個案總覽

Mirakl provided the technology and seller network required for companies like Macy's, Best Buy, Walmart, Siemens, or Carrefour to simply design, launch, and administer a marketplace that included products from third-party sellers. What began as a basic business idea in 2012 had grown into a thriving venture ten years later. Mirakl expected to grow its $100 million revenue by fivefold during the next five years. But how was this to be accomplished? Mirakl had previously prioritized the development of solutions for marketplace operators, the "Macy's of the world". It was now working on a new solution, Mirakl Connect, with the goal of becoming the premier destination for third-party sellers. But, how should Mirakl monetize these services? Should they charge for them, and if so, to whom? Should Mirakl explore adjacent opportunities, such as financial services, fulfillment, or advertising? Finally, should Mirakl expand its services to marketplaces outside the Mirakl ecosystem?

Many executives are struggling to determine the extent to which they should integrate the management of employees and external contributors. The MIT SMR-Deloitte Future of the Workforce team looks at Cisco as a case example of one organization focused on a more strategic integration of its contingent workforce.

While not every organization will need to build a brand in the metaverse or transact with cryptocurrencies, all organizations manage credentials as issuers, holders, and verifiers. That's why the one piece of Web3 technology that managers should begin looking at is decentralized credentialling systems based on blockchain. While it's still early days for this application area, pilot projects are showing promise, in addition to highlighting issues that must be addressed.

Developers of many new AI solutions produce statistics showing that the tools make critical decisions with greater accuracy and efficiency than humans. But managers tasked with evaluating these applications need to peel back the layers of developers' performance claims and focus on the ground truth used to train and validate the AI tool. This article explains what ground truth is, how to identify it for a given tool, and how to understand the implications for tool quality.

Smart products like robotic vacuums and smart speakers have become common in many households. However, although consumers value the convenience and time savings smart products offer, the accompanying feeling that they've relinquished some control to a machine is inhibiting the products' broader adoption and customer satisfaction. Based on their research, the authors recommend ways that smart-product companies can increase consumers' sense of control and autonomy when using smart technology.

Synthetic media presents looming threats to managers in a business setting. To address this issue, we first offer a short overview of the evolution of media manipulation to contextualize the new era of synthetic media. Then, we present the problems associated with synthetic media via veridicality and heuristics to illustrate how consumers have little choice but to believe what they see, read, and hear online. We outline the most likely and impactful types of synthetic media threats and attacks and present a synthetic media incident response playbook. Our aim is to inform managers about six specific phases so they can prepare, assess, detect, analyze, and recover from synthetic media incidents and coordinate their lessons learned.

Is there a model of the ideal business school instructor? I asked myself this question many times when first teaching in a business school classroom. Assessing the style and approach of other effective instructors signaled that there are many models for success in the classroom, but how should I think about developing my own successful teaching style and approach? To address this challenge, I developed the concept of the personalized teaching signature, which accounts for teaching as a complex, multidimensional activity that includes preparation, practicality, rigor, entertainment, empathy, experiential engagement, enthusiasm, novelty, and surprise. I argue that an instructor should strive to exceed an expected threshold of proficiency in each teaching dimension and then focus on specific dimensions in which they can excel to develop a personalized teaching signature. In this article, I describe the concept of a personalized teaching signature, outline its basic dimensions, and offer a step-by-step process for using this signature to become a better business school instructor.

Monifa Porter was in the process of building a product that could work her and her team out of a job. As head of 2401, a subdivision of the corporate innovation and growth company Mach49, Porter was building a technology product that would assist corporations as they designed, built, and scaled new ventures within their existing organization. In late 2022, Porter's team released a minimum viable product of 2401, and was now in the process of designing a product feedback and enhancement cycle that would hopefully turn 2401 into a major revenue stream for Mach49. To do so, Porter would need to run her team with discipline, all while weighing important questions about whether the innovation process could be automated and if so, what implications that had for her team of product managers.

The negative effects of bias in artificial intelligence models' underlying data has made headlines, and companies need to find ways to address it. But it's impossible to completely abolish bias in AI data to equitably account for diverse populations so instead, companies should remediate it to deliberately compensate for unfairness. The author describes a three-step process that can yield positive results for leaders looking to reduce the impact of AI bias.

For all the hype around bossless companies, managerial authority and hierarchy are still the best ways to ensure coordination, cooperation, creativity, predictability, and accountability in large organizations. But authority and hierarchy must be aligned to a company's specific needs. To fine-tune them, leaders should consider the need for speed in decision-making, the extent of employee knowledge, what knowledge is truly required, employee feelings of ownership, and procedural justice.

In increasingly diverse workplaces, imposter syndrome may inhibit the performance of some of the best and brightest rising stars. Managers can help transform the hold that these fears can have on employees and make that concern a fleeting thought instead of a permanent restraint.

When the author launched what would become Marico as a division within his family's business, Bombay Oil, it was with product innovation: Instead of selling edible oils in bulk to other businesses, it would sell in smaller, branded packages directly to consumers. Eventually the division became a separate entity, which is now one of India's largest homegrown CPG companies. Its growth has depended on constant innovation-around not just products, packaging, and pricing but also supply chain, talent management, and business models. Over the past decade Marico has branched out into services with its Kaya skin-care spas, pioneered the use of premium hair oils, and added savory oats to Indian diets. Through the Marico Innovation Foundation, Mariwala also promotes innovative thinking outside the company, supporting small businesses and entrepreneurs in their efforts to scale up new ideas. The key to doing that well, he says, is to be ever curious about customer needs, to create a flat hierarchy that rewards risk-taking, to learn from every failure, and to constantly prototype, experiment, refine, and retest.

AI initiatives at many organizations are too small and too tentative. They never get to the only step that can add economic value-being deployed on a large scale. Testing the waters may deliver valuable insights, but it probably won't be enough to achieve true transformation. A pilot program or experiment can take you only so far. The authors have identified 30 companies that have gone all in on AI-and achieved success-as well as 10 actions those companies took to become successful AI adopters: (1) Know what you want to accomplish. (2) Work with an ecosystem of partners. (3) Master analytics. (4) Create a modular, flexible IT architecture. (5) Integrate AI into existing workflows. (6) Build solutions across the organization. (7) Create an AI governance and leadership structure. (8) Develop and staff centers of excellence. (9) Invest continually. (10) Always seek new sources of data. In other words, you need to be aggressive enough with AI that the technology eventually transforms every aspect of your business.

Daniel Jones and Niki Lewis of Bext360 are struggling to persuade a potential corporate client to use their software as a service (SaaS) solution that focuses on Environmental, Social, and Governance (ESG) metrics. In a meeting with a big-box retailer, initial enthusiasm for the platform's capabilities, including measuring Scope 3 emissions, turned to caution when a senior corporate attorney raised concerns about legal liability. The deep transparency offered by the platform could expose the company to risks related to their suppliers' actions, leading the executives to postpone their decision. Jones and Lewis now face a paradox. Their technology offers in-depth ESG oversight, which companies should theoretically want. However, this transparency also introduces risks and responsibilities that potential clients may be hesitant to assume. The challenge lies in persuading clients to see the transformative potential of their ESG practices despite these risks.

Organizations that aim to develop and deploy responsible AI systems must begin by considering what implicit and systemic biases reside in a technological system itself. Not doing so can result in catastrophic failures resulting from unintended consequences of the system's normal operations. The authors advise managers on how to mitigate the risk that algorithmic systems can cause organizational or social harm if left unchecked.

When Yonyou's founder Wang Wenjing was considering the pros and cons of fully embracing the cloud-based SaaS (software as a service) model, the company was on the road to becoming China's leading ERP software provider. It took years for Yonyou to entirely change its business model from selling software licenses to collecting services fee. Industry-wide, the trend was more than clear that providing cloud-based service would become a predominant business model, but it was still a difficult decision for individual competitors because it would mean a series of changes, including reshaping executives' mindset, reorganizing teams, cutting staff, redefining key performance indicators and incentive plan, to support its shift from the traditional cash cow business to an emerging one. Some actions were painful but necessary. Additionally, Yonyou's management needed to keep a close eye on its financials just in case that investors might not be satisfied with the results. The new business model that Yonyou adopted was a cloud platform that provided a base and a set of uniform standards for partners such as independent developers and service providers to join and prosper. The biggest challenge ahead was to take every opportunity to educate the market and grow its subscription-based revenue so that it was able to stand fast in terms of domestic market share.

Developing AI systems based on neural networks can require large volumes of labeled training data, which can be hard to obtain in some settings. New techniques for reducing the number of labeled examples needed to build accurate models are now emerging to address this problem. These approaches encompass ways to transfer models across related problems and to pretrain models with unlabeled data. They also include emerging best practices around data-centric artificial intelligence.

While the popular view is that insights are the key benefit of artificial intelligence, in truth AI creates value by improving the quality of decisions. The good news is, the opportunities for it to do that in business are countless. But because decisions in one area of an organization usually have an impact on decisions in other areas, introducing AI often entails redesigning whole systems. In that way, AI is similar to groundbreaking technologies of the past, like electricity, which initially was used only narrowly but ultimately transformed manufacturing. Decisions involve a combination of prediction and judgment, and because AI makes highly accurate predictions, it will shift decision rights to where judgment is still needed, potentially changing who makes decisions and where, when, and how. More-accurate predictions in one part of a value chain will also have ripple effects on other parts. For instance, if a restaurant can reliably forecast the amount of ingredients it needs each week, its orders will fluctuate, making its suppliers' sales more uncertain. Strong communication is needed to synchronize effort and resources in a system, and modularity will help prevent changes in one area from disrupting others.

The new system in a day-care center is producing biased results--setting off an alarm when some dark-skinned parents enter the lobby. The director of the center must decide whether improving security is worth the possible legal ramifications. This fictional case study by Mary C. Gentile, David Danks, and Maralee Harrell features expert commentary by Joseph Steinberg, Cedric L. Alexander.

The new system in a day-care center is producing biased results--setting off an alarm when some dark-skinned parents enter the lobby. The director of the center must decide whether improving security is worth the possible legal ramifications. This fictional case study by Mary C. Gentile, David Danks, and Maralee Harrell features expert commentary by Joseph Steinberg, Cedric L. Alexander.