Architecting experience a marketing science and digital analytics handbook

Contents Introduction ix The Foundations of Personalization 1 1.1 The New Business Value: Analytics Strategy, Technology, Science & Art 21 2.2 The Changing Nature of Marketing Data 322.3

Digital Analytics Handbook

Vol 1: Architecting Experience:

A Marketing Science and Digital Analytics Handbook

by Scot R Wheeler

Library of Congress Cataloging-in-Publication Data

Names: Wheeler, Scot R.

Title: Architecting experience : a marketing science and digital analytics handbook /

Scot R Wheeler, Medill-Northwestern University, USA.

Description: | Series: Advances and opportunities with big data and analytics; 1 |

Includes bibliographical references and index.

Identifiers: LCCN 2015028389| ISBN 9789814678414 (hardcover : alk paper) |

ISBN 9814678414 (hardcover : alk paper) | ISBN 9789814725651 (softcover : alk paper) |

ISBN 9814725651 (softcover : alk paper)

Subjects: LCSH: Communication in marketing | Digital media.

Classification: LCC HF5415.123 W48 2016 | DDC 658.8/02 dc23

LC record available at http://lccn.loc.gov/2015028389

British Library Cataloguing-in-Publication Data

A catalogue record for this book is available from the British Library.

Copyright © 2016 by World Scientific Publishing Co Pte Ltd

All rights reserved This book, or parts thereof, may not be reproduced in any form or by any means,

electronic or mechanical, including photocopying, recording or any information storage and retrieval

system now known or to be invented, without written permission from the publisher.

For photocopying of material in this volume, please pay a copying fee through the Copyright Clearance

Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, USA In this case permission to photocopy

is not required from the publisher.

In-house Editor: Philly Lim

Typeset by Stallion Press

Email: enquiries@stallionpress.com

Contents

Introduction ix

The Foundations of Personalization 1

1.1 The New Business Value: Analytics

Strategy, Technology, Science & Art 21

2.2 The Changing Nature of Marketing Data 322.3 The Fundamental Analytics Architecture:

The Applied Digital Analytics Playbook

3.1 ADAP Section One: Problem Definition 503.2 ADAP Section Two: Solution Definition 55

The Changing World of Owned Media 71

4.1 Web Architecture & Web Data Collection 73

Earned Media: Organic Social & SEO 115

5.4 Inbound Organic Social

6.4 DSPs and Programmatic Real-time

Data Management, Models, and Algorithms 199

8.1 The Applied Digital Analytics Playbook

8.2 Data Mining & Data Visualization 2058.3 Predictive Analytics & Machine Learning 208

The Cultural and Organizational

9.2 The Information Society: Media

9.3 Organizational Change for Effective

About the Author

Scot Wheeler is a leader in tal analytics delivery, overseeing

digi-a tedigi-am which develops consumer intelligence, prospect conver-sion propensity scoring, cross-channel performance evaluation, environmental trend analysis, testing, targeting and optimiza-tion, and predictive modeling for budget allocation and response forecasts

He is also an adjunct lecturer in Northwestern University’s Master’s Degree program in Integrated Marketing Communications, where he teaches Digital Analytics and Statistics

Scot received his MBA in Strategy, Finance and Marketing from Northwestern University’s Kellogg School of Management Prior to his current roles, Wheeler was Group Director of Marketing Science for the digital agency Critical Mass Before that, he ran product development, marketing and sales for the social media analytics platform Evolve24 Wheeler’s professional background spans a variety of technology, consulting and agency roles From his start in software development, Scot’s 20 years of experience at the inter-section of technology and marketing includes work with Yahoo!, GE, Electronic Arts, AT&T, MasterCard, State Farm, USAA and HP

Introduction

In the Integrated Marketing Communications approach taught at Northwestern University, the con-sumer is placed at the center of all marketing prac-tice Unfortunately, this customer-centricity is not always as common in practice as it should be in the real-world of digital marketing In actual practice, brands often place concern for awareness of their message at the center of their marketing practice, and much “digital strategy” is simply an effort to ensure consistent branding and “messaging” across digital channels However, any digital marketing practice that is focused on brand message and struc-tured primarily by channel and function (the paid media team, the social media team, the web team) will typically fail to create a truly integrated and rel-evant experience as a consumer moves across digital channels The capability to capture and use data from any consumer’s digital engagement, and the growing expectation for content personalization that consumers have as a result (beginning with each user’s Amazon and Netflix experience), means that the disconnection of data across channels will be felt

by the user and will adversely impact their ence with the brand Conversely, the effective collec-tion and connection of data across channels will play

experi-a significexperi-ant role in creexperi-ating experi-and mexperi-aintexperi-aining brexperi-and relationships with the digitally embedded consumer

Thus, the question this book sets out to answer is the

biggest question facing digital marketers today: how

do I deliver content and experience around my brand that is relevant enough to drive engagement in the user’s current context? The quick answer to this

is of course through the application of data and lytics to drive highly relevant, contextual targeted content and adaptive experience, but since this

ana-answer is not as easy to achieve as it is to say (and it

is a mouthful), this book has been designed to help you develop the understanding and skills required

to make this happen

The path to delivering relevant, contextual and even adaptive digital experiences is not one for the mar-keter to walk alone, and this book will explore the relationships that must emerge between marketing, technology, research and operations to bring about truly effective 21st digital experience delivery At the end of the day however, the envisioned reader of this book has the strongest interest in the marketing per-spective on these conversations, with a 21st century marketing mindset that understands marketing as innovation and technology driven customer-centric relationship building for long-term customer value versus message dissemination for the masses myopi-cally focused on driving business transactions above all else

Digital communications long-ago turned mass-media

on its head, a fall from which mass-media as the top effective communications form will never recover In

a world with a seemingly infinite amount of content and scores of methods for consuming that content, communication today is about appealing to individu-als, person by person, and appeal requires relevance

in context In any conversation, delivering relevance

in context requires understanding the person you’re speaking with This is true for digital marketing as well

This book will focus on the impact of data and nology on marketing both within businesses and for consumers as well It will allow you to guide your organization in a necessary process of continuous evolution to effectively collect and use the right data, analytics, technology platforms and algorithms to achieve valuable outcomes

tech-This evolution is built around a six-stage process which is facilitated through an Applied Digital Ana lytics Plan (ADAP), which is introduced in Chapter 3:

1 Define the problems that data can solve

2 Identify sources of data (existing and potential)

3 Collect, manage and analyze data

4 Overcome organizational and cultural inertia

5 Apply data and analysis to solve the problems

6 Evaluate the outcomes

This book will explain how evolution within the cess detailed above is achieved through the follow-ing activities:

pro-1 Data-driven problem identification and ented strategic communications design (design research)

data-ori-2 Strategic alignment of customer and business objectives

3 KPI development and documentation from objectives

4 Marketing channel digital data collection strategy and implementation

5 Multi-channel data integration

6 Testing and optimization across all channels

7 Integrated planning models and performance reporting

8 Predictive analytics and adaptive digital ence enablement

experi-These activities build upon and interact across each other and are discussed in detail through the remaining chapters, with the first several chapters focusing on the specifics of data and collection and analysis for owned media, earned media and paid media channels, and with the later chapters focused on integrating data across channels and applying it to continual optimization of results in omni-channel engagement with customers through applied analysis and technology Before proceeding into these activities however, it is worth our while to begin with a deeper examination of the question of relevance in digital marketing What is relevant to our customers at any given point in time? How do we know? And how do we take that knowledge and use

it to deliver better experiences that in turn yield ter results?

The Foundations of Personalization

As the digitization of life proceeds with seemingly exponential progress, we continually find ourselves in

an ever changing cultural landscape, where each day the amount of new information recorded is greater than all of the world’s recorded information prior to the digital age, where the average citizen of nearly all nations has unprecedented access to knowledge, entertainment and opinion, and of course where those same citizens are exposed to hundreds of adver-tisements a day on screens both stationary and mobile

We find ourselves in a world where social life ingly means digital life, and where success in busi-ness and marketing require advanced capabilities to access and interpret data In short, we’ve crossed the horizon into a world where it can be argued that culture (i.e work, arts and entertainment, customs, habits and pastimes) has largely become a product

increas-of information technology, and that ingly, information has become the core of culture in the developed world

correspond-Living in a technologically and digitally driven world means living with constant change In the 20th century, the economic and cultural base of the developed world transformed from being agriculturally (and

Chapter

ONE

land) based to manufacturing and technology based over half a century, with the Second World War finally cementing the rise of the technocrat over the gentry

as culture’s new elite The transformations that have occurred as developed culture has then shifted from capital intensive analog technology to skill and infor-mation intensive digital technology have come more and more rapidly, and many businesses are still trying

to catch up with the changes in technology and society that have come at them over the last two decades

The continual development of information ogy and applications arises from a human drive to continually expand both our knowledge and conveni-ence, a drive that has been at the core of advancement

technol-in science and technology for centuries In the middle

of the 20th century, the study of such advancement in communication technology was taken up by a profes-sor of communications at the University of Toronto

in a way that forever changed our understanding of the relationship between media and society

Given the breadth and depth of any individual’s exposure to media today, it seems inconceivable that the phrase “Media” and the concepts associated with

it would at one time have required an introduction and development within popular thinking, but in fact there was such a time not that long ago, and the man who made the introduction was Professor Marshall McLuhan

McLuhan’s 1965 book Understanding Media: The

Extensions of Man introduced the concept of Media

where previously there had simply been notions of independent communication technologies such as

‘the press’, ‘television’ and ‘advertising’ which were

recognized to synthesize into a single unit in practice, but were nonetheless typically evaluated individually with regards to their impact on culture

McLuhan observed that these and many other mation and communication technologies and prac-tices not only work together and proceed from one another, but that in doing so they actually extend the perceptual power of individuals and mediate com-munication and thinking in ways that significantly influence culture and human affairs; thus his appli-cation of the term Media to these mediating ‘exten-sions of man’

infor-Perhaps the most lasting convention introduced by McLuhan (and the most useful for the subject of this book) is the notion of “hot” and “cool” mediums, or elements of culture Very few people wonder who should be credited for coining the use of ‘cool’ in social context, e.g a ‘cool’ new band or the ‘cool’

kids at school, since the expression seems to have always been a part of the vernacular Equally, the idea

of a ‘hot’ new sound or a person with a ‘hot’ body are commonly used in American parlance without con-sideration of origin Today, these notions of “cool”

and “hot” seem natural in the ways they are applied, but it was just as recently as the beginning of the Cold War that Marshall McLuhan observed that different media exerted different influences on people’s per-ceptions and engagement with those media, and clas-sified those media into “hot” and “cold” categories

McLuhan’s theory has held up incredibly well into the early 21st century media environment, and still provides an excellent framework for understanding the influence of mediated information on culture,

and the countervailing response of culture in the subsequent development of new information tech-nology Everyone wants their content to be “hot”,

“sticky” and “viral”, and these metaphors for tion owe much to McLuhan’s understanding of how people engage with messages in media McLuhan’s thinking is worth consideration by anyone interested

informa-in the science of marketinforma-ing communications, and will be explored in detail in Chapter 9 But to begin with, we’ll first boil down the application of McLuhan’s ideas to marketing to a fundamental principle — the message is either relevant in the receiver’s current context, or it is not

1.1 The New Business Value: Analytics Increase

Relevance

In 20th century marketing, messages from the brand were thought to be uni-directional, and consumer interaction with these messages was thought of as passive receipt and absorption of the message Digital communications turned that idea on its head Digital communications are multi-directional, as consumers have the capability to respond and interact directly to and/or about the brand through both shared and owned media channels This interaction between brand and consumer or by consumers about/around

a brand in all digital channels is commonly referred

to as “engagement”, and effective engagement is the objective of all digital marketing Engagement result-

ing from digital marketing may be as simple as ing on a link or liking/sharing content (and thus passing it along through a network), or it may be more involved, such as returning to a brand’s digital experience, going deeper into content and tools,

click-adding their own new content, attending an event, completing a form, making a purchase or sharing a referral

When it comes to engagement, one rule stands out

over all others: relevance drives results We are no

more likely to engage with activities and tions that have no appeal or value to us in digital than we are in the real world In fact, digital gives us much better ways than we have in real-life to filter out irrelevant and uninteresting content Digital also gives us much more content to filter than we face in real-life, which for most users creates a high thresh-old between what is potentially viewable for them and what actually elicits engagement

conversa-This returns us to the largest problem that the digital

marketer faces today: how do I deliver content around

my brand that is relevant enough to drive engagement

in the user’s current context? The answer to this is of course through the application of data and analytics

to drive highly relevant, contextual targeted content and adaptive experience.

The figure below tells a story about the rise of vance in digital communications channels over the past two decades

rele-As we see in Figure 1.1 in the earliest days of stream digital communication, marketing was con-ducted through email, display advertising and websites

main-Of course, the dawn of email marketing brought the immediate dawn of spam, since in the early days, sim-ply having an email address qualified you for targeting

by anyone who could get that email In early digital marketing, display advertising was made more relevant than email by virtue of some occasional effort to align

advertising with the content of the page on which it was advertised, at least by marketers who didn’t want to throw their digital budget down a black hole And marketing on websites was the most relevant content on the web for those exposed to it since they were qualified to see it based

on having sought it out

Search marketing arrived on the scene in earnest in 1999

to take the top position for delivering relevant marketing content through algorithms that matched expressed inter-est or intent with digital content results Being based on explicit interest cues and algorithmic matching of content

to those cues, search has remained toward the top of the relevance ladder ever since, being surpassed recently only

by content that has interest cues, algorithmic content geting and memory of user history in a more specific con-text than the blank page of a new search However, Google’s interest in having more user context data from across all platforms will likely see the return of “predictive”

tar-search (exhibited currently in the Google Now application)

Figure 1.1

The Rise of

Relevance

as the most relevant form of content delivery around any users immediate content needs in context

What Figure 1.1 shows us — beginning with search then extending to social content, social ads and eventually lifting all boats — that increasing data about context and algorithms for matching content with context are helpful in delivering relevance But where is this data, and how do we use it to discover what is relevant, and apply that understanding to driving results?

1.2 Introducing the “Demand Chain”

A company’s supply chain and the practice of supply chain management is critical to that company’s ability

to produce and deliver its goods to its customers An organization’s supply chain is the linkage of material, processes and people that proceeds from the initial procurement of the raw materials needed to product

a product, all the way through production to the final delivery of that product to the end customer Without careful management of the supply chain, among other problems, materials required for production might not be available when needed, warehouses could be overflowing with un-needed raw material or finished product, or more product than needed could be produced to sit on shelves in stores without buyers Supply chain management begins with pro-jections around the demand for products, then puts into motion all of the gears required to produce and then distribute the right amount of product in the right places at the right time based on that demand

While a complex array of production inputs, outputs and logistics provide the material for supply-chain management, it is the projection or forecast of

demand from the market that provides the impetus

If the forecasted demand for product is incorrect, then the best a highly effective supply-chain manage-ment process can do is try to adjust to the estimating error once it becomes apparent

Creating perfectly precise demand forecasts is nearly impossible, so producers of goods and services have options If the good is something packaged and sold

in a store, then a target for sales is set, and tion (in actual units or in production cost-to-profit ratio for something like software) is run to that target

produc-Once produced, the product is put up for sale in stores and/or online Once the product is up for sale

in some location, the supply chain has done its job until more product is needed — which can be quickly for made-to-order goods or services However, once the product is up for sale, the product has entered the “demand chain” — a less recognized and less understood area of the marketing equation

If the supply chain is the process that pushes a

prod-uct out to where it can be bought, the demand chain

is the counterpart process through which the

cus-tomer ultimately pulls the product into their basket

Similarly, if the supply chain is the process that duces product supply, then the demand chain is the process that produces demand

pro-1.3 The Customer Journey

While marketing strategy and marketing tions tries to understand and tap into the demand chain, it is a common and disadvantageous mistake to think that marketing drives the demand chain, and is

communica-the source of product demand Demand for products starts with a need or desire in a consumer It is very true that advertising applies psychology to evoke needs and desires, but it is also true that most goods also fulfill an actual need or demand Advertising can

be used to cultivate the perception of a certain brand

of clothing as more sexy or sophisticated than another, but the demand for clothes was already there Demand also requires a stimulus to buy — I may be exposed to advertising that guides me to fully perceive a brand as tied to some characteristic or quality, but perception

is not purchase To become a customer, there must be some kind of trigger prompting me to buy some-thing in that product category Only then will my pre-established perceptions of the characteristics of various options begin to matter

So, the demand chain begins with a “trigger” to sider a purchase This can be as simple as running out of toilet paper, or as complex as recognizing the need to determine a care plan for an aging parent

con-The most traditional concept of the consumer path

to purchase (along the demand chain) envisioned marketing as a funnel that brought the consumer from awareness, to interest, to desire and then finally

to action, or purchase

This way of thinking of customer engagement has guided generations of marketing planners and mar-keting campaigns, with no expense spared on aware-ness and branding campaigns based on the idea that more volume at the top of the funnel has to translate to more volume out the bottom of the funnel Of course, the funnel was never a funnel as there was never 100%

retention of what went into the top Instead, it was more of a sieve, with much of what went into the top

spilling out before it ever reached the bottom Thus, the idea that increasing volume at the top of the funnel would increase sales at the bottom has never been guaranteed In fact, depending on how many and how large the holes in the process of moving consum-ers from awareness to purchase, there has always been strong potential to waste huge amounts of time and effort moving people into the top of a process from which they would immediately fall out

This recognition of prospect attrition throughout the traditional “funnel” to purchase and the question about how to decrease such attrition necessitated a new way of thinking about the path to purchase In

2009, McKinsey Consulting introduced the idea of the Customer Decision Journey, which has subse-quently become the new standard in thinking about the path consumers take from awareness through purchase and importantly, even after purchase Since its introduction, it has gone through several stages of evolution and refinement, such as the version of the journey we will reference throughout the pages of this book

Figure 1.2 takes the original notions of the McKinsey Customer Decision Journey and adds two additional dimensions: the role of external “life events” as trig-gers to the customer decision journey, and the inter-action points between customers on the journey and data about those customers

The process begins on the far left with a “life event”, which is some piece of context that provides the impetus to take action in our product category Life events are diverse, and relevant life events for any business will vary based on the nature of that busi-ness Life events range from major events such as

marriage, a new job, a new house or the birth of a child to everyday events such as hosting a party or even just having time for lunch, reaching the week-end, getting off from work in the afternoon, or need-ing toilet paper Life events do not need to be major

in order to be significant triggers for the customer decision process The size and scope of the life event

is not what matters in itself What matters most is that

we, as marketers, are cognizant of the fact that there

is always an external context to a customer’s entry into

a decision journey — that the customer has a life side the decision process, and that something about that life brought them in to the decision process

out-Understanding this, the marketer should treat every life event as something important enough to their customer to trigger the expenditure of thought and energy through the decision process, and should of course recognize that the more significant the life event, the more significant the customer problems, objectives and needs

From the triggering event, we proceed clockwise through the diagram Customers in the first round of

Figure 1.2

The Customer

Decision Journey

the journey will travel around the outside path labeled “active evaluation” While on this path of active evaluation, they are developing an “initial con-sideration set” with regard to the options available to them to address the problems, objectives and needs established with the triggering event Awareness of their options is of course very important at this stage,

so it is good that awareness building is already a nificant piece of most marketing programs But ini-tial consideration extends well beyond just awareness and is in fact a process of comparison and evaluation leading to a decision

sig-To survive the active evaluation stage, brands must not only stay within consideration throughout the process, but must also stand-out from other options

by the time the process reaches a decision point

While understanding the importance of awareness has led to a focus on that stage, with data being used

in increasingly sophisticated ways to optimize ing on targeted impressions for awareness develop-ment, the entire active consideration process is still under-addressed by most marketing programs This

spend-is changing with the evolution of marketing mation software, and those firms that are address-ing this change with the most focus are poised to reap the rewards To repeat one of the most com-mon themes of this book, relevance delivers results, and if there is a critical time to establish relevance,

auto-it is during the period of time when your company

is being considered against competitors for a fit with your customers’ needs The means to establish-ing relevance is of course by generating meaningful points of engagement from insights about each customer’s context as drawn from your data The method for this will emerge through the rest of the

book, so suffice it to say here that knowing when and why to apply data to engagement is at least as critical

as knowing how to do so

Moving clockwise along the diagram, the customer moves through active evaluation and ultimately reaches a decision If the trigger was not strong enough to overcome dissatisfaction with all of the options considered, then the customer may decide not to make a purchase at all If the trigger was strong but no option was truly satisfactory, then the cus-tomer will begrudgingly select the least dissatis-factory option With the long-tail of options ranging from standard to niche presented today in most mar-kets, this is increasingly less of a concern to most consumers, who are happy for example to switch from hotels to Airbnb to meet a set of requirements that the hotels could not match, but which they had

to accept before Airbnb was an option And if during active consideration one option differentiated itself according to the customer’s needs, at this point the decision will be made to purchase from that brand

Continuing clockwise around the consumer decision journey, the next stage focuses on the post-purchase,

or post-decision, experience that the customer has with a brand Typically, this is thought of as “cus-tomer service” for customers who have bought some-thing from us, and that is a large segment of the population that a company will engage at this stage

in the journey But those customers who decision led

to an option other than ours still can, and should, also be engaged here

For those potential customers who chose another option, our post purchase engagement will be

focused largely around paid media Perhaps most importantly — and currently unfortunately uncommon — we need to recognize when these prospects have made a decision and are no longer in active evaluation, and we need to suppress serving them ads around our product Most readers of this book, as users of the web, will know how annoying it

is to be served ads about something we have already purchased So, even if we have lost a sale, it is impor-tant that we do not further dissuade the prospect from engaging with us by continuing to try and get them to buy something they are no longer shopping for Having provided that level of relevance in their experience, we then use data (primarily from paid media and Data Management Platforms [DMPs]) to anticipate their future needs and to be ready to deliver greater relevance when they reach their next trigger

For active customers, engagement in the decision stage of the journey is the key to guiding them into the “loyalty loop” This loyalty loop comes from establishing a strong preference for our brand with the customer so that at the next trigger, the customer’s active evaluation will default to our com-pany’s options and will bypass the active evaluation

post-of our competitors Suppression post-of no longer vant paid media and conversion-oriented email is equally important here to avoid alienation from our brand post-purchase Nothing seems sillier (or is more wasteful to the marketing budget) than seeing targeted advertising from a brand for a product recently purchased from that same brand Having shown that we can achieve minimal context recogni-tion by knowing that this customer has purchased from us, we then use 3rd party DMP data as well as first-party data from the customer’s purchase

rele-experience and post-purchase engagement with our earned and owned media channels (e.g app, web, email, social, customer service) to ensure satis-faction with their decision, and to anticipate and prepare relevant responses to their potential upcom-ing needs If we’ve done this effectively (through methods this book will provide in subsequent chap-ters), we are then much better poised to respond faster and with more relevance to the customers next round of needs than any other options, elevating our brand in their next round of consideration

The Customer Decision Journey (CDJ) provides a very helpful sense of flow from trigger to purchase and post-purchase behavior that allows more refined marketing strategy within the demand chain It is worth noting that even the most helpful models tend to oversimplify the reality they seek to repre-sent, and as two of my Northwestern University Integrated Marketing Communications (IMC) pro-gram colleagues have observed, the customer deci-sion journey is no different At any level of perspective

on a process, there is a usually a more complex layer beneath Professors Ed Malthouse and Tom Collinger in the IMC program conduct research into drivers of the decision process through the Spiegel Research Institute, and have developed a conceptualization of customer demand that consid-ers the process as less of a flow and more of a con-tinuous interaction of multiple drivers to the purchase decision

The visualization (see Figure 1.3) represents the dependence of several factors in the purchase deci-sion, as movement in any of the gears in the diagram will turn all the other gears This conception of the deeper mechanics behind the decision to purchase

inter-from a brand fits perfectly with the customer decision journey with the understanding that the gears are the mechanism by which the consumer is driven through the CDJ flow.

1.4 Research and Analytics

The Spiegel Institute is one of many academic tions dedicated to research that helps organizations better understand the consumer and the type of brand engagement that produces business outcomes And certainly every company that survives or thrives in its market has some person or organization dedicated to that same understanding As digital marketing channels and the collection and application of data in those chan-nels continues to evolve, market research, customer insights and analytics are playing an integrated role in digital marketing design and execution In a market moved by the facilitation of relevant experiences, the organization with the best understanding of their

customers, the demand for their products, and the ers through the path to purchase have a clear advantage over competitors with less understanding of any of these drivers of sales and loyalty

driv-I am often asked to clarify how digital analytics differs from the legacy market research function extant in most organizations Figure 1.4 shows the way in which research and analytics wrap around and through the marketing process from pre-trigger to post-purchase

As discussed, the consumer decision journey starts with

a trigger — which we see at the bottom center of Figure 1.4 The trigger itself occurs within an indivi dual, some-one with needs, wants, interests, attitudes, beliefs family and friends The individual is also of a certain age, gen-der, ethnicity, income range, geography, education level and acting within a set of current circumstances In the center of the diagram above we see a flow from trigger

to the decision to purchase or not The flow is driven through the customer decision journey process by the mechanics defined by the Spiegel institutes gears

Figure 1.4

Research and

Analytics

To the left and right of the path to purchase lie the outside influences on that path From the left come the macro-level external influences over the path to purchase, including economic factors, competitors’

actions in the marketplace, and changes in ogy that alter production, supply or demand factors within our market From the right come influences related specifically to our brand, including the quality and value of our product and accompanying customer service, the public response to our product, and of course, our marketing efforts Both of these sets of influences serve as inputs and catalysts to the decision that takes place through the path to purchase

technol-The various flavors of research and analytics reside around and within this process and its influences

From a supply-chain standpoint, the most tal (though not simple) method of market research occurs on the arrow marked “Research (1)”, which considers segments of consumers, their purchases, and the influence of external factors on those pur-chases and then develops a model to predict expected future sales such that production numbers and mar-keting budgets can be built from those predictions

fundamen-This approach to research is improved via “Research (2)” which adds consideration of the brand experi-ence influences on the outcomes in the purchase path, improving the understanding of the variables that influence “purchase or not” and thus improving the accuracy of forecasts and the marketing mix models built against those forecasts

The branches of research shown in the arrows marked “Research (3)” give us improved understand-ing of our customer segments in demographic, psychographic, attitudinal and behavioral terms, and

provide customer insights on which our marketing communications and service delivery should be built

To be most effective, this research combines tive and quantitative methods, and delivers results that drive the accuracy and applicability of the more quantitative research approaches discussed above

qualita-Finally we come to your reason for picking up this book: analytics As seen in Figure 1.4, analytics is rooted in the understanding of each individual, their influences and their circumstances, and runs parallel

to the path to purchase through the status of the chase decision and then around into the formation

pur-of insights and tactics (e.g spend) that influence the next set of brand experience inputs to the path to purchase/CDJ Importantly, whereas research is typi-cally conducted in a batch process against large sam-ples, this analytics process is conducted in a streaming, real-time timeframe on a case-by-case basis for each individual within the path to purchase/CDJ This characteristic is what allows the output of analytics to drive real-time marketing decisions such as real-time bidding on media, real-time content customization, and cross-channel marketing automation coordina-tion In the digital environment, each time a touch-point is engaged by a potential customer, the analytics function should have algorithms in place that adjust the propensity for, or the probability of, a purchase

by that individual, and that determines what to do next in marketing to this individual given this new probability for purchase

This summary of the function and objective of digital analytics provides us with the launching point for the remainder of this book As mentioned, the subsequent chapters will help you to understand the fundamental

components and steps required to build effective data-driven experience delivery, beginning with the common current state within marketing analytics, and finishing with the expected near-term future state in programmatic experience delivery, and the influence this may have on businesses and cultures.

Strategy, Technology, Science & Art

The organizational discipline that defines the collec­

tion and integration of data, effectively translates that data into useful information, and applies that infor­

mation to optimizing marketing delivery in a variety

of forms is most clearly thought of as “Marketing Science” As is apparent from this prior definition, data and technology reside at the core of marketing science While this data can be derived from measure­

ment in digital channels, it can also be accumulated through more traditional market research methods

In fact, to remain relevant in the 21st century, corpo­

rate market research functions are finding it neces­

sary to build on their traditional market research data sources and capabilities to better understand digital consumers using digitally collected data On the con­

tinuum of this evolution, what distinguishes mar­

keting science for digital marketing from traditional research functions is that marketing science not only understands digital data and technology, but is able to strategically and tactically guide the organization in the application of insights from this data through both content and media technologies as the core driver in delivering relevance in the digital user experience

Chapter

TWO

Marketing Science: The collection, aggregation and application of business data and consumer insights

to optimize customer experience and business results through marketing technology

Thus, the Marketing Science position in the delivery

of digital marketing is a hybrid of several traditional specialized roles First and foremost, there is no sense in developing digital marketing strategy with­

out insights from data, and there is no sense trying to develop insights from data without an understanding

of the business and marketing strategy these insights will serve Accordingly, the Marketing Science Analyst

is truly the next generation of what has traditionally been considered the Digital Strategist role

As McLuhan showed us in Chapter 1, the medium is the message Thus, effective delivery of a message in digital marketing requires not just an understanding

of the vast proliferation of digital media technologies (or digital touch­points), but also an understanding

of how information can be applied to optimize the effectiveness of message delivery through these tech­

nologies Put more succinctly, digital marketing nology is information technology Thus, in guiding

tech-the application of data to digital experience, tech-the Marketing Science Analyst must be an applied infor­

mation technologist and data scientist:

1 Understanding marketing technologies and digi­

tal media touch­points’ code enough to ensure that the right data is being collected from each digital touch­point;

2 Understanding data structures and sales/

marketing IT systems enough to ensure that all of this data is being integrated together for cross­

channel analysis;

3 Being capable of defining and/or developing explanatory/predictive models from data that can

be applied programmatically (e.g algorithmi­

cally) to the delivery of digital experiences; and

4 Defining how to apply analytics from streaming data to optimize the delivery and performance of these digital experiences for each visitor across channels

Thus, the Marketing Science Analyst must be able to move between and translate across marketing strat­

egy, information technology and data science But their work is not done yet While, strategy, informa­

tion technology and data science are critical to facili­

tating the delivery of messages to the right people in the right touch­points at the right time, there is one more fundamental factor to every communication through every touch­point: the right message Here

is where the art of marketing science comes to play

The data, analytics and algorithmic delivery of con­

tent will never be optimized if the content itself is not effective Thus, the data, insights and strategy devel­

oped by the Marketing Science Analyst must also be brought to bear in better understanding the people who are engaging with our brand through digital (and other) touch­points, and in artfully designing content and user experience (UX) in accordance with that deep and detailed understanding

A simple process for the delivery of digital experi­

ences from the perspective of the Marketing Science Analyst is laid out in Figure 2.1 Beginning in the lower left section, the collection of data is the core of our marketing science/digital analytics capabilities

As has been true in computer science since the first card was punched: what you put in determines what

you get out In digital marketing, the collection of data happens through software code Every analyst should desire to verify the quality of the source data they will use to develop insights In digital marketing, this means looking into code, and being able to explain your requirements for data collection from a touch­

point experience to the software developers who are building that experience

With data flowing into the experience delivery process, the Marketing Science Analyst can begin to apply their strategy, science and art The bottom right section represents the “design and development” stage of a digital experience Here, the analyst is exercising their data science, strategic and creative capabilities to:

1 Discover and explain what makes an engagement effective;

2 Define strategies around those explanations and determine how the effectiveness of delivery against these strategies will be measured;

3 Build models and algorithms that will allow the optimal delivery of those strategies; and

Figure 2.1

Experience

Delivery Cycle

4 Work with the content and experience design creative teams to ensure that the experience is built around these effectiveness insights

With the experience built and put into production, the attention of the Marketing Science Analyst then shifts to the top section Here, the analyst measures the effectiveness of the delivery and guides the tech­

nical and creative optimization of that delivery This can be done through testing and experimentation and through quantitative programmatic optimiza­

tion With performance results being returned and optimization efforts underway, the Marketing Science Analyst returns to the first step in the next round of design, asking what data from the existing delivery can be used to optimize the next round, and defining how to collect needed information that is not cur­

rently available

Since we start and end this process with the collec­

tion of data, let’s first consider the sources for this data from across the organization While brands have become careful to ensure consistent branding across their traditional and digital channels, their digital marketing practices are still structured pri­

marily by channel and function (with a paid media team, a social media team and a web team for exam­

ple) Such a structure typically fails to create a truly integrated and relevant experience as a consumer moves across digital channels The capability to cap­

ture and use data from any consumer’s digital engagement, and the growing expectation for con­

tent personalization that consumers have as a result (beginning with their Amazon and Netflix experi­

ences), means that the disconnection of data across channels will be felt by the user and will impact their

relationship with the brand Thus, to secure brands’

relationships with digitally embedded consumers, Marketing Science must first collect and connect the systems (and their corresponding data) that are shown in Figure 2.2

The gaps between each of the boxes in Figure 2.2 reflect the siloed or segmented nature of digital mar­

keting practice — and accordingly the data generated from that practice — which Marketing Science must help brands overcome

2.1 Paid, Earned, or Owned Breakdown

In the center of this Digital Data Stack are the communications channels through which digital engagement occurs At the top are the “owned media” channels, which have been the primary focus

of Marketing Science for decades Digital analytics

in marketing began with web tagging, and in many cases this is still the primary analytics competency within marketing or the agencies producing a brand’s owned channel marketing Chapter 3 will focus on the owned media channel — and data collection, connection and application in that

Figure 2.2

The Marketing

Science Stack

channel — beginning with the website but also extending to web services, email and mobile apps

Next come the “Earned Media” channels, which are the focus of Chapter Four A brand’s engagement with consumers through earned media occurs through search and social media, and Marketing Science for earned media includes SEO analytics, social media content tagging, social engagement analytics and social media listening

Last are the “Paid Media” channels, which typically command the majority of a digital marketer’s spend across The major paid media outlets include search marketing and paid display advertising, but social media advertising is increasingly adding options (and complexity) for media planners Paid media has seen the earliest and most widespread application of ana­

lytics to content delivery through “retargeting” which analyzes a user’s past online behaviors to identify the best advertising opportunity, through social content network analysis, which allows social networks to use insights about the interests of a user and their social network connections to increase the relevance of advertisements, and through “programmatic buying”

which uses business rules and real­time behavioral feedback to make a split­second decision on whether

or not to buy an impression based on that particular user’s history and characteristics

Our owned, earned and paid marketing communica­

tions channels are surrounded by additional data sources On the left side of the document we have first party sources of data that should be integrated with the data drawn from our Marketing Communications channels to increase the contextual relevance of digital

engagement at any point in time Customer service records, sales data and survey responses should all support more traditional consumer insights research and usability analysis to increase the relevance of the experience being delivered to the digital user

On the right side we have third party data that should also be used to support predictive models that best match content with consumers and thus deliver the highest level of relevance within digital engagement

Data Management Platforms (DMPs) and Demand Side Platforms (DSPs) play a central role in retargeting and real­time buying by tying together users’ demo­

graphic characteristics with their exposure to and engagement with advertising across the web as well

as anonymized information about online and offline spending

So, with the necessary background now established and with a model of the various pieces of media that need to be connected together as our foundation for moving forward, we are now ready to explore how to collect and connect data to drive relevant digital engagement for your brand

The Challenge to “Collect & Connect”

The Latin root of the word “data” means “a given or a thing”, and that is really what data is, a collection of

“givens” Data is the source from which information can be created, but it is not quite information in itself

The transition of a set of givens from data to informa­

tion requires analysis Thinking in terms of a coin flip, data is the possession of a coin with one side up, and information is the recognition of which side that is