Consumer Behaviour and Analytics

Data Driven Decision Making

Andrew Smith

• Publisher: Routledge
• Year: 2019
• ISBN:
  9780429953354
• Edition: 1
• Topic: Business
• Subtopic: Business General
• Index: Business

Consumer Behaviour and Analytics provides a consumer behaviour textbook for the new marketing reality. In a world of Big Data, machine learning and AI, this key text reviews the issues, research and concepts essential for navigating this new terrain. It demonstrates how we can use data-driven insight and merge this with insight from extant research to inform knowledge-driven decision making.

Adopting a practical and managerial lens, while also exploring the rich lineage of academic consumer research, this textbook approaches its subject from a refreshing and original standpoint. It contains numerous accessible examples, scenarios and exhibits and condenses the disparate array of relevant work into a workable, coherent, synthesized and readable whole. Providing an effective tour of the concepts and ideas most relevant in the age of analytics-driven marketing (from data visualization to semiotics), the book concludes with an adaptive structure to inform managerial decision making.

Consumer Behaviour and Analytics provides a unique distillation from a vast array of social and behavioural research merged with the knowledge potential of digital insight. It offers an effective and efficient summary for undergraduate, postgraduate or executive courses in consumer behaviour and marketing analytics or a supplementary text for other marketing modules.

Chapter 1

An introduction to consumer analytics

Introduction

This chapter introduces some key issues and concepts that lie at the heart of an analytics-driven approach to consumer marketing. The chapter reviews the various types of inquiry and inference, cause-effect, descriptive and predictive analytics. It also reviews the various forms of purposive data capture used to augment analytics insight (e.g. surveys, ethnography, neuroscience etc.). A key premise is that marketing education has a greater chance of relevance if aligned with (or cognizant of) practice; this underpins the rationale for starting with issues central to behavioural analytics insight (the ‘engine room’ for contemporary consumer marketing practice). Those who have not read the preface should do so now, since it sets out the overall approach to the topic and the ethos of Consumer Behaviour and Analytics (CB&A).

The context of contemporary marketing

Consumers and marketing have changed radically in the last decade or so. Consumer marketing is increasingly analytics-driven and data-driven. This has become the new normal. Seemingly, practitioners are in a position to ‘know’ more about what customers actually do than ever before. This knowledge is increasingly reliant on insights generated by algorithms leading to ‘automated marketing’. However, humans oversee the configuration of data processing, the interpretation of output and the subsequent tactical and strategic decisions that this data processing informs. Figure 1.1 exemplifies the mindset required to interact with this new reality.

Figure 1.1The contemporary mindset and skill set

The functional boundaries between ‘technical’ aspects of data science and the ‘traditional’ skills of the marketer are blurring (if not within individual actors’ skill sets then within organizations or teams). The modern marketer needs to be able to see things through the lens of the data scientist (see, for example, Provost and Fawcett 2013). The functions cannot be divorced or discrete, as depicted on the left of the diagram. At the very least they are required to overlap, as the upper right Venn illustrates. Data scientists need to be able to think like marketers and marketers like data scientists. So, the lower Venn is the ideal; a blending of the mindset and even skill set. Commercial organizations (and not-for-profit ventures) need to ensure that these functions are blended with cross-disciplinary teams and individuals. If they are siloed and compartmentalized then the outcome is bound to be sub-optimal. For example marketers will pose research questions which are not viable and interpret data and output in a way that is questionable (or just plain wrong). Conversely, data scientists will tend to pursue actions and projects that might not lead to actionable marketing or posit research questions that already have answers (via the back catalogue of academic/generic consumer behaviour and marketing research). So, these human intermediaries require sources of extant knowledge (regarding marketing and consumer research), such as that summarized in CB&A, and insight into analytic thinking (as discussed in this and subsequent chapters).

Why data-driven?

There are distinct imperatives that explain how and why data-driven marketing has come about:

Operational imperatives

Why ignore all that data?

Data capture has never been easier. That is, transactional and other behavioural indicator data, or verbal and written (sentiment) data. It would seem perverse to ignore that data. It has obvious self-evident potential. Indeed, data streams can only get richer. Amazon Echo or Google Home point the way to the future, as does the internet of things (the network of smart devices that can autonomously communicate – e.g. smart refrigerators). There are more windows on our thoughts and our behaviour than ever before in the history of humankind. We used to leave a few pot fragments behind us for archaeologists to dig up; now we have a gargantuan data library on everyone.

Ostensibly the mobile phone or tablet is there for your convenience, to help you live your life; this is true. But it is also telling people what you’re doing, where you are, what you’re buying, who you know and many other things, too numerous to list. It is performing these functions ceaselessly and continually. Google is a data management company that makes its money from targeted marketing communications (MC); it sells the insight to other businesses. It epitomizes the new economy. It provides services free at the point of use in return for you ‘paying’ with data. It is truly global, truly cross-national.

Cost

Data capture is less costly than it was. Moreover, many businesses capture data as a matter of routine in order to operate their services. The data is there. However, the notion that data capture and use is cost-free is a fallacy. Additional costs include the storage of data, high performance computing (HPC), the expertise required to process the data (in-house or outsourced), the infrastructure and organizational changes to ensure that the data leads to strategic usage and data-driven decision making. Nonetheless, analytics-driven marketing appears to be cost-effective.

Analytical imperatives

Let’s find out what people actually do before we attempt to explain it

The late Andrew Ehrenberg was critical of (academic) marketing’s deductive approach. Models and ideas about consumer behaviour can be generated ad nauseam. ‘Models without facts’ as he put it (Ehrenberg 1988). This is possibly a questionable label but it exemplifies a period when academic and practitioner marketing research often tried to explore and explain behavioural drivers and antecedents, or rely on reports of behaviour (via surveys) before conducting so-called descriptive research on what people are actually doing. If we’re not sure of what people are doing and how they are behaving, should we really be asking the question ‘why are they doing it?’

His work and the work of fellow travellers and associates remains singular and influential. It led to some rare ‘laws’ of consumer behaviour and the methods and analytical model developed have been widely used in commercial market research. However, this body of work did not spawn a flurry of data-driven, more inductive work. This is explained by inertia and prevailing orthodoxy dating back to the genesis of scholarly and commercial consumer behaviour research. A brief account of these origins is essential; it is executed in Chapter 4 (the most suitable location in terms of this book’s narrative).

Pre- and post-war marketers used to rely on anecdote and instinct about what the customer required and desired. However, a series of unexpected product¹ failures provoked a rethink of their approach to customer insight and new product development and targeting. This culminated in customer segmentation and categorization exemplified by the Values, Attitudes and Lifestyle approach (VALs) in the 1970s. Suffice to say that the approach attempted to predict consumer preferences by psychographic measurement. This means that consumers were categorized according to their psychology, aspirations, their intellectual and financial resources and crucially their motivations. The VALs method was influential and its impact still reverberates. Segmentation is still core to modern marketing (many things need to be aimed at groups or blends of groups not individuals) although the starting point is increasingly transactional/behavioural data. Questions remain over the ability of psychographics to predict behaviour. In the age of analytics-driven consumer research, psychographics have found a role in explaining observations on behaviour or providing evidence to support the conclusions drawn from behavioural data.

Data identifies individuals/households

This might seem normal now but it was a long time coming. Financial services have always required identity. However, prior to the loyalty card or online purchase many retailers and service providers struggled to link purchases with people. Knowing who bought what is taken for granted now. This allows marketing communications to target individually rather than through channels or media (i.e. like traditional advertising). Most products are still not individualized, communications increasingly is.

The upshot is that retailers and those last in the chain/closest to the consumer are the new ‘druids’. If you are further upstream (i.e. if you make fast-moving consumer goods – FMCGs) then you need the entity who sells on to the consumer to help you understand behaviour. In past decades you would have relied on market research companies and surveys for insight. Now, it’s the retailer or internet service provider. Online and offline retailers and service providers are simultaneously in the data business in a big way and can sell the insight back down the supply chain.

Data and insight

There are numerous ways of classifying data (...

Table of contents

• Cover
• Half Title
• Title Page
• Copyright Page
• Contents
• Figures
• Tables
• Preface
• Acknowledgements
• Chapter 1 An introduction to consumer analytics
• Chapter 2 Purchase insight and the anatomy of transactions
• Chapter 3 Web and social media activity
• Chapter 4 Extant research and exogenous cognition
• Chapter 5 Elemental features of consumer choice: Needs, economics, deliberation and impulse
• Chapter 6 Perceptual and communicative features of consumer choice
• Chapter 7 Individual and social features of consumption
• Chapter 8 Knowledge-driven marketing and the Modular Adaptive Dynamic Schematic
• Index