Now , a leader of Northwestern University's prestigious analytics program presents a fully-integrated treatment of both the business and academic elements of marketing applications in predictive analytics. Writing for both managers and students, Thomas W. Miller explains essential concepts, principles, and theory in the context of real-world applications. Building on Miller's pioneering program, Marketing Data Science thoroughly addresses segmentation, target marketing, brand and product positioning, new product development, choice modeling, recommender systems, pricing research, retail site selection, demand estimation, sales forecasting, customer retention, and lifetime value analysis. Starting where Miller's widely-praised Modeling Techniques in Predictive Analytics left off, he integrates crucial information and insights that were previously segregated in texts on web analytics, network science, information technology, and programming. Coverage includes: The role of analytics in delivering effective messages on the web Understanding the web by understanding its hidden structures Being recognized on the web – and watching your own competitors Visualizing networks and understanding communities within them Measuring sentiment and making recommendations Leveraging key data science methods: databases/data preparation, classical/Bayesian statistics, regression/classification, machine learning, and text analytics Six complete case studies address exceptionally relevant issues such as: separating legitimate email from spam; identifying legally-relevant information for lawsuit discovery; gleaning insights from anonymous web surfing data, and more. This text's extensive set of web and network problems draw on rich public-domain data sources; many are accompanied by solutions in Python and/or R. Marketing Data Science will be an invaluable resource for all students, faculty, and professional marketers who want to use business analytics to improve marketing performance.
Explore new and more sophisticated tools that reduce your marketing analytics efforts and give you precise results Key Features Study new techniques for marketing analytics Explore uses of machine learning to power your marketing analyses Work through each stage of data analytics with the help of multiple examples and exercises Book Description Data Science for Marketing Analytics covers every stage of data analytics, from working with a raw dataset to segmenting a population and modeling different parts of the population based on the segments. The book starts by teaching you how to use Python libraries, such as pandas and Matplotlib, to read data from Python, manipulate it, and create plots, using both categorical and continuous variables. Then, you'll learn how to segment a population into groups and use different clustering techniques to evaluate customer segmentation. As you make your way through the chapters, you'll explore ways to evaluate and select the best segmentation approach, and go on to create a linear regression model on customer value data to predict lifetime value. In the concluding chapters, you'll gain an understanding of regression techniques and tools for evaluating regression models, and explore ways to predict customer choice using classification algorithms. Finally, you'll apply these techniques to create a churn model for modeling customer product choices. By the end of this book, you will be able to build your own marketing reporting and interactive dashboard solutions. What you will learn Analyze and visualize data in Python using pandas and Matplotlib Study clustering techniques, such as hierarchical and k-means clustering Create customer segments based on manipulated data Predict customer lifetime value using linear regression Use classification algorithms to understand customer choice Optimize classification algorithms to extract maximal information Who this book is for Data Science for Marketing Analytics is designed for developers and marketing analysts looking to use new, more sophisticated tools in their marketing analytics efforts. It'll help if you have prior experience of coding in Python and knowledge of high school level mathematics. Some experience with databases, Excel, statistics, or Tableau is useful but not necessary.
Optimize your marketing strategies through analytics and machine learning Key Features Understand how data science drives successful marketing campaigns Use machine learning for better customer engagement, retention, and product recommendations Extract insights from your data to optimize marketing strategies and increase profitability Book Description Regardless of company size, the adoption of data science and machine learning for marketing has been rising in the industry. With this book, you will learn to implement data science techniques to understand the drivers behind the successes and failures of marketing campaigns. This book is a comprehensive guide to help you understand and predict customer behaviors and create more effectively targeted and personalized marketing strategies. This is a practical guide to performing simple-to-advanced tasks, to extract hidden insights from the data and use them to make smart business decisions. You will understand what drives sales and increases customer engagements for your products. You will learn to implement machine learning to forecast which customers are more likely to engage with the products and have high lifetime value. This book will also show you how to use machine learning techniques to understand different customer segments and recommend the right products for each customer. Apart from learning to gain insights into consumer behavior using exploratory analysis, you will also learn the concept of A/B testing and implement it using Python and R. By the end of this book, you will be experienced enough with various data science and machine learning techniques to run and manage successful marketing campaigns for your business. What you will learn Learn how to compute and visualize marketing KPIs in Python and R Master what drives successful marketing campaigns with data science Use machine learning to predict customer engagement and lifetime value Make product recommendations that customers are most likely to buy Learn how to use A/B testing for better marketing decision making Implement machine learning to understand different customer segments Who this book is for If you are a marketing professional, data scientist, engineer, or a student keen to learn how to apply data science to marketing, this book is what you need! It will be beneficial to have some basic knowledge of either Python or R to work through the examples. This book will also be beneficial for beginners as it covers basic-to-advanced data science concepts and applications in marketing with real-life examples.
Discover how data science can help you gain in-depth insight into your business - the easy way! Jobs in data science abound, but few people have the data science skills needed to fill these increasingly important roles. Data Science For Dummies is the perfect starting point for IT professionals and students who want a quick primer on all areas of the expansive data science space. With a focus on business cases, the book explores topics in big data, data science, and data engineering, and how these three areas are combined to produce tremendous value. If you want to pick-up the skills you need to begin a new career or initiate a new project, reading this book will help you understand what technologies, programming languages, and mathematical methods on which to focus. While this book serves as a wildly fantastic guide through the broad, sometimes intimidating field of big data and data science, it is not an instruction manual for hands-on implementation. Here’s what to expect: Provides a background in big data and data engineering before moving on to data science and how it's applied to generate value Includes coverage of big data frameworks like Hadoop, MapReduce, Spark, MPP platforms, and NoSQL Explains machine learning and many of its algorithms as well as artificial intelligence and the evolution of the Internet of Things Details data visualization techniques that can be used to showcase, summarize, and communicate the data insights you generate It's a big, big data world out there—let Data Science For Dummies help you harness its power and gain a competitive edge for your organization.
Marketing has changed substantially in the last few years. With more and more research conducted in marketing and consumer behaviour fields, and technological advances and applications occurring on a regular basis, the future of marketing opens up a world of exciting opportunities. Going beyond a state-of-the-art view of the discipline, this innovative volume focuses on the advances being made in many different areas such as; critical thinking, new paradigms, novel conceptualisations, as well as key technological innovations with a direct impact on the theory and practice of marketing. Each chapter presents an expert overview, and an analytical and engaging discussion of the topic, as well as introducing a specific research agenda paving the way for the future. The Routledge Companion to the Future of Marketing provides the reader with a comprehensive set of visionary insights into the future of marketing. This prestigious collection aims to challenge the mindset of marketing scholars, transforming current thinking into new perspectives and advances in marketing knowledge. Foreword Wayne S. DeSarbo, Smeal College of Business, Pennsylvania State Univerity, USA "The Future of Marketing" presents 22 different chapters written by some of the top scholars in the field of Marketing. These 22 chapters are organized into four topical areas: (1) New paradigms and philosophical insights (Chapters 1-5), (2) Contributions from other scientific fields (Chapters 6-9), (3) Reconnecting with consumers and markets (Chapters 10-17), and (4) New methodological insights in scholarly research in the field (Chapters 18-22). Thus, there are a number of diverse areas treated here ranging from futuristic managerial philosophies to state of the art qualitative and quantitative methodologies applicable to the various types of Marketing problems to be faced in the future. There are a number of implicit guidelines (and future research areas and needs) that can be gleaned for (quantitative) modelers in terms of the issues and considerations that their constructed models should explicitly accommodate in future empirical endeavors: Heterogeneity When modeling consumer perceptions, preferences, utility structures, choices, etc., it is important to avoid potential masking issues that aggregate models are subject to in many cases. In the simple case, consider a regression scenario where there are two equal sized segments whose utility functions (as a function of price) are opposite reflections of each other. Aggregating the sample in one large analysis yields a non-significant price elasticity coefficient, whereas estimating separate utility functions by segment displays the true structure in the data. While latent structure and hierarchical Bayesian methods have been developed for disaggregate analyses, a number of methodological issues exist with such existent approaches that provide fertile ground for future research. Competition Many quantitative models are estimated at a brand level and reflect only the efforts of that sole brand. For example, in many customer satisfaction studies, attention is often paid to the consumers of a particular client brand or service in an effort to portray their performance and derive the important drivers of satisfaction. Financial optimization models are then often constructed to examine where a company should invest its resources to best improve sales, retention, word of mouth, loyalty, etc. These studies need to occur in a fully competitive setting where one derives a full picture of the competitive market place. Managers need to know the relative importance of the drivers of satisfaction for their brand/service as well as for their competitors. In addition, knowledge of the relative performance of their brand relative to competitors is necessary information for strategy formation. Ideally, one would hope to see modeling efforts which also examine cross effects in terms of how Brand A’s policy affects other brands. Over time, competitive dynamics are also important as discussed next. Dynamics As seen in the various chapters, this can assume many different manifestations. Related to the previous category above related to competition, it is often necessary to examine competitive dynamics as opposed to comparative statics where the modeler of the future examines simultaneous and/or sequential optimization by each of the competitors in a market place in a game theoretic context. In such a manner, it will not be the case that all competitors end up enacting the same exact identical strategies. Alternatively, the models of the future should be adaptive and have the ability to "learn" from past data, as well as benefit from informed managerial expert input and constraints. Parameter values that change/adapt during the duration of the data are also a desirable feature. Non-Linearity Traditional linear response functions do not typically yield realistic normative managerial guidelines or optimized solutions. End point solutions that suggest "all or none" types of resource allocations are useless in most realistic Marketing applications. A large amount of work is required in this area as Marketing often lacks the strong theory necessary to provide such insight regarding the models that are constructed. In addition, multiple objective functions need to be accommodated with the use of multicriterion optimization methods Endogeneity Often times, there are hidden effects embedded in the various independent variables the Marketer believes are exogenous and truly independent. These may be due to effect of lagged variables, managerial decision making practice, etc. To ignore such effects, threatens the integrity of the models Marketers construct. For example, in traditional regression models, such endogeneity often produces a correlation between the independent variable in question and the error term, often resulting in biased estimates when employing ordinary least-squares estimation. Moderation/Mediation There are times particularly in regression approaches where the relationships between two variables are affected by values of a third variable. In such cases, we need to employ selected interaction effects to measure such moderated effects. Interaction effects are often needed to model the synergistic or catalytic effects of various independent variables. Alternatively, in a mediation regression model, rather than hypothesizing a direct causal relationship between the independent variable and the dependent variable, a mediational model hypothesizes that the independent variable influences the mediator variable, which in turn influences the dependent variable. Thus, such moderator and mediator variables serve to clarify the nature of the relationship between the independent and dependent variables. Marketers need to be aware of such potential inter-relationships. Models Guided by Theory Ideally, the models we construct should be more than just data analytic structures which approximate the relationships found in the data. Where possible, models should be constructed on the basis of available sound Marketing theory describing the process being modeled. One of the advantages of structural equation models is that one can utilize such a methodology to test and implement some a priori theory describing the relationship or causal nature of various inter-related constructs. This feature has been lacking in the general modeling efforts to date. A major reason for this is due to the lack of adequate theory development for most of the processes encountered in Marketing. For example, we have no solid Marketing theory regarding the structure of marketing mix response models. Thus progress must be advanced in such areas so that the models we construct are more robust and explainable. I wish to personally thank the co-editors and various authors of the "Future of Marketing" for opening the door to get a glimpse of the future in the field of Marketing. The hope is that this new book will provide fresh ideas to guide future research to improve the field of Marketing and define the next generation of research efforts as the torch gets passed to future generations.
Who is most likely to buy and what is the best way to target them? Marketing Analytics enables marketers and business analysts to answer these questions by leveraging proven methodologies to measure and improve upon the effectiveness of marketing programs. Marketing Analytics demonstrates how statistics, analytics and modeling can be put to optimal use to increase the effectiveness of every day marketing activities, from targeted list creation and data segmentation to testing campaign effectiveness and forecasting demand. The author explores many common marketing challenges and demonstrates how to apply different data models to arrive at viable solutions. Business cases and critical analysis are included to illustrate and reinforce key concepts throughout. Beginners will benefit from clear, jargon-free explanations of methodologies relating to statistics, marketing strategy and consumer behaviour. More experienced practitioners will appreciate the more complex aspects of data analytics and data modeling, discovering new applications of various techniques in every day practice. Readers of Marketing Analytics will come away with a firm foundation in markets analytics and the tools they need to gain competitive edge and increase market share. Online supporting resources for this book include a bank of test questions as well as data sets relating to many of the chapters.
To enhance marketing analytics, approximate and inductive reasoning can be applied to handle uncertainty in individual marketing models. This book demonstrates the use of fuzzy logic for classification and segmentation in marketing campaigns. Based on practical experience as a data analyst and on theoretical studies as a researcher, the author explains fuzzy classification, inductive logic and the concept of likelihood and introduces a blend of Bayesian and Fuzzy Set approaches, allowing reasonings on fuzzy sets that are derived by inductive logic. By application of this theory, the book guides the reader towards a gradual segmentation of customers which can enhance return on targeted marketing campaigns. The algorithms presented can be used for visualization, selection and prediction. The book shows how fuzzy logic can complement customer analytics by introducing fuzzy target groups. This book is for researchers, analytics professionals, data miners and students interested in fuzzy classification for marketing analytics.
The volume presents innovations in data analysis and classification and gives an overview of the state of the art in these scientific fields and applications. Areas that receive considerable attention in the book are discrimination and clustering, data analysis and statistics, as well as applications in marketing, finance, and medicine. The reader will find material on recent technical and methodological developments and a large number of applications demonstrating the usefulness of the newly developed techniques.
This book is a complete introduction to the power of R for marketing research practitioners. The text describes statistical models from a conceptual point of view with a minimal amount of mathematics, presuming only an introductory knowledge of statistics. Hands-on chapters accelerate the learning curve by asking readers to interact with R from the beginning. Core topics include the R language, basic statistics, linear modeling, and data visualization, which is presented throughout as an integral part of analysis. Later chapters cover more advanced topics yet are intended to be approachable for all analysts. These sections examine logistic regression, customer segmentation, hierarchical linear modeling, market basket analysis, structural equation modeling, and conjoint analysis in R. The text uniquely presents Bayesian models with a minimally complex approach, demonstrating and explaining Bayesian methods alongside traditional analyses for analysis of variance, linear models, and metric and choice-based conjoint analysis. With its emphasis on data visualization, model assessment, and development of statistical intuition, this book provides guidance for any analyst looking to develop or improve skills in R for marketing applications.
Master predictive analytics, from start to finish Start with strategy and management Master methods and build models Transform your models into highly-effective code—in both Python and R This one-of-a-kind book will help you use predictive analytics, Python, and R to solve real business problems and drive real competitive advantage. You’ll master predictive analytics through realistic case studies, intuitive data visualizations, and up-to-date code for both Python and R—not complex math. Step by step, you’ll walk through defining problems, identifying data, crafting and optimizing models, writing effective Python and R code, interpreting results, and more. Each chapter focuses on one of today’s key applications for predictive analytics, delivering skills and knowledge to put models to work—and maximize their value. Thomas W. Miller, leader of Northwestern University’s pioneering program in predictive analytics, addresses everything you need to succeed: strategy and management, methods and models, and technology and code. If you’re new to predictive analytics, you’ll gain a strong foundation for achieving accurate, actionable results. If you’re already working in the field, you’ll master powerful new skills. If you’re familiar with either Python or R, you’ll discover how these languages complement each other, enabling you to do even more. All data sets, extensive Python and R code, and additional examples available for download at http://www.ftpress.com/miller/ Python and R offer immense power in predictive analytics, data science, and big data. This book will help you leverage that power to solve real business problems, and drive real competitive advantage. Thomas W. Miller’s unique balanced approach combines business context and quantitative tools, illuminating each technique with carefully explained code for the latest versions of Python and R. If you’re new to predictive analytics, Miller gives you a strong foundation for achieving accurate, actionable results. If you’re already a modeler, programmer, or manager, you’ll learn crucial skills you don’t already have. Using Python and R, Miller addresses multiple business challenges, including segmentation, brand positioning, product choice modeling, pricing research, finance, sports, text analytics, sentiment analysis, and social network analysis. He illuminates the use of cross-sectional data, time series, spatial, and spatio-temporal data. You’ll learn why each problem matters, what data are relevant, and how to explore the data you’ve identified. Miller guides you through conceptually modeling each data set with words and figures; and then modeling it again with realistic code that delivers actionable insights. You’ll walk through model construction, explanatory variable subset selection, and validation, mastering best practices for improving out-of-sample predictive performance. Miller employs data visualization and statistical graphics to help you explore data, present models, and evaluate performance. Appendices include five complete case studies, and a detailed primer on modern data science methods. Use Python and R to gain powerful, actionable, profitable insights about: Advertising and promotion Consumer preference and choice Market baskets and related purchases Economic forecasting Operations management Unstructured text and language Customer sentiment Brand and price Sports team performance And much more