clicks, the algorithms maximized clicks at the expense of quality. Second, algorithms are black boxes. Though they can predict the future with great accuracy, they won't say what will cause an event or why. They'll tell you which magazine articles are likely to be shared on Twitter without explaining what motivates people to tweet about them, for instance. To avoid missteps, you need to be explicit about all your goals--hard and soft--when formulating your algorithms. You also must consider the long-term implications of the data the algorithms incorporate to make sure they're not focusing nearsightedly on short-term outcomes. And choose the right data inputs, being sure to gather a wide breadth of information from a diversity of sources.

This case presents a multi-party negotiation among Yelp, current partner OpenTable, and two startups in the online restaurant reservation industry.

The Behavioural Insights Team case introduces students to the concept of choice architecture and the value of experimental methods (sometimes called A/B testing) within organizational contexts. The exercise provides an opportunity for students to apply these principles to solve a managerial problem - increasing tax compliance rates among delinquent taxpayers. Students are asked to rewrite the letter that the UK tax department (HMRC) sends to delinquent taxpayers; this exercise is based on a successful behavioral field experiment run by the UK government.

Supplements the (A) case.

After widely-publicized complaints of destructive guests and unreliable hosts, online apartment rental site Airbnb explores mechanisms to facilitate trust between guests and hosts. Flexible online reputation systems can collect and share information with ease, but Airbnb must decide which information guests and hosts should have to provide and how much flexibility each should have in selecting whom to do business with. A full-featured system could provide all the information users have been requesting, but would it be too complicated for routine use?