Customer Needs and Solutions的封面图片
Customer Needs and Solutions

Customer Needs and Solutions

高等教育

CNS explores critical issues concerning the relationships among organizations and the customers they serve.

关于我们

Customer Needs and Solutions is an innovative, multidisciplinary journal exploring critical issues concerning the relationships among organizations and the customers they serve. ? Includes and integrates insights from such fields as marketing, strategy, psychology, and organizational behavior. ? Contains three types of content: Research Papers, Perspectives, and Unsolved Problems and Call for Solutions. ? Aims to facilitate communication between academics and stakeholders. ? Features overviews of leading marketing programs around the world.

网站
https://link.springer.com/journal/40547
所属行业
高等教育
规模
201-500 人
类型
教育机构

动态

  • ???????????????????? ?????????? ???????????????? ???????????? (????????) ?????? ???????????????????? ???????????????? ???????????????? ?????????? ???? ????????-???????? ? In latest research authored by Songting Dong on using Large Language Models (LLMs) to interpret consumer decision rules, published in Customer Needs and Solutions. ?? ? In today’s fast-paced and complex market, understanding consumer preferences is more challenging—and essential—than ever. Traditional methods often rely heavily on human judgment, which can be costly, time-intensive, and subjective (e.g., Ding et al. 2011). This study explores how fine-tuned LLMs, like GPT-4, can automate these insights, providing a scalable, real-time alternative to manual coding. ? ?? ?????? ????????????????: ? 1. ???????????????? ???????????????????? ????????????????: Fine-tuned LLMs outperform previous unstructured direct elicitation (UDE) models, capturing up to 25% more predictive information, especially in complex categories like automotive preferences. ? 2. ?????????????????????? ?????? ??????????????????????: LLMs eliminate the need for extensive manual input, maintaining high consistency and enabling large-scale applications in marketing. ? 3. ???????? ?????? ???????? ????????????????????: By replacing human agents, LLMs make data processing more affordable and instantaneous—crucial for industries working with real-time consumer data. ? ?? ?????? ???????? ??????????????: For brands aiming to stay ahead, these models offer a way to quickly comprehend consumer decision rules from unstructured data. This means more agile, informed decision-making and more personalized marketing strategies. Moreover, with LLMs’ ability to interact with users, they may serve as a valuable knowledgebase, helping to summarize preferences and decision rules, and support the creation and simulation of marketing strategies. ? As we look ahead, LLMs’ ability to handle nuanced decision-making will continue to improve. Future research will explore refining fine-tuning methods and integrating multimodal capabilities, enabling models to interpret increasingly complex consumer data. ? For more on this research, please access the article here: https://lnkd.in/eXFWp_Di. ? Reference: the dataset on which this research is built comes from Ding, M., Hauser, J., Dong, S., Dzyabura, D., Yang, Z., Su, C., & Gaskin, S. (2011). Unstructured Direct Elicitation of Decision Rules. Journal of Marketing Research. 48(1), 116-127. https://lnkd.in/e6by4yDK ? #Research #AI #MachineLearning #MarketingAnalytics #ConsumerBehavior #LLM #DataScience #Innovation

    • 该图片无替代文字
  • A really cool paper by Wayne Taylor and Anand Bodapati ... CNS would love to see more papers like this published in our journal.

    查看Wayne Taylor的档案

    Assistant Professor at Southern Methodist University

    I’m very excited to finally see this paper in print: The Effect of Gambling Outcomes on Casino Return Times with Scalable DDC, co-authored with Anand Bodapati at UCLA Anderson School of Management, is now published at Customer Needs and Solutions (link in the comments). This paper considers the direct marketing problem of whom to target when the customer learns about the firm through multiple interactions. We focus on situations where the customer outcome is random and can vary from occasion to occasion. A casino provides a nice empirical setting because it is easy to quantify differences between actual outcomes and expected outcomes (the house advantage, which the gamblers learn about over time). This phenomenon carries over to many other industries. For example, suppose your initial ride with Lyft was terrible due to chance - this sets low expectations about Lyft's ride quality. Lyft might consider a marketing offer to encourage the rider to give them another chance (i.e., more “draws”) where you will be more likely to experience better rides. To estimate the model, we combine forward simulation from dynamic discrete choice modeling with parallelization via Amazon Web Services (AWS) EC2 servers. By properly setting up the utility function we create a massively parallel estimation algorithm which allows us to incorporate complex learning and updating. In theory, we could fully parallelize this process across ~8.6 billion servers and estimation would be nearly instantaneous. Why a structural model? A causal ML learning might be useful for parts of the analyses (since the casino outcomes are nearly random), but to take full advantage of the policy simulations we prefer the structural parameters associated with the utility function. This allows us to simulate a variety of policies such as varying gambler beliefs and priors, evaluating marketing as a function of sequences of outcomes, and searching for optimal marketing policies. The counterfactuals suggest that casino profitability can increase substantially when marketing incorporates gamblers’ beliefs and past outcome sequences into the targeting decision. Interestingly, we find that marketing offers tend to be more effective when posterior beliefs of the house advantage are high, but uncertainty is low.

    • 该图片无替代文字
  • 查看Customer Needs and Solutions的组织主页

    267 位关注者

    Enhance Your Marketing Courses with AI-Driven Simulations ? AI-driven simulations are a powerful way to enrich marketing learning experience. By simulating real-world scenarios, students can gain practical insights and hands-on experience. Here are three examples of how simulations can be used: 1.?????Simulate Consumers with Different Expertise: Explore how consumers with varying levels of expertise behave differently when purchasing a fishing rod. This simulation helps students understand the impact of consumer knowledge on purchasing decisions and marketing strategies. Read more about this simulation here: https://lnkd.in/eFXb-kkD. 2.?????Simulate A Pricing Consultant for Pricing Decisions: Learn how to develop and implement effective pricing strategies for a café. This simulation allows students to experience value-based pricing, second-price discrimination, and other pricing tactics in a competitive market environment. Read more about this simulation here: https://lnkd.in/exKnjfMV. 3.?????Simulate An Interview with Steve Jobs: Engage in a role-play interview with Steve Jobs for a marketing position. This simulation provides insights into what top industry leaders might look for in a candidate, helping students prepare for real-world job interviews. Read more about this simulation here: https://lnkd.in/eHtmCcSP. These simulations demonstrate the potential of integrating Generative AI into marketing education, offering students a dynamic and immersive learning experience. ?? For more comprehensive insights and practical examples on the innovative uses of GenAI in marketing education, explore the latest paper on CNS: https://lnkd.in/efWHm_9N #Marketing #EdTech #GenAI #Simulations #EducationInnovation #AIinEducation

  • Here is a way to create an administrative customized chatbot for your undergrad and grad courses, as well as the link to the Customer Needs and Solutions article with more such ideas ...

    查看Customer Needs and Solutions的组织主页

    267 位关注者

    Everyone Can Use a Customized Chatbot for their Marketing Courses Customized chatbots can streamline administrative tasks and improve engagement with students. Here is a toy example of custom chatbot created using ChatGPT Plus for a Marketing Analytics course, you can test it here https://lnkd.in/gXUB_RdD. ?At this moment, users need to have a ChatGPT Plus account (paid account) to use custom GPTs. It is rumored that OpenAI will allow free users to access them in a few weeks. We have attached a few screenshots on how to construct such a customized chatbot using GPTs in OpenAI, as well as some sample usage of the chatbot. If you want to build your own chatbot, see the end of the post on how we did it, and you can do it too (spoiler alert, it’s easy!) . We have also included some screenshots (both for the process of creating it and some sample queries) For more comprehensive insights and practical examples on the innovative uses of GenAI in marketing education, explore the latest paper on GenAI and Marketing Classroom in CNS: https://lnkd.in/efWHm_9N #Marketing #EdTech #GenAI #Chatbots #EducationInnovation #AIinEducation ? Initiate the GPT Builder in the “Create” tab After logging onto ChatGPT using a ChatGPT Plus account, click “Explore GPTs” -> “MyGPTs” -> “Create a GPT” (as shown in the screenshot below). In the “Create” tab, you can chat with the GPT Builder to initiate the chatbot. In our example: GPT Builder: What would you like to make? User upload file and prompt: I would like to build a teaching assistant chatbot to help me handle administrative issues for my Marketing Analytics course. Please use the course outline and marking rubrics (the uploaded file) as the knowledge base. GPT Builder update GPT setting and ask: Great! The GPT is now set up as a teaching assistant for a Marketing Analytics course. It will assist with administrative queries using the course outline and marking rubrics as its knowledge base. Next, let's give this GPT a name. How about "Course Assistant"? Do you like this name, or would you prefer something different? User prompt: Please name it TA for Marketing Analytics. GPT Builder update GPT setting, generate profile image, and ask: Here's the profile picture for "TA for Marketing Analytics". Do you like it or would you like any changes? User prompt:???…… [some contents omitted here] -- Continue in the comment below --

    • 该图片无替代文字
    • 该图片无替代文字
    • 该图片无替代文字
    • 该图片无替代文字
    • 该图片无替代文字
  • Everyone Can Use a Customized Chatbot for their Marketing Courses Customized chatbots can streamline administrative tasks and improve engagement with students. Here is a toy example of custom chatbot created using ChatGPT Plus for a Marketing Analytics course, you can test it here https://lnkd.in/gXUB_RdD. ?At this moment, users need to have a ChatGPT Plus account (paid account) to use custom GPTs. It is rumored that OpenAI will allow free users to access them in a few weeks. We have attached a few screenshots on how to construct such a customized chatbot using GPTs in OpenAI, as well as some sample usage of the chatbot. If you want to build your own chatbot, see the end of the post on how we did it, and you can do it too (spoiler alert, it’s easy!) . We have also included some screenshots (both for the process of creating it and some sample queries) For more comprehensive insights and practical examples on the innovative uses of GenAI in marketing education, explore the latest paper on GenAI and Marketing Classroom in CNS: https://lnkd.in/efWHm_9N #Marketing #EdTech #GenAI #Chatbots #EducationInnovation #AIinEducation ? Initiate the GPT Builder in the “Create” tab After logging onto ChatGPT using a ChatGPT Plus account, click “Explore GPTs” -> “MyGPTs” -> “Create a GPT” (as shown in the screenshot below). In the “Create” tab, you can chat with the GPT Builder to initiate the chatbot. In our example: GPT Builder: What would you like to make? User upload file and prompt: I would like to build a teaching assistant chatbot to help me handle administrative issues for my Marketing Analytics course. Please use the course outline and marking rubrics (the uploaded file) as the knowledge base. GPT Builder update GPT setting and ask: Great! The GPT is now set up as a teaching assistant for a Marketing Analytics course. It will assist with administrative queries using the course outline and marking rubrics as its knowledge base. Next, let's give this GPT a name. How about "Course Assistant"? Do you like this name, or would you prefer something different? User prompt: Please name it TA for Marketing Analytics. GPT Builder update GPT setting, generate profile image, and ask: Here's the profile picture for "TA for Marketing Analytics". Do you like it or would you like any changes? User prompt:???…… [some contents omitted here] -- Continue in the comment below --

    • 该图片无替代文字
    • 该图片无替代文字
    • 该图片无替代文字
    • 该图片无替代文字
    • 该图片无替代文字
  • Adapting Marketing Education to the Generative AI Revolution ? In an open access article just published in Customer Needs and Solutions, Min Ding, Songting Dong, and Rajdeep Grewal explore the significant impact of Generative Artificial Intelligence (GenAI) on marketing education (see https://lnkd.in/efWHm_9N). Their research dives into how these technologies are reshaping the landscape of marketing practice and pedagogy, providing a timely response to the evolving demands for marketing education. ? The integration of GenAI tools into marketing courses is not merely an enhancement but a necessity. The article provides educators with an actionable framework to incorporate GenAI into their curricula, emphasizing tasks such as content generation, data analysis, and simulation exercises. These tools offer educators and students alike the opportunity to engage with marketing strategies that mirror the complexities and dynamics of the modern marketplace. ? Join the dialogue: How are you integrating GenAI into your marketing courses and strategies? What changes are you seeing in the field, and how are you preparing students for these new dynamics? We look forward to comments, feedback, and reactions. ? The paper comes with an associated PowerPoint slide deck that should help permeate the marketing classroom with GenAI worldwide. Do reach out the authors for details, questions, and clarifications. Also, for more in GenAI follow Customer Needs and Solutions. ? #MarketingEducation #GenAI #DigitalMarketing #EducationalInnovation

    • 该图片无替代文字
  • Here is another genAI application in marketing ... if you are interested in writing an article for CNS to share your experience using genAI, please contact Min Ding or Rajdeep Grewal

    查看Rajdeep Grewal的档案

    Townsend Family Distinguished Professor of Marketing, Kenan-Flagler Business School, University of North Carolina, CH.

    Step into the world of AI-powered review articles! ? Here is a study that Anindita Chakravarty embarked on. She published a review article in Customer Needs and Solutions (the bottom article in the picture) titled “Review of Marketing Relevant Real Activity Management.” We (Min Ding and Rajdeep Grewal) then asked her to explore the capabilities of ChatGPT-Plus to produce an academic review article. ? In a replication piece that relied on #PromptEngineering, appeared this year in CNS: “Replicating Published Literature Review using ChatGPT-Plus: Observations” (top article in the picture). In this research note, Anindita Chakravarty examines the performance of ChatGPT-Plus on six key dimensions: (A) Provide a structure for a manuscript with guidelines for sections, their sequence and general content. (B) Generate the constructs provided by the scholar and provide multiple definitions of the constructs based on how many contexts it can draw upon. (C) Provide constructs that are related to the initial constructs provided by the scholar. (D) Can draw upon multiple theories to provide relationships among all constructs. The specificity of the theory and relationship is contingent on the “nudge” provided by the researcher, i.e., the prompt used by the scholar. Nowadays there are guides for prompt engineering that help in narrowing down the most useful text, though there are few available for writing for academic journals. As such much depends on the expertise, effort, and time of the scholar to narrow down the prompt that will generate the most appropriate response sought. (E) Can generate tables with structured format, which might be especially useful for literature review tables. (F) Can generate a bibliography of relevant literature. ? Explore the articles to see the performance of ChaptGPT-Plus on these six attributes. CNS will be open to publish review articles?where authors follow this approach, if you are interested contact the Min Ding or Rajdeep Grewal. For more materials on GenAI and marketing in general, follow Customer Needs and Solutions.

    • 该图片无替代文字
  • Customer Needs and Solutions转发了

    Sentiment Analysis in the Age of Generative AI ? Jan Ole Krugmann and Jochen Hartmann from TUM School of Management at the Technical University of Munich (TUM) investigate Generative AI‘s capacity to perform sentiment analysis in their recent article published in Customer Needs and Solutions. Sentiment analysis helps understand consumers’ emotions and opinions towards brands, products, people, and organizations and?is one of the most important text classification tasks in marketing.? ? In their comparative, open-access study, Krugmann and Hartmann evaluate leading large language models (#LLMs), namely, GPT-3.5, GPT-4, and Llama 2, against established transfer learning models like SiEBERT. Across three experiments, they assess how different prompting approaches, linguistic characteristics, and data type affect accuracy of sentiment classification, providing practical insights for both marketing scholars and practitioners on selecting appropriate methods depending on the application context. In addition, the article sheds light on LLMs‘ capability to provide helpful explanations for their sentiment classifications.? ? Collectively, their findings suggest that LLMs, even when used as zero-shot classifiers, can often match and occasionally surpass specialized transfer learning models in performance. ? As #GenAI continues to evolve, how do you see its application for textual analyses such as sentiment analysis? What are your thoughts or experiences with this disruptive technology in your own work? ? #GenerativeAI #SentimentAnalysis #NaturalLanguageProcessing #DigitalMarketing #TUM #LLMs #GPT3 #GPT4 #Llama2

    • 该图片无替代文字
  • Sentiment Analysis in the Age of Generative AI ? Jan Ole Krugmann and Jochen Hartmann from TUM School of Management at the Technical University of Munich (TUM) investigate Generative AI‘s capacity to perform sentiment analysis in their recent article published in Customer Needs and Solutions. Sentiment analysis helps understand consumers’ emotions and opinions towards brands, products, people, and organizations and?is one of the most important text classification tasks in marketing.? ? In their comparative, open-access study, Krugmann and Hartmann evaluate leading large language models (#LLMs), namely, GPT-3.5, GPT-4, and Llama 2, against established transfer learning models like SiEBERT. Across three experiments, they assess how different prompting approaches, linguistic characteristics, and data type affect accuracy of sentiment classification, providing practical insights for both marketing scholars and practitioners on selecting appropriate methods depending on the application context. In addition, the article sheds light on LLMs‘ capability to provide helpful explanations for their sentiment classifications.? ? Collectively, their findings suggest that LLMs, even when used as zero-shot classifiers, can often match and occasionally surpass specialized transfer learning models in performance. ? As #GenAI continues to evolve, how do you see its application for textual analyses such as sentiment analysis? What are your thoughts or experiences with this disruptive technology in your own work? ? #GenerativeAI #SentimentAnalysis #NaturalLanguageProcessing #DigitalMarketing #TUM #LLMs #GPT3 #GPT4 #Llama2

    • 该图片无替代文字

相似主页

查看职位