Volume 10 (2024-25)

Each volume of Applied Marketing Analytics consists of FOUR 100-page issues, published in print and online. Articles published in Volume 10 include:

Volume 10 Number 4

Editorial
In support of equivalent experience for analytics leadership
Chris Hogan, Managing Vice President, Data Science and Analytics Operations, Gartner Digital Market
Practice Papers
From test to scale: Leveraging generative AI for the five ‘C’s of digital marketing
Kelly Cutler, Lecturer, Northwestern University

Digital marketing has become increasingly complex with the proliferation of new technologies, platforms and data sources competing for shrinking marketing budgets and waning consumer attention spans. Increased government scrutiny and focus on consumer protection have also created additional difficulty and competition for digital marketers. Simultaneously, innovations in machine learning and generative artificial intelligence (GenAI) are driving new levels of operational efficiency, offering marketers transformative tools to enhance productivity. However, many organisations are not prepared to introduce this new technology and create a culture of testing and scaling AI technologies within the organisation. Adoption of advanced tools represents significant opportunities for digital marketers combined with challenges in a business environment bogged down by regulatory concerns, employee apprehension and rising operational costs. To navigate these tumultuous times successfully, digital marketers require a healthy blend of advanced technical knowledge and calculated risk-taking. By concentrating on the five ‘C’s of digital marketing (content, creative, campaigns, conversions and customer), marketers can identify the best instances for adopting GenAI into their process. To maximise impact and efficiency, marketers must commit to bringing GenAI into the organisation with a streamlined method. This paper introduces the ITMS (introduce, test, measure, scale) framework, offering a four-step progression for AI projects from start to scale. With these frameworks, digital marketers can lead an organisation to increased productivity by introducing new and useful GenAI tools to create meaningful value and drive quantifiable efficiency.
Keywords: generative AI; GenAI; AI; artificial intelligence; digital marketing; marketing and AI; digital marketing and AI
How data-driven personalisation allows marketers to meet customers’ ever-increasing expectations for relevance
Zontee Hou, President and Chief Strategist, Media Volery

Consumer expectations for relevance are higher than ever. Research conducted by the author found that consumers want brands to offer personalisation, with 85 per cent of US adults saying that it is ‘somewhat’ to ‘extremely’ important for brands to share personalised recommendations with them. While marketers agree that it is important for brands to offer personalised marketing messages and offers to consumers, less than half of marketers say their organisations collect enough data. This paper explores three of the key roadblocks to building a robust data-driven marketing organisation and the methods by which organisations can address those challenges. The paper concludes with a look at the opportunities to build competitive advantage through more personalised and relevant marketing messages, in an environment where the majority of organisations have not yet acted upon these customer expectations.
Keywords: data-driven personalisation; consumer expectations; marketing personalisation; data challenges; relevance in marketing
The effect of landing page design and structure on conversion rates
Marcin Hylewski, Content and Marketing Specialist, Artur Pluskota, Engineering Manager and Błazej Abel, CEO & Founder, Landingi

This paper investigates the effect of landing page design and structure on conversion rate, based on an analysis of 25,027 landing pages created by anonymous Landingi customers. For the purposes of the research, a script in Julia language was developed to automatically recognise the elements of individual landing pages, as well as parameters such as number, position, size, colour, etc. Data extracted this way were then used to train a model examining the relationship between the aforementioned parameters and the conversion rate. Finally, the material obtained was subject to interpretation by the research team. The paper presents and discusses the most relevant factors, the effect of which on conversion rate has proven indisputable, including the height of the landing page, the volume of space occupied by images, the use of video, the length of copy, the number and position of buttons, the shape of forms and the use of a promotional counter.
Keywords: landing page; conversion rate; web design; digital marketing; machine learning; UX; page structure
Using conjoint analysis across the marketing value chain
Marco Vriens, CEO, Kwantum and Felix Eggers, Chief Consumer Science Officer, Kwantum and Professor, Copenhagen Business School

Conjoint analysis, best known in its choice-based form, has become one of the most commonly applied techniques in marketing research since it was introduced in the early 1970s. Its popularity stems from its ability to provide a systematic, experimental framework for collecting and analysing data on how product attributes and their levels influence consumer preferences and decision making. By simulating real-world trade-offs, conjoint analysis generates managerial insights such as the relative importance of product features, consumers’ willingness to pay and predicted market shares, making it an essential tool for product design, pricing strategies and competitive positioning. Methodological advances have enabled researchers to apply conjoint across the entire marketing value chain, as well as to deal with the associated challenges, such as how to deal with (too) many attributes, which type of experimental design to use, how to minimise hypothetical bias, whether to include benefits or other meta-attributes, how to account for non-compensatory decision making, how to account for consumer budgets, etc. This paper discusses the challenges encountered when applying conjoint across the marketing value chain, and the methods best suited to manage these challenges. In this way, the paper provides a concise user’s guide to making good methodological choices without getting drowned in the vast literature on this topic.
Keywords: marketing; marketing analytics; market research; conjoint; generative AI; GenAI; AI; large language models; LLMs
Enhancing market research with a GPT-based API wrapper: A leap towards advanced data analysis
Michael Dupin, Adjunct Professor of Data Science, Merrimack College and M. Furkan Oruç, Researcher, Maritime, Spire

This paper introduces an innovative application programming interface (API) wrapper built around OpenAI’s GPT, designed to significantly enhance the capabilities of large language models (LLMs) for market research. Traditional applications of GPT in survey crafting, response analysis and open-ended query interpretation, although effective, are limited by the standard interfaces provided. These interfaces (such as ChatGPT) often fall short when handling complex analytical tasks and ensuring data security, particularly in data-sensitive sectors like healthcare and finance. Our proposed API wrapper addresses these shortcomings by extending GPT’s functionalities beyond basic text generation and interaction. By integrating a sophisticated user interface (UI) and query management system, the wrapper simplifies interactions between users and the LLM. This allows researchers and analysts, regardless of their programming expertise, to leverage advanced data analysis tools and gain deeper insights from large datasets. Moreover, the wrapper ensures high standards of data privacy and security. Key features of the API wrapper include flexible UI options, ranging from open-source platforms to commercial software-as-a-service solutions that cater to diverse organisational needs and technical capabilities. The wrapper also incorporates advanced query handling and error management techniques to enhance user experience and efficiency. Through realworld applications and case studies, we demonstrate the wrapper’s ability to facilitate complex data analysis tasks such as thematic analysis of narrative data, comparative studies and multilingual data processing. The development and implementation of this wrapper marks a significant step towards democratising access to powerful AI tools in market research. It opens new possibilities for extracting nuanced insights and conducting sophisticated analyses without the need for deep technical knowledge or significant manual labour, thereby broadening the scope and impact of research methodologies in various industries.
Keywords: large language models (LLMs); GPT; text analysis; API wrapper; automated insights
Research Papers
Redefining business with artificial intelligence: Strategies for innovating consumer experience
Sari Yehuda Nassar, Doctoral Researcher, University of Barcelona

Artificial intelligence (AI) stands as a transformative force in the era of the Fourth Industrial Revolution, offering unprecedented improvements in efficiency, accuracy and decision-making across various sectors. As AI becomes increasingly integral to marketing, companies leveraging AI-powered strategies are poised to gain a significant competitive edge. This paper explores the intersection of AI and marketing, with a particular focus on its effect on consumer behaviour — the core target of modern marketing efforts. Employing a qualitative data analysis approach, the study involves four key stages: data collection, organisation, interpretation and presentation. This methodology is chosen to provide an in-depth understanding of consumer perceptions and attitudes towards AI in marketing. Rather than merely describing AI’s benefits, the paper adopts an analytical and narrative approach to reveal how AI enhances business strategies and operations while improving consumer satisfaction. The analysis delineates patterns of AI integration and contributes to the theoretical framework of AI’s role in marketing, highlighting its strategic importance and implications for future research.
Keywords: artificial intelligence; AI; consumer behaviour; marketing strategies; business innovation; AI-driven services; cosmetic industry
Data governance in the age of artificial intelligence: Challenges, best practices and regulatory compliance
Animesh Kumar Sharma, Research Scholar and Rahul Sharma, Professor, Mittal School of Business, Lovely Professional University

The growing use of artificial intelligence (AI) across businesses has created serious issues concerning data governance and privacy. As AI technologies rely significantly on massive datasets to learn, adapt and make choices, enterprises must provide strong data governance to protect data integrity, provide security and ensure compliance. Data governance encompasses the policies, procedures and standards that guarantee data are accurate, accessible and properly used throughout their life cycle. Data governance frameworks are becoming important in the context of AI, as AI systems handle and analyse vast amounts of data, frequently containing sensitive or personal information. This research paper examines the importance of data governance in the age of AI, stressing both the benefits and problems it provides. Effective data governance frameworks can assist firms in making better decisions, ensuring regulatory compliance and protecting user privacy. Traditional governance systems face substantial challenges from issues such as data bias, data quality and the complexity of managing AI-driven datasets. To reduce these dangers, best practices in data governance are highlighted, such as data classification, metadata management and the introduction of AI-specific governance standards. As AI technologies advance, the significance of adaptable and transparent data governance frameworks cannot be overemphasised. This study adds to our understanding of how corporations should reconcile the rapid advances in AI with the importance of strong data governance and privacy regulations.
Keywords: data governance; data management; data governance frameworks; data governance best practices; regulatory compliance; artificial intelligence

Volume 10 Number 3

Special issue: Customer data
Guest-Editor: Oliver Schiffers, Head of Customer Data Consulting, adesso SE

Editorial
Customer data in marketing analytics
Oliver Schiffers, Head of Customer Data Consulting, adesso SE
Special Issue Papers
A brief history and the future of customer data
Jim Sterne, Applied Technology Evangelist and Thomas H. Davenport, Professor, President’s Distinguished Professor of Information Technology, Babson College

This paper explores the evolution of customer data technologies from their inception in the 1980s to the present and speculates about the future. The paper highlights promises made and challenges encountered, examining the rise of database marketing, the impact of the Internet and social media, and the ongoing struggles with personalisation and customer relationship management. The paper reviews several different aspects of the customer relationship and discusses how and why technology and data have not been as effective as promised. By reviewing case studies, including Amazon’s pioneering efforts, and recent advancements in artificial intelligence — particularly generative AI — the paper provides a comprehensive overview of the current state of customer data technology. The paper also argues that the current array of tools that marketers can employ for customer relationships is much richer than in the past and could be used to dramatically improve customer satisfaction and value. Looking forward, the paper explores the potential for data integration, federation and advanced AI technologies to finally deliver on the long-promised benefits of personalised customer experiences. Given past outcomes, the current authors hesitate to suggest that it will solve all customer-related marketing problems, but are optimistic about improvement.
Keywords: AI; customer data, personalisation, generative AI, customer relationship management (CRM), data integration, data federation, sentiment analysis
Augmenting customer lifetime value with omnichannel engagement
Carolyn Tang Kmet, Associate Professor, and Jonathan Copulsky, Senior Lecturer, Northwestern University

Organisations have long used lifetime value (LTV) to measure the value a customer represents to the business over the course of the relationship. Once a business quantifies this value, LTV informs decisions ranging from how much to spend on acquisition and engagement campaigns to how much to expend on customer service and retention efforts. In this paper, current philosophies and measurement approaches used to calculate customer LTV are briefly reviewed and an evolution of the LTV framework that incorporates omnichannel customer engagement is recommended. While current LTV inputs rely mostly on first party data such as website behaviour, e-mail open and click-through rates and paid media engagement, the value of a customer-brand relationship extends beyond interaction with owned and paid media channels. The framework proposed in this paper provides practitioners with a new approach to evaluate customer value that additionally provides insight into a new segment of potential influencer and affiliate partners and suggests that the value of customers may extend beyond their spend with a brand to include their impact on other customers’ spend.
Keywords: customer lifetime value, CLV, omnichannel engagement, customer engagement value, CEV, paid media engagement, owned media engagement, earned media engagement, predictive customer behaviour
How does the customer experience benefit from better customer data?
Tarik Kilinc, Senior Digital Marketing Consultant, adesso SE

This paper explores how improved customer data enhances the customer experience through advanced personalisation, data-driven decision making and optimised marketing strategies. By leveraging zero party data and first party data, organisations can gain deep insights into customer preferences and behaviours. Integrating these types of data into customer data platforms allows for the creation of unified customer profiles, enabling seamless, cross-channel personalised experiences. Additionally, A/B-testing plays a crucial role in refining marketing efforts based on these comprehensive data insights.
Keywords: customer data, personalisation, customer experience, zero party data, first party data, customer data platforms, A/B-testing
The four Cs of B2B targeting: Using company, customer, channel and contextual data to shrink the audience bullseye
Stephen Diorio, Managing Director, Revenue Enablement Institute and Rich Howarth, Chief Executive Officer, Terminus Software

The combination of highly focused audience populations and much more precise industry, account and buyer persona-based segments make using deterministic approaches to plan, target, deploy and optimise media and marketing programmes in a business-to-business (B2B) environment a major challenge. Recent media benchmarking data confirms that hyper-pareto economics of audience targeting in B2B marketing can dramatically improve the yield and business impact of investments in paid, owned, earned and shared media programmes. This paper builds on existing academic research into audience targeting by adding qualitative interviews with 100 B2B marketing leaders and specific case examples from Terminus to examine how leading marketers are using probabilistic analytical approaches to shrink the audience bullseye and improve the business impact of media. The paper explores how B2B marketers are taking advantage of advances in analytics and artificial intelligence (AI) to combine company, customer, channel and context data from a variety of internal and external sources to model highly targeted and accurate audiences and segments. It provides a blueprint for how the best marketers are using internal company, opportunity and sales coverage data from customer relationship management (CRM) and first party customer engagement data from paid, owned, earned and shared digital marketing channels as the foundation for targeting audiences and identifying signals of interest, consideration, response and intent. It shows how marketers can further supplement this targeting data foundation with third party contextual data around account structures, audience graphs and buyer intent to continuously improve targeting accuracy with closed loop analysis of customer response data. Shrinking the audience bullseye in this manner allows marketers to improve the financial impact of marketing investments by a third or more, by targeting marketing programmes and resources to prospects in specific accounts, job roles and sales territories while zeroing in on the fraction of prospects who are actively in the market to buy at any time.
Keywords: account based marketing, revenue operations, connected media strategies, data-driven audiences, persona-based targeting, dark funnel, deterministic targeting
Deriving consumer insights with segmentation for identity and consent
Pamela Castricone, Emerging Solutions Strategist - Data Science, and Lucas Long, Head of Global Privacy Strategy, InfoTrust

The emerging privacy landscape is placing constraints on how consumer data can be collected and processed, especially for advertising use cases. Given the privacy-centric environment, organisations should approach their first party consumer data through the lens of what can be done with the data. This paper proposes a framework for segmenting consumer data based on identity and consent state, with strategies to derive the most consumer insights possible from each. Strategies focus on the role of data modelling in uncovering insights from first party data, and various modelling techniques are explored, including customer lifetime value modelling, propensity modelling, regression-based attribution and aggregate data modelling. By adopting a privacy-centric approach to data collection and leveraging advanced modelling techniques, organisations can gain valuable insights into consumer behaviour, optimise marketing and advertising efforts, and thrive in the evolving landscape of consumer data analysis.
Keywords: consumer data privacy, first party data, data segmentation, data modelling techniques, consumer insights, privacy-centric advertising, consumer behaviour analysis
Optimising lead qualification through machine learning: A customer data-driven approach
Helen Josephine V L, Associate Professor, Business Analytics, Vasudevan Moorthy, Associate Professor, Marketing, and Bindhia Joji, Assistant Professor, Business Intelligence, Christ University and Chandana Sai, Manager, Enterprise Solution Architect, Airtel Business

Lead generation is the process of turning an outside person or business into a customer of the business. Traditionally, marketing personnel must conduct significant follow-ups in order to convert even one potential consumer. Converting bad client leads can cause businesses to burn through cash reserves. As a result of this, it is now necessary to develop an automated system that can correctly anticipate whether or not a lead should be explored (converted to a customer or not). In this study, an attempt is made to evaluate historical data for leads produced by other businesses in order to train and validate a machine learning (ML)/deep learning (DL) model and test it against real-world characteristics to categorise them as hot leads (convert to customers) or cold leads (failed leads). This can be achieved by employing ML algorithms, low code–no code libraries, such as PyCaret in Python, and can be used to make predictions regarding probable lead creation, propensity to convert generated leads and optimal actions on the leads by communications teams. Supervised ML algorithms such as logistic regression, decision trees, random forests and other models using a Python library were built to score leads for identifying potential conversions. With good and broad lead-scoring models in place, businesses can optimise their CTI actions on the basis of lead prioritisation and let go of non-prospect leads at the right time to cut costs and enable efficiency. The result of this study reveals that 52 per cent of the sample of 74,779 leads are cold leads and 48 per cent are hot leads that are sales qualified. The leads are qualified using the lead score matrix. This method can aid digital businesses to remove unqualified leads and manage leads better, and therefore improve the quality of the leads sent to clients. This, in turn, will improve conversion rates for individual customers. These increased conversion rates will enhance the business strategy of digital marketing firms.
Keywords: machine learning, supervised ML algorithms, logistic regression, decision tree, hot leads, cold leads
Powerful tools for personalisation: Using large language model-based agents, knowledge graphs and customer signals to connect with users
Seth Earley, Founder and CEO, Earley Information Science and Sanjay Mehta, Engineering Director, AI, eCommerce and Search, Kin + Carta

This paper discusses how large language models’ (LLMs) agentic workflows powering ChatGPT types of applications can use a combination of enterprise data sources to hyper-personalise information at scale for customers or employees. Typical use cases include marketing communications, customer support, content creation and digital assistants. The approaches described are at one level established in theory; however, practical adoption has been challenging and the combination of templated prompts with LLMs and agent call outs to external application programming interfaces and knowledge sources are new. The data sources using these approaches include knowledge, content and transactional data with near real time and real time customer signals. Customer signal data can include first, second or third party data that describes the characteristics of a customer or employee, as well as real time ‘digital body language’ — click paths, searches, responses to campaigns and chatbot dialogues. Two use cases in two industries — automotive and industrial manufacturing — will be detailed to illustrate how the same principles and approaches can be applied in situations that are very different, and how a knowledge architecture combined with retrieval augmented generation (RAG) should be developed and applied. Analytics to monitor outcomes and enable manual and automated course corrections will be discussed. The outcomes are unified and contextualised experiences realising the sometimes ambitious designs of user experience developers. It is easier to storyboard a design than it is to make it a reality. Marketing organisations are more and more responsible for the end-to-end customer journey and experience. However, the customer journey is a knowledge journey. At each step of the process, they are asking questions about the company, product or service. What product and solutions do you offer? Which ones are right for me? How do I choose a particular offering? How do I purchase or procure the product or service? How can I maintain it, and get service or support? How do I get the most from my purchase? What are the options for upgrading or enhancing my solution? These are marketing communications that consist of educating the prospect rather than selling to them. Today’s prospects are empowered with greater information and understanding of offerings and the competition than ever before. Marketing is therefore responsible for helping them make the decision based on information and references that are presented at each stage of the journey.
Keywords: marketing personalisation, customer journey, RAG, retrieval augmented generation, hyper-personalisation, customer experience, generative AI, artificial intelligence, ChatGPT, LLMs, large language models, knowledge management, LLM challenges, LLM solutions, knowledge models, metadata models, knowledge architecture

Additional Practice Paper
How to retain website traffic in a generative AI world
Laura Patterson, President, VisionEdge Marketing and Jacqueline Sinex, Managing Director, WEBii (WebXess, Inc.)

Gartner predicts that generative AI is becoming a substitute for search engines leading to a 25 per cent drop in search engine volume. We live in a digital world, and as such, most organisations rely on search to drive website traffic and ultimately sales. How is search changing? What are the implications and benefits of generative AI in understanding user intent? What can organisations start doing now to retain and increase website traffic in a generative AI world? This paper addresses these questions.
Keywords: website traffic, web traffic, search, SEO, search engines, generative AI, AI, customer behaviour, user intent

Volume 10 Number 2

Editorial
Paul Lima, Managing Partner, Lima Consulting Group, LLC
Practice Papers
Embracing cookieless advertising with AI
Ian Thomas, Founder/Chief Data Officer, Yew Tree Data Consulting

After several delays, the end of unrestricted use of third party cookies is now drawing near, forcing all parts of the digital advertising industry to reconsider how they can drive campaign performance and inventory monetisation without gathering user data. Major browser-makers such as Google are deploying and testing technologies to replicate some of the key capabilities of cookies to enable advertisers to continue working as they have before but the industry needs to find a new way of thinking about driving campaign performance which relies less on the idea of finding the perfect audience and more on implementing a set of connected optimisation techniques that drive performance while maintaining privacy. Fortunately, recent developments in AI (artificial intelligence), especially Generative AI, provide some valuable techniques for achieving this, while at the same time improving the online experience for consumers by serving them more relevant ads that more closely match the context in which they are seen.
Keywords: AI; Google; cookies; creative automation; optimisation; privacy; targeting
How consumer perceptions drive online ratings
Bradley Taylor, Company Owner, GITR Innovation, Insights and Analytics

In marketing, it is a commonly accepted belief that the image of a brand will drive, to some extent, the sales of the brand. The better the image within key areas, the better the sales. In the world of online retailers, the rating is assumed to play an important function, too, when it comes to new purchases. This paper seeks to demonstrate the relationship between ratings and brand image. However, it introduces the nuance that not all brand image perceptions drive ratings to the same degree. We will test these beliefs by using product reviews for snacking goods within the USA. Depending on the statistical results, correlation, causation or matching to randomness, we can strengthen or correct the beliefs that brand image and product experience affect the online rating. We also desire to show which is more robust, the image or experience, and if there is any interference in the brand from the realities of purchasing goods online, such as delivery issues.
Keywords: brand image; emotional connections; online ratings
Using patient lifetime value to future-proof your dental practice
Rosie Pritchett, Dentist, Dental Core Training, South Tyneside and Sunderland NHS Foundation Trust, et al.

Dental practices are a small business. Like any other business, they need cash flow management and financial planning to be viable, if not highly profitable. What a lot of practices may not realise is that they are sitting on a treasure trove of data to be used in more ways than plain accounting and financial forecasting. This paper focuses on longitudinal data, such as the timing of each patient's visits and the value of their treatments since joining the practice. It aims to show how such data can be used by practices to understand their patient base and make plans for the future development of the business. There are plenty of business metrics. Here, the focus is on one metric: patient lifetime value (PLV), derived from customer lifetime value (CLV). CLV is well established in retail and other sectors. This paper shows that patient appointment data can be used in methods adapted from CLV via the bespoke concept of PLV. The paper describes different approaches to calculating PLV, the advantages and disadvantages of each approach and the ways in which they can benefit a dental practice.
Keywords: business improvement; commitment; customer lifetime value; data science; decision making; loyalty; patient lifetime value; statistical models; trust
The three critical thinking skills marketers need to excel at for impactful data storytelling
Caroline Florence, Insight Narrator

Data democratisation means marketers have ever-greater access to data to help inform decisions. No longer in the hands of specialists, data is more readily available to use than ever before. However, despite increasing access to data tools, capabilities within marketing teams have not kept up with the pace of change. Investment in data tools will not lead to better data-enabled decision making unless supported by the right marketing capabilities. Critical thinking skills are often overlooked in data storytelling best practice, with more emphasis on visualisation and creative communications. However, it is the critical thinking skills that enable marketers to extract the value from data to ensure the stories both stand up to scrutiny and stand out from the noise. This paper focuses on the needs to build capabilities in three core areas: identifying the ‘so what?’ insights from the data, distilling insights into a few salient points of view and clarifying the ‘now what?’ recommendations.
Keywords: analysis; critical thinking; data literacy; data storytelling; insight; marketing effectiveness; skills
Research papers
Overcoming analysis paralysis in bulk: Efficient methods for extensive key driver analyses
Michael Dupin, Adjunct Professor, Merrimack College and Sophia Tannir, Data Scientist

In the rapidly changing field of market research, identifying the key drivers of customer relationships is essential for enhancing business strategies and customer satisfaction. This paper explores the application of driver analysis, a critical methodology that assists businesses in pinpointing the crucial factors influencing customer behaviour and satisfaction. By effectively distinguishing impactful elements from less relevant ones, this technique enables more precise decision making and strategy development. The core of this paper introduces an innovative method for categorising drivers into primary and secondary groups, simplifying the complex data landscape and focusing on the most influential factors. This new grouping method significantly reduces the analytical complexity typically associated with traditional models, making the insights more accessible and actionable for businesses. A case study utilising the Kiwis Count survey — a comprehensive public service evaluation in New Zealand — serves to illustrate this methodology in a real-world context. By applying the proposed method to this survey, the paper provides a detailed examination of how various demographic groups perceive public services and what drives their satisfaction. The results reveal distinct patterns in how different demographics value aspects of service delivery, from staff competence to trust and transparency. By focusing on the most impactful drivers, organisations can allocate resources more effectively, enhance customer experiences and ultimately achieve greater customer loyalty and success.
Keywords: Key Driver Analysis (KDA); categorisation; data-driven strategy; enhanced data interpretation; noise reduction
Machine learning and AI in marketing analytics: Leveraging the survey data to find customers
David Fogarty, Associate Professor, National University and Xinlei Cui, Graduate Research Assistant, New York University, Stern School of Business

The field of marketing analysis in the digital era faces numerous challenges. Despite the availability of vast amounts of structured and unstructured data, practitioners have yet to fully harness the potential of machine learning models. This paper addresses this gap by investigating how to find targeted customers and expand the emerging market by implementing machine learning models to process survey text data and provides empirical evidence through model evaluation experiments. The research problem focuses on demonstrating the effectiveness of machine learning and AI models in optimising value creation and enhancing competitive advantages in marketing practices. The paper employs mixed methods and presents experimental results, leading to conclusions highlighting the benefits of improving data quality to strengthen the performance of machine learning models. This research also provides insights into model selection and offers a foundation for future researchers and marketing analysts to interpret and evaluate machine learning models effectively by multiple efficient metrics.
Keywords: analytics; artificial intelligence; automated classification; consumer targeting; machine learning; marketing
Crossed signals: The negative effects of manufacturer warranty length on brand share in an independent retail channel
Robert J. Fisher, Alberta School of Business Research Chair in Marketing, University of Alberta, Yu Ma, Associate Professor of Marketing & Bensadoun Faculty Scholar, McGill University, and Barry Scholnick, Professor, Alberta School of Business, University of Alberta

Manufacturers’ attempts to signal quality via longer warranties can have unintended effects on independent retailers’ behaviour in categories where extended service contracts (ESCs) are offered. The results of this research indicate that independent retailers are likely to promote brands with shorter rather than longer manufacturer warranties to maximise the likelihood of selling an ESC as an option. This research provides an explanation as to why manufacturers find it difficult to use warranty length as a signal of product quality. This research is based on a field study of brand shares of household appliances and consumer electronics by a major retailer over a ten-year period. The results have important implications for manufacturer and retailer marketing strategies.
Keywords: extended service contracts; extended warranties; quality signalling; retailer incentives; warranty length

Volume 10 Number 1

Editorial: What customer experience professionals should be doing as the battle lines between capitalism and individual rights continue to take shape
Simpler is better
Denis Malin, Editorial Board Member
Practice Papers
The evolution of digital marketing in the era of AI
Kelly Cutler, Business Leader, Author and Educator, Northwestern University

This paper explores the far-reaching impact of artificial intelligence (AI) on the evolving digital marketing landscape. In the digital age, marketing is undergoing a profound transformation led by data privacy and changing technology, with a focus on the benefits and the risks of adopting AI. Increased productivity and the value of AI must be balanced with careful ethical, legal and privacy considerations. Technologies like third party cookies, once the cornerstone of digital advertising, are being retired and replaced. Simultaneously, ad blockers, wielded by users seeking respite from intrusive advertisements, are reshaping the marketing paradigm. Marketers are at a crossroads. Insights and ideas will be explored pertaining to how AI is poised to reshape the way that marketers do business. While emphasising the role of AI in aspects of marketing such as customer engagement, personalisation, marketing automation, content curation, predictive analytics, campaign creation and more, there is also the need for oversight, management and responsible deployment. This can be accomplished by combining automation and technology with human intervention and direction. Careful examination of the potential benefits and the multifaceted risks posed by AI will define how marketers move into a privacy-centric digital future. This paper delves into the rapidly evolving digital marketing ecosystem as it adapts to changing user behaviours, tightening regulations and new technologies.
Keywords: digital marketing; data privacy; transparency; artificial intelligence and marketing; generative AI; AI and digital marketing
Customising generative AI: Harnessing document retrieval and fine-tuning alternatives for dynamic marketing insights
Dakota Crisp, Senior Manager of Data Science, Jacob Newsted Data Engineer and Data Scientist, Brendon Kirouac, Data Scientist, Danielle Barnes Senior Director of Data Science, Catherine Hayes Senior Director of IT, and Jonathan Prantner Chief Analytics Officer, OneMagnify

This study delves into the transformative impact of leveraging large language models (LLMs) in marketing analytics, particularly emphasising a paradigm shift from fine-tuning models to the strategic application of document retrieval techniques and more. Focusing on innovative methods, such as retrieval augmented generation and low-rank adaptation, the paper explores how marketers can now activate against vast and unstructured datasets, such as call centre transcripts, unlocking valuable insights that were previously overlooked. By harnessing the power of document retrieval and adaptation, marketers can bring their data to life, enabling a more nuanced and adaptive approach to understanding consumer behaviour and preferences. This research contributes to the evolving landscape of applied marketing analytics by demonstrating the efficacy of document retrieval in enhancing the utilisation of LLMs for dynamic and data-driven marketing strategies.
Keywords: generative AI; marketing analytics; call centre; natural language processing; document retrieval techniques; retrieval augmented generation; low-rank adaptation
Improving voice of the customer analysis with generative AI
Jim Sterne, Applied Technology Evangelist and Thomas H. Davenport, Professor, President’s Distinguished Professor of Information Technology, Babson College

This paper explores the integration of generative artificial intelligence (GenAI) in voice of the customer (VoC) analysis to provide deeper understanding of prospects and customers. GenAI has enormous potential to enhance customer satisfaction, refine products and services and improve the customer experience. This speculative paper illustrates how GenAI can keep pace with increasing customer expectations and the volume of feedback by uncovering nuanced sentiments, trends and customer needs through context comprehension and its conversational query capabilities. The paper explores the power of GenAI in VoC analysis for improving customer satisfaction, accelerating troubleshooting and resolution and upgrading products and services. Additionally, this paper addresses the role of GenAI in advanced communication routing, agent support, multilingual support and sentiment analysis, showcasing its ability to provide comprehensive and context-aware insights.
Keywords: generative AI; GenAI; voice of the customer; VoC; customer satisfaction; sentiment analysis; customer feedback analysis
The recipe for success in creating frictionless customer journeys
Stephanie Burton, Expert Solution Consultant, Data & Insights, Adobe

Customers now demand seamless experiences across all points of contact. To deliver on this expectation, organisations need a reliable marketing architecture for accurate omnichannel customer journey analysis. This allows them to identify and eliminate friction points that hinder customer satisfaction. However, disparate channels of data often utilise different customer identifiers, creating a challenge in unifying data for comprehensive analysis. By focusing on a strong customer identity strategy and leveraging technological advancements to seamlessly combine data at the individual level, organisations gain an advantage in crafting consistent and frictionless customer experiences. This approach follows a specific recipe, with key steps to identify and remove friction points. Professionals involved in creating or measuring omnichannel experiences will find valuable insights and practical tips within this paper, along with learnings from renowned customer-centric companies like Uber, Netflix and Amazon. Delivering consistently exceptional customer experiences allows companies to command premium prices, build long-term brand loyalty and solidify their reputation as truly customer-focused organisations.
Keywords: customer experience; customer journey; customer identity; frictionless; seamless; omnichannel analysis; cross-channel analysis
The power of clarity: Understanding how the effective use of data storytelling can improve the decision-making process
Giovanna Fischer, Independent Consultant and Founder, Escola de Insights

This paper discusses the importance of communication in the data field and its potential to improve decision-making processes in corporations. This study adopts an approach intended to understand the current state of the art in the field. The paper proposes a framework for developing data storytelling as a tool that supports data-driven decisions.
Keywords: data; communication; storytelling; business intelligence; data storytelling; data-driven decisions; management
Strategies to mitigate hallucinations in large language models
Ranjeeta Bhattacharya, Senior Data Scientist, BNY Mellon, AI Hub

In the world of enterprise-level applications, the construction and utilisation of large language models (LLMs) carry a paramount significance, accompanied by the crucial task of mitigating hallucinations. These instances of generating factually inaccurate information pose challenges during both the initial development phase of LLMs and the subsequent refinement process through prompt engineering. This paper delves into a variety of approaches such as retrieval augmented generation, advanced prompting methodologies, harnessing the power of knowledge graphs, construction of entirely new LLMs from scratch etc, aimed at alleviating these challenges. The paper also underscores the indispensable role of human oversight and user education in addressing this evolving issue. As the field continues to evolve, the importance of continuous vigilance and adaptation cannot be overstated, with a focus on refining strategies to effectively combat hallucinations within LLMs.
Keywords: LLM; large language model; hallucination; prompt engineering; RAG
Research Papers
Predicting maintenance costs of an IT system using AI models
Nathan Bosch, Machine Learning Engineer, Lyft, Emmanuel Okafor, Postdoctoral Researcher, SDAIA-KFUPM Joint Research Center for Artificial Intelligence, King Fahd University of Petroleum and Minerals, Marco Vriens, CEO, Kwantum and Lambert Schomaker, Professor in Artificial Intelligence, University of Groningen

Predictive maintenance is a maintenance policy where the goal is to detect potential future maintenance risks in a system so that the maintenance process can be optimised before system faults occur. This paper describes a deep learning model that does not require domain expertise. Deep learning approaches have several benefits over explicit statistical modelling: (1) they require far less domain-specific knowledge; (2) if the underlying data-generating mechanism of assets changes, a deep learning model would only need to be retrained to learn these new changes; (3) they can capture non-linear and complex multidimensional relationships; and (4) they may outperform rule-based or statistical methods. The paper describes how the model predicts maintenance-relevant events, along with the cost of the upcoming event and the time when it will happen. The paper describes the use of a long short-term memory architecture for our deep learning model. By doing so, the cost values represent a real, quantitative value of the potential maintenance costs in the future of an asset. Event, cost and time prediction are all achieved with high accuracy. This allows for the development of maintenance solutions without the initial high degree of domain process knowledge required. The artificial intelligence model can be used to raise an alarm when the cost values exceed some threshold, when the frequency of high-cost events increases significantly over the lifetime of an asset, or when the expected cost exceeds the cost of maintenance.
Keywords: predictive maintenance; deep learning; long short-term memory; LSTM; cost prediction; time prediction
Navigating compliance and regulations in marketing analytics: Upholding ethical standards and consumer trust
Animesh Kumar Sharma, Research Scholar, and Rahul Sharma, Professor, Lovely Professional University

This paper delves into the multifaceted marketing analytics compliance and regulation landscape across diverse business sectors and legal frameworks. It discusses a spectrum of norms with respect to overseeing data collection, processing and utilisation in marketing endeavours. Stringent global laws govern the handling of personal data, necessitating strict adherence. The paper scrutinises pivotal compliance elements like consent, transparency and data security alongside pivotal legislation like the California Consumer Privacy Act and the General Data Protection Regulation. It assesses the implications for marketing analytics, emphasising rights regarding personal data access, erasure anonymisation methods and ethical data use. Non-compliance repercussions, encompassing legal and financial risks and reputational harm, are highlighted, as many industries are facing distinct regulatory challenges. The paper details the essential components of policies, training, monitoring and enforcement that are crucial to ensuring marketing compliance. It stresses the role of technology, advocating for marketing compliance software to streamline processes, monitor compliance and adapt swiftly to regulatory shifts. It elucidates the collaborative nature needed within marketing teams to achieve effective compliance management. The conclusion highlights how compliance software helps with regulatory updates, data privacy, monitoring and content assessment. This paper emphasises the dynamic nature of marketing analytics compliance, urging vigilance with regard to legislative alterations and technological advancements. The paper provides a comprehensive insight into managing compliance challenges in this evolving field while upholding ethical standards and fostering consumer trust.
Keywords: marketing analytics; compliance and regulation; data protection legislation; non-compliance consequences; data compliance; risk management
Book Review
Search Marketing: A Strategic Approach to SEO and SEM, by Kelly Cutler
Reviewed by Shashi Bellamkonda, Principal Research Director, Info-Tech Research Group, and Adjunct Professor, Digital Marketing, Georgetown University