LinkedIN -Jerome Vittoz - CRM 12 - Thesis MA Final
•
1 like•1,012 views
This document provides a summary of a master's thesis on marketing automation strategies. The thesis examines how marketing automation can help companies achieve CRM leadership and increase marketing efficiency and revenue generation.
The thesis consists of an introduction to marketing automation, a description of the research methodology, findings from interviews with 12 marketing automation experts, and conclusions. Key findings include that marketing automation is most suitable for companies with large customer interactions across complex digital journeys. Successful implementation requires integration with data management platforms and clear customer profiles. The most commonly cited effective automated marketing rules and flows included welcome programs, abandoned cart campaigns, and programs to retain existing customers.
1 of 277
Download to read offline
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
More Related Content
LinkedIN -Jerome Vittoz - CRM 12 - Thesis MA Final
1. Marketing Automation Strategies
for CRM leadership
Marketing efficiency and revenue generation, analysis of
methods that works
Jerome Vittoz, Buehlstrasse 47, 8055 Zürich
Masterarbeit am Institut für Marketing Management,
Zürcher Hochschule für Angewandte Wissenshaften
Master Thesis Director:
Dr. Roger Seiler
Zürcher Hochschule für Angewandte Wissenshaften
Supporting Lecturer:
Prof Dr. Frank Hannich
Zürcher Hochschule für Angewandte Wissenshaften
2. 2
1 Introduction
1.1 Preface
A special thanks to Petra Staudenmeier, former Head of International Marketing at Lindt
& Srpüngli for her support and in making my attendance of this Master program
possible.
Thanks to Uwe Sommer, Lindt & Sprüngli group Vice President and member of Group
Management for his confidence and encouragement in developing digital marketing at
Lindt & Sprüngli.
I do thank Dr. Roger Seiler and Prof Dr. Frank Hannich – Zürich University of Applied
Science for their guidance along the redaction of this paper.
3. 3
2 Management Summary
Name of the thesis: Marketing Automation Strategies for CRM leadership: Marketing
efficiency and revenue generation, analysis of methods that works.
Objectives: The objectives of the thesis is to help the reader understand how Marketing
Automation strategies lead companies to CRM leadership and most importantly
increase marketing efficiency and revenue generation.
Lead Research questions: What are the required means, skills, tools and processes for
successful Marketing Automation, leading to marketing efficiency and more revenue
generation?
Methods and procedure: The desk research has built the base for the qualitative
content research study. Then the study was based on the interviews transcription of
twelve Marketing Automation experts representing companies of different sizes,
regions business models. The interviews have been interpreted using the technique of
qualitative content analysis exposed in chapter 6.1.2. Finally, we have discussed the our
findings and conclusions in the chapter 6.3 and 6.4.
Major findings: Marketing Automation should be considered as fundamental
component of any customer centric strategy. It is most suitable for companies with large
volume of customer interactions across complex digital customer journeys and to
improve marketing efficiency processes.
Root cause: Marketing Automation can be implemented and used for many different
reasons and not only for marketing efficiency or revenue generation. Managing an
important volume of customer interactions is the primarily reasons for implementing it.
However, it can also bring advantages in data quality management, in developing a
customer centric strategy, sorting out organizational issues especially coordinating
marketing and sales efforts, increasing customer satisfaction, develop customer
personalization and customer scoring. What are the methods that works?
System architecture: The ideal marketing solution does not exist. Nevertheless,
conceptually the integration should consist of a data management layer, enriched with
4. 4
another marketing orchestration layer and finally a layer for customer interfaces such
as CMS, e-commerce or any advertising medium. The whole works in a closed-loop in
order to constantly enrich customer ID profiles and improve profiling by using scoring
methods.
Data Management: Data quality is the most important element of a solid marketing
automation strategy. Methods such as data normalization, progressive profiling should
be carefully implemented. Marketing Automation in combination with a data
management platform, a clear customer referential with a unique record ID is the
required conditions to secure elementary data quality.
Organization & Processes: Successful Marketing Automation’s manager’s benefits from
the support of their management with clearly defined business question and know their
success factors. The implementation of Marketing Automation is an iterative and
multidisciplinary project between marketing, sales, IT and sometimes with the support
of a consulting agency. The initial process of creating Marketing Automation flows
should be using the customer journey mapping method.
Best practices Marketing Automation: Most successful marketing automation rules are
the ones that answer business questions. In our research, most often cited Marketing
Automation programs were: the look alike modelling program, the welcome nurturing
series, the first order push, the abandoned cart, the drive to store, the search triggered
campaign, the lapsing prevention program, the happy customers programs (see chapter
6.2.5 for details). Programs are sets of Marketing Automation flows put together with
the aim to reach one particular business objective across the customer life cycle.
Marketing Automation is suitable for companies that manage important volume of
customer interactions across channels and devices with limited resources.
5. 5
Table of content
1 Introduction ......................................................................................................2
1.1 Preface ............................................................................................................... 2
2 Management Summary .....................................................................................3
3 Table.................................................................................................................9
3.1 Table of illustrations .......................................................................................... 9
4 Thesis architecture .......................................................................................... 10
4.1 Why this topic? ................................................................................................ 10
4.2 Practical relevance........................................................................................... 11
4.3 Thesis structure................................................................................................ 12
4.4 Objectives of the thesis.................................................................................... 12
4.5 The research question ..................................................................................... 12
4.6 Theoretical foundation .................................................................................... 12
4.7 Methods and procedure.................................................................................. 13
4.8 Thesis thread summary.................................................................................... 14
5 Introduction to Marketing Automation............................................................ 15
5.1 History.............................................................................................................. 15
5.2 Today................................................................................................................ 18
5.3 Definitions........................................................................................................ 18
5.4 Other forms of business automation............................................................... 20
5.4.1 Sales automation...................................................................................... 21
5.4.2 Service automation................................................................................... 21
5.4.3 Advertising automation (programmatic advertising)............................... 22
6. 6
5.4.4 Benefits of Marketing Automation........................................................... 23
5.4.5 How does Marketing Automation differ from CRM?............................... 24
5.4.6 For whom is Marketing Automation most relevant?............................... 26
5.5 Marketing Automation theoretical models..................................................... 26
5.5.1 Lead to revenue management by Forrester............................................. 27
5.5.2 The strategic business question according to Carter’s model ................. 30
5.5.3 Irina Heimbach’s model............................................................................ 32
5.6 Enhanced New Marketing Automation Model................................................ 33
5.6.1 The campaign management..................................................................... 34
5.6.2 Marketing Automation Rules ................................................................... 34
5.6.3 Data Management platforms ................................................................... 36
5.6.4 Data collection.......................................................................................... 38
5.6.5 Data processing ........................................................................................ 39
5.6.6 Activation.................................................................................................. 42
5.6.7 Application example................................................................................. 43
5.6.8 Summary................................................................................................... 46
6 Research.......................................................................................................... 47
6.1 Study design..................................................................................................... 47
6.1.1 Desk research ........................................................................................... 47
6.1.2 Qualitative Content Analysis .................................................................... 48
6.1.3 The researched questions ........................................................................ 49
6.1.4 Sampling ................................................................................................... 49
6.2 Research findings............................................................................................. 51
6.2.1 The root causes and benefits for implementing Marketing Automation 51
7. 7
6.2.2 System architecture.................................................................................. 53
6.2.3 Data Management.................................................................................... 54
6.2.4 Organization & Processes......................................................................... 55
6.2.5 Marketing Automated, most efficient Rules & Flows .............................. 57
6.2.6 Best practices recommended by the interviewees.................................. 61
6.2.7 Mentioned obstacles and challenges by interviewee.............................. 63
6.2.8 What is next for Marketing Automation? ................................................ 64
6.3 Discussions....................................................................................................... 66
6.3.1 Summary of the researched questions .................................................... 70
6.4 Conclusions ...................................................................................................... 74
7 Appendix......................................................................................................... 76
7.1 Interview guide ................................................................................................ 76
7.2 Transcript of interview 1.................................................................................. 79
7.3 Transcript of interview 2.................................................................................. 90
7.4 Transcript of interview 3.................................................................................. 98
7.5 Transcript of interview 4................................................................................ 105
7.6 Transcript of interview 5................................................................................ 112
7.7 Transcript of interview 6................................................................................ 118
7.8 Transcript of interview 7................................................................................ 125
7.9 Transcript of interview 8................................................................................ 129
7.10 Transcript of interview 9................................................................................ 137
7.11 Transcript of interview 10.............................................................................. 145
7.12 Transcript of interview 11.............................................................................. 150
7.13 Transcript of interview 12.............................................................................. 155
8. 8
7.14 Codification of transcripts.............................................................................. 163
8 Bibliography.................................................................................................. 274
9 Declaration of Authenticity............................................................................ 277
9. 9
3 Table
3.1 Table of illustrations
Illustration 5-1 - References of Marketing Automation occurred 1975 - 2008 ............. 15
Illustration 5-2 - First Marketing Automation Vendors (Source: Marketing Insider)..... 16
Illustration 5-3 - Closed-loop marketing cycle................................................................ 23
Illustration 5-4 - How is Marketing Automation different from CRM ?......................... 24
Illustration 5-5 – The L2RM Process (Source: Forrester Research, Inc.) ........................ 28
Illustration 5-6 - The strategic business questions from Carter's model ....................... 31
Illustration 5-7 - General Framework of Marketing Automation................................... 32
Illustration 5-8- The Enhanced New Marketing Automation Model.............................. 33
Illustration 5-9 - Functional aspects of DMP (Source: Comprendre les DMP, 2015)..... 37
Illustration 5-10 - Scoring concept (Source: EC4U, Delphine Arvangas)........................ 40
Illustration 5-11 - Media optimization and scoring (Source: Converteo, 2015)............. 41
Illustration 5-12 - The 6 steps of the customer life cycle............................................... 44
Illustration 5-13 - Examples of Marketing Automation flows ........................................ 45
Illustration 6-1 - Study design for qualitative content research .................................... 47
Illustration 6-2 – 12 interviewers’ sample list................................................................ 50
10. 10
4 Thesis architecture
4.1 Why this topic?
The major motivation in choosing this topic lays on the on-going improvement of the
consulting and tools I want to provide to Lindt & Sprüngli subsidiaries to grow their
business. In August 2014, I had required to attend the program “Master of Advanced
Studies in Customer Relationship Management (CRM)” of ZHAW in Zurich. At the time,
I had discussed the following elements to convince my management to support me in
this endeavour.
10% of all retail sales will be transacted online by 2020. In 2016, 50% of all
purchases (in-store and online) will be influenced by digital touch points. For Lindt
& Sprüngli AG, transforming its organization to fully embrace Digital and CRM is
a critical factor in succeeding to grow market shares faster than the market and
require a deep understanding of CRM and multi-channel strategies. Since 2008,
Lindt & Sprüngli has continuously invested in building up a large community of
consumers across several social and sales channels such as e-commerce and
Retail stores, Mobile, Newsletters, Facebook, YouTube, Twitter. This growing
volume of customer contacts makes its management more complex and
hazardous especially across several countries. Lindt & Sprüngli needs more
integrated tools, processes, and know-how to face these changes. We all
observed the growing importance of digital marketing touch points for our
consumers and the impacts on organization. These changes imply that we
consider not just our target groups as whole homogenous segment but as single
direct customer with particular needs. We must slowly move away from “One to
Many” or mass towards “One to One” communication in real-time. I’m convinced
that soon, not only knowing consumer’s needs and profile will be crucial to
succeed in the business, but also to maintaining a high quality dialogue with our
customers. This will become a key objective for all leading companies. By
choosing this online educational program, I would like to better support Lindt &
11. 11
Sprüngli AG in implementing concrete CRM solutions, most probably for our
retailing organizations and help our teams collect, manage and analyse our
consumer data in order to enhance Lindt brand equity and grow sales.
Today, I simply have put a name on what I believe may help my company to achieve this
objective. Marketing Automation. I am proud to share with you the results of my
research and I hope you will find it useful.
4.2 Practical relevance
While literature provides an important volume of information about CRM, it is more
difficult to find academic literature about the functioning of Marketing Automation in
the context of CRM. The empirical proof that Marketing Automation is helping
companies to save costs and increase revenue generation is missing from an academic
point of view. More, our preliminary researches have shown no or very little
standardized and structured researches about the impact of Marketing Automation on
revenue generation and marketing efficiency. The existing literature provide guidance
about the implementation of Marketing Automation software but it lacks a standardized
model to highlights some of the key components and processes to realizing the true
value of Marketing Automation as a marketing strategy.
12. 12
4.3 Thesis structure
The first part of the paper introduces the purpose of the thesis and its objectives,
construction and overall goals. The second part is dedicated to an introduction of
Marketing Automation through history, application, process and methods including key
existing theoretical models that we used to establish the “Enhanced New Marketing
Automation Model” described in chapter 5.6. The Third part is dedicated to the findings,
discussions and conclusions by analysing the practice of interviewed companies using
Marketing Automation.
4.4 Objectives of the thesis
The objective of the thesis is to help the reader to understand how Marketing
Automation strategies leads to CRM leadership and most importantly increase
marketing efficiency and revenue generation. By strategy, we mean the mastering of
means, skills, tools and processes that answer a company strategic business questions.
4.5 The research question
The research question for this paper is: What are the required means, skills, tools and
processes for successful Marketing Automation, leading to marketing efficiency and
revenue generation?
4.6 Theoretical foundation
The fundamental reason for companies wanting to build relationships with customers is
economic. If they manage their CRM well, they have two important key business
benefits as follow:
Marketing efficiency
Marketing Automation is in fact a derivative from the business process automation
defined as a process of managing information, data and processes to reduce costs,
resources and investment. Business Process Automation increases productivity by
13. 13
automating key business processes through computing technology (Technopedia,
2015). Literature also stress out the importance of keeping existing customer buying
over customer acquisition, especially from the cost perspective. It costs a company six
times more to sell a product to a new customer than an existing one (Dyché, 2008).
Marketing Automation benefits will ultimately improve lead management, nurturing,
provide measurable results, enhance targeting and personalization with seamless
campaign execution, increasing efficiency and productivity, align marketing and sales,
support CRM, develop business intelligence and improve improved user engagement
(Regalix, 2014).
Better Financial performances
Companies generate better results when they manage their customer base in order to
identify, acquire satisfy and retain profitable customers. These are key objectives of
many CRM strategies (Buttle, Customer Relationship Management: concepts and
technologies, 2011). Organizations that embrace marketing metrics and create data-
driven marketing culture, have a competitive advantage that results in significantly
better financial performances than of their competitors (Jeffery, 2010).
4.7 Methods and procedure
To answer the research questions, desktop research will be conducted in order to get
in-depth understanding of the research field, in particular the Marketing Automation
history, principles, definitions, processes and tools. The desk research will build the base
for the qualitative research. This paper will focus on the main theories and models from
which Marketing Automation has derived from and can be found in chapter 5.5.
Literature research included books, peer-reviewed and non-peer reviewed articles from
various journals and studies from consulting and research companies. The purpose of
the data collection is to identify the current state of research in this particular research
field. Because the research questions cannot be answered by gathering information
exclusively from desk research, a qualitative research study will be conducted. Part of
the data used in this paper will be collected in performing interviews with Marketing
14. 14
Automation experts. The questionnaire used for the qualitative content research will be
designed according to the “Enhanced New Marketing Automation Model” proposed in
the chapter 5.6. The aim of the interviews is to meet various Business to Consumer and
Business to Business companies’ professionals using Marketing Automation across
various industries of different sizes and countries such as CRM, Retail e-commerce or
marketing departments, the detailed list of the interviewees can be found in chapter
6.1.4. The interviews will be transcribed and interpreted using the technique of
qualitative content analysis exposed in chapter 6.1.2.
4.8 Thesis thread summary
The below illustration will give the readers the lead thesis architecture while reading this
thesis.
Illustration 4-2 - Thesis thread
15. 15
5 Introduction to Marketing Automation
5.1 History
According to Heimbach, the term Marketing Automation as we know it today, was first
introduced by John D.C Little in his presentation at the 5th Invitational Choice
Symposium UC Berkeley 2001. Little refers to it as the automated marketing decision
support on the Internet. Little formulated its essence with this phase: ‘‘What do we tell
retailer X to do when customer Y arrives on Monday morning?”. However, Marketing
Automation term seems to be older than that 2001. In using Google Books Ngram
Viewer, we have created a graph showing how Marketing Automation have occurred in
a corpus of books in between 1975 and 2008. The graph below show that Marketing
Automation is not a new idea and seems to have been used since the late seventies.
Illustration 5-1 - References of Marketing Automation occurred 1975 - 2008
This discovery led us to research more in depth about the term Marketing Automation
and we found out that the very first mention appeared in 1962 when Charles R. Goeldner
published an article about it. Automation in marketing is taking several forms, such as
automatic stores, vending machines, electronic data processing, and automatic
warehousing (Goeldner, 1962). While automation was existing at the time in only a small
16. 16
experimental number of institutions, the author believed at the time it would have a
profound influence on the future of American marketing institutions.
The history of Marketing Automation as we know it today, really started in the late
1990s. The first rise and fall of Marketing Automation was tied to the dot.com boom and
burst. The 1990s saw a myriad of Marketing Automation vendors come up with solutions
on the market including SAS, UNICA and Eloqua to name a few. After the burst only a
few survived as standalone vendor such as SAS. Some of them are still in the market
today, such as UNICA which is part of IBM or Eloqua part of Oracle. The middle of the
2000 marked several shifts for Marketing Automation when next generation vendor like
Infusion Soft or Marketo emerged. Over the next years, the pioneers’ s success inspired
a number of other competitors to enter the market. These included companies such as
Pardot now part of Sales Force.
Illustration 5-2 - First Marketing Automation Vendors (Source: Marketing Insider)
In the late 2000s, Marketing Automation started to speed up in awareness and interest,
essentially due to three main reasons.
First, changing buyer’s behaviours forced companies to change how they sell and
market. Before the Internet and social networks, buyers had limited options to obtain
the products and services information they wanted, so the seller controlled the buying
process. Then customers moved to the power position. They could get the information
they wanted on their own by delaying the initial contact with the sale representative
until they know as much or even more than the salesperson. To address this challenge,
17. 17
Marketing Automation started to play an increasing role in the revenue process. They
nurtured relationship with early stage prospects until they became ready (Jon Miller,
2013). Nevertheless, how do you manage individual dialogue with hundreds of
thousands, even millions of potential customers? This is precisely why having a
Marketing Automation platform became so critical in the mid-2000. There was no any
other way to keep up with the demand of modern marketing.
The second factors is linked to the 2008 crisis, which forced companies to change their
approach to market in favour of more lead and revenue generation focus (Jon Miller,
2013). Almost all companies have been affected by the crisis. While smaller companies
literally searched to cut cost and headcounts to reinvest in lead to revenue generation
management, larger companies have seen potential by introducing Marketing
Automation not only to survive, but also take advantage of this evolution to reinforce
their leadership. They invested in technology that automated and streamlined critical
revenue process. Particularly in crisis time, when marketing considered as expenses,
measurability became key to demonstrate efficiency across digital channels in order to
reallocate marketing budget to higher ROI channels. Marketing Automation became also
more popular because of its capability to prove return on investment.
The third trends that favoured the development and adoption of Marketing Automation
Software came from a new software delivery model called “Software as a Service”
(SaaS). This means that marketing managers could access any software through their
browser without or very little support of their IT. Because these software are sold using
the subscription fee model, marketing manager could buy it using marketing expenses
budget rather that capital investments allowing them to control and anticipate their
expenses. These three reasons strongly contributed to make from the newly Marketing
Automation offer a rapid adoption in the 2000s.
Between 2010 and 2014, there was over $5.5 billion worth of acquisitions made in the
Marketing Automation industry. The main reasons were simple. Fragmented marketing
systems will need to be replaced by integrated marketing suite (Raab, 2011). Larger
Marketing Automation vendors started to acquire smaller. The largest one was
18. 18
Salesforce’s ‘Russian doll’ acquisition of ExactTarget for $2.5 billion, after ExactTarget
acquired Pardot for $95 million (Taylor, 2015).
5.2 Today
Marketing Automation software editors are providing today more options than ever to
face more complex customer lifecycle, a broader array of digital marketing channels and
unprecedented volume of customer data. Marketers who implement Marketing
Automation increased their sales-pipeline contribution by 10% (Vendors, 2014).
Marketing Automation software vendors are now benefiting from their success, as
systems Revenues for B2B Marketing Automation systems will grow 60% to reach $1.2
billion in 2014, according to Raab Associates. Industries such as telecommunication,
consumer packaged goods manufacturer and financial services providers now comprise
as much as 75% of the customer base for several vendors. Software Advise, a division of
Gartner found that only 9% of potential software buyers currently use a Marketing
Automation system. Just 3% of micro-businesses and less than 10% or large firms use
Marketing Automation software, according to Raab Associate. Marketing Automation
vendors are also taking a certain number of strategic approaches, including acquisition,
mergers and market repositioning to leverage the opportunities leverage by these
challenges.
5.3 Definitions
The various definitions below indicate how differently Marketing Automation may be
defined across experts or vendors. In writing this paper and while conducting the
research interviews, we have used the following main definition:
Main definition: Marketing Automation is a category of technology that allows
companies to streamline, automate, and measure marketing tasks and workflows, so
they can increase operational efficiency and grow revenue faster (Jon Miller, 2013).
We have selected the above definition as the lead definition for writing this paper and
in designing the research study. One of the reason was that it was generic enough to
19. 19
encompass all forms of business automation. This first definition underlines the facts
that Marketing Automation’s main benefit is operational efficiency and revenue growth,
which is one of the initial assumption of the paper. Marketing Automation is still new
and complex enough that experts themselves struggle to agree on a unique definition.
Therefore, we would like to propose additional definitions to complete the main one
above:
Definition two: Marketing Automation software collects and uses data to send
personalized messages to contacts at different times based on their actions (Taylor,
2015). This second definition underlines the process, starting by data collection in order
to send personalized message and to engage prospects at different times based on their
actions. The definition is very close to Heimbach Marketing Automation workflow
described further in chapter 5.5.3. The definition suggests that data management,
personalization of content and action in time are the central elements of the Marketing
Automation processes. Marketing Automation here allows marketer to send the right
message to the right person at the right time because it is triggered by an action. We
could reasonably think that a phone call or a one-to-one conversation may do the same.
In fact, what Marketing Automation offers on top is the management of an important
quantity of conversations at the same time, automatically and without hiring a horde of
staff. Summarized, Marketing Automation allows companies to solve a fundamental
business issue: scaling the ability to communicate with a large volume of contact in a
personalized manner.
Definition three: Marketing Automation focuses on the lead acquisition and demand
generation activities within a marketing group, as opposed to the sales activities where
CRM systems as a whole tend to focus. Simply put, these tools automate marketing
processes — everything from strategic planning and campaign design to customer
segmentation, lead generation, nurture campaigns, prospect scoring, and closed loop
analytics (Schwartz, 2015). This third definition highlights the different approaches of
marketing and sales, marketing being focused on acquisitions and sales on revenue
growth. It contextualizes Marketing Automation in relation to CRM, making of both
methods a complementary approach. Some key fundamental functionalities are listed
20. 20
here and will be further exposed in the paper, especially in chapter 5.6. It also underlines
the automation of marketing processes, from strategic to operational part and suggests
the retro-looping functioning of it.
5.4 Other forms of business automation
Without well-designed and applied operational processes, there is little possibility of
implementing the CRM strategy (Buttle, Customer Relationship Management: concepts
and technologies, 2015). This capability often require IT investments in the following
four fundamental solutions: Sales force automation, Marketing Automation, Ad
automation and Service automation. The purpose of that paper is to introduce
Marketing Automation; therefore, it is important to be able to contextualize the role of
each well-established type of automated business processes below.
Business automation which Marketing Automation belongs to has always been a
strategic aspect of business operations. Many organizations today are relying on upon
digital technology for a wide range of application and automation has moved into a
central position in the IT investment landscape (Paul, 2015). The general principles of
automation in the modern workspaces include the deployment of solutions and
frameworks that will reduce and rework inefficient processes, improve the accuracy and
free up resources to focus on performance improvement (Paul, 2015). According to the
research institute Gartner, by 2020, customers would manage 85% of the relationship
with a company without even talking to a human (Gartner, Gartner Customer Summit,
2011). Automation value does not rely relies on sales growth, but for 75% on costs saving
leading to resources reallocation (Fétique, 2015). Isn’t Marketing Automation
everything that a computer does for us in marketing? Let’s think of a website: once
planned, programmed and put in production, what isn’t automated? It is available 24/24
a day, loading up webpage without the support of a horde of staff, collecting user data,
sending out emails or invoices. All these tasks are automated! A website is fully
automated, and it is part of the marketing arsenal. Is this Marketing Automation? Yes it
is. So why are we talking about Marketing Automation here? In fact, Marketing
Automation is an umbrella concept. Before Marketing Automation vendor agreed on a
21. 21
shared and widely adopted name, some vendors or consultant tried to impose other
names such as Forrester “Lead To Revenue Generation (L2RG)” or “data marketing”.
Since the growing interest of Business to Consumer companies for it and the increasing
number of customer contact across digital channels, Marketing Automation has been
adopted as an umbrella name, which regroup several marketing disciplines and
methodologies. Here are the other forms of business automation:
5.4.1 Sales automation
Sales force automation can be define as follow: Sales force automation is the application
of computerized technologies to support salespeople and sales management in the
achievement of their work-related objectives (Buttle, Customer Relationship
Management: concepts and technologies, 2015). Most of Sales Force Automation
software are designed to collect, store analyse and distribute user customer-related
data for sales purposes. This software’s enables managers to operate the following key
activities: account management, contract management, document management, event
management, lead management, order management, pipeline management, and
product configuration, amongst other things. Here the primarily focus is to support
essential sales processes. According to Buttle, most benefits for using sales automation
ar more cash flow, shorter sales cycles leading to faster inventory turnover, proved
customer relationships, accurate management reports, increase sales revenue, market
share growth, higher win rate, reduced cost of sales, more closing opportunities and
improve profitability (Buttle, Customer Relationship Management: concepts and
technologies, 2011).
5.4.2 Service automation
Service automation can be defined as follow: Service automation is the application of
computerized technologies to support service managers, customer service agents in call
centres, help-desk staff and mobile service staff operating in the field, in the
achievement of their work-related objectives (Buttle, Customer Relationship
Management: concepts and technologies, 2015). Customer service department are
22. 22
responsible for managing inbound call centers, operations, complaints handling and
resolution, order entry and processing, providing field sales support, managing
outbound call center operations and acting as liaison to other departments. When
service is delivered through a central call center, in a multi-channel environment there
need to be a tight integration between various communication systems including
telephony, email, and the web. Access to the right customer-related data to enable the
service agent to identify and fix the issue promptly is critical to the delivery of responsive
customer service.
5.4.3 Advertising automation (programmatic advertising)
Advertising deals used to be made by phone, fax and e-mail. Now much of that work is
being done on purpose-built digital marketplaces. Traditionally, companies are firing
message at huge audience on print or TV, hopping they’d reach out the right audience.
Ad trading desk are helping to do that more exact. Known as “programmatic
advertising,” this process involves computerized systems to sell online ad space to
advertisers and their agencies. It is comparable to a stock market, instead of trading
shares, these markets trade digital ad space, or impressions (Krashinsky, 2015). It is
another form of automation, which is reshaping the advertising industry. Today, about
half of the advertising expenses is automated, especially for display ads. Automated
trading increase efficiency by reducing costs and increasing targeting efficiency. An
advertiser or representative add agency registers at the trading desk, also known as
“Demand-Side-Platform” to place an ad that target a specific audience. E.g. a Swiss 35-
44 woman who likes French cooking, without kids, speaking German and living in the
canton of Zurich. They also put in the price range they are willing to pay for that digital
ad space. That information is processed in the ad exchange, also known as “Supply-Side-
Platform” where ad inventory’s of publishers is offered, and in fractions of a second, the
auction is done and the winning bidder’s ad is placed on a website. Meanwhile, dozens
of ad-tech companies are fighting for a piece of this market, the industry is call to evolve
a lot. In combination with its Data Management platform, which is a key component of
23. 23
Marketing Automation - see chapter 5.6.3, the advertiser will be able to automate the
ad buying process based on educated information extracted from its customer database.
5.4.4 Benefits of Marketing Automation
Marketing Automation is all about understanding customer needs and meet their
preference (Buttle, Customer Relationship Management: concepts and technologies,
2015). Customer develops a stronger sense of emotional and behavioural identification
with the company when they experience offers and communications that base on deep
understanding of their needs and preferences. One of the very first benefit when talking
about IT enabled CRM programs is cost saving. This is essentially due to more formalized
and standardized processes, generating operational costs savings. Normally, a company
willing to introduce a customer-centric-strategy invests primarily in building a single
view of the customer also known as “SVOC”. This approach integrate all form of data
from all operational units that involves customers such as sales, marketing customer and
customer service in order to create a unified view of the customer interactions with the
company. Once this is in place, the company’s employee may engage the customer one-
to-one based on his activity. Once this customer-related data infrastructure is in place,
companies may introduce analytical software and develop the ability to data mine in
order to produce actionable insights.
Enhanced marketing efficiency. Marketing Automation
allows marketer to automate what is known as the closed
loop marketing (plan-do-measure-Lear cycle) as illustrated in
the figures 5.3. Closed loop marketing approach makes sure
companies learn continuously from their marketing
initiatives, achieving higher level of marketing effectiveness.
Another source of efficiency is the identification of
inefficient or falling marketing initiatives in order to
reallocate resources to more successful activities. The
increasing number of customer contacts and sales
promotions companies must face today makes the
Illustration 5-3 -
Closed-loop
marketing cycle
24. 24
advertising campaign management much more complex and challenging. This main idea
behind the replication of marketing processes is that it should delivers greater control
over costs. Unsurprisingly, manual processes are often the source of errors and
inefficiencies. Consequently, automation of process is a way to reduce costs (Buttle,
Customer Relationship Management: concepts and technologies, 2015). Marketing
Automation allows companies to execute hundreds of campaign and events
simultaneously across multiple channels without the proportional increase of costs and
complexity in running the marketing activities. Marketing plans are often designed
months ahead. Marketing Automation allows manager to respond more instantly to
opportunities even if they are not part of a plan. Some features enable firms to engage
in real-time, responding immediately to an identified opportunity. For example if a user
is visiting a given product category for the third time in the same week, marketer can
send and automated offer with a reduction voucher.
5.4.5 How does Marketing Automation differ from CRM?
Illustration 5-4 - How is Marketing Automation different from CRM ?
25. 25
Marketing Automation and CRM are very similar marketing methods that have much
aspect in common. Having long proved their strategic value, CRMs are now an essential
tool for hundreds of thousands of businesses in almost every industry (Peterson, 2015).
CRM is not a technology but a strategy. It is nevertheless very hard to do without
technology (Kirby, 2015). Even if CRM is often considered as a technology, it is more a
business philosophy or a business model. CRM is a business model, which covers people,
customers, strategy, organization, processes and technology. It integrates all these
elements in the direction of the customer satisfaction with as a main objective a balance
between cost and return on investment (Huldi, 2015).
In most people’s mind, CRM is more a software or a tool than a strategy. In fact, CRM
software is a category of software that covers a broad set of applications and software
designed to help businesses to manage customer data and customer interaction, access
business information, automate sales, marketing and customer support and manage
employee, vendor and partner relationships (Beal, 2015). From that perspective, CRM
covers Marketing Automation. From a functional perspective, CRM systems typically do
not provide functionality for email marketing, prospect behaviour tracking and
marketing program management. It is true that many CRM systems can be customized
to operate these tasks like automated campaign flows, lead scoring and lead
deduplication, but it is hard and expensive. In the end, Marketing Automation focuses
on the need of marketing departments, lead generation while CRM provides must-have
solution to the sale department. CRM software is a database for storing user or
customer data and managing the revenue process. Alone, the company will have the
basic ability to store contacts and track sales stages, but limited insight into the buyer’s
journey (Peterson, 2015). The buyer’s journey start long before they become a
customer. With no information on where the leads came from, what their interests are,
and how they have interacted with the various marketing touch points, the initial
conversation will be a lot harder to navigate, analysed and understood.
Technically, a Marketing Automation platform is in other word a lead generation engine
without a CRM. Here, the perfect relationship between marketing and sales can be
expressed — at least in part by the structural relationship between Marketing
26. 26
Automation and CRM. When both processes are integrated, this means the end of
marketing and sales data silo for better operation alignment. Marketing can focus on
bringing in qualified lead and tracking customer journeys across digital touch point
before sales take the order completion with a full history. For sales, Marketing
Automation delivers a high quality leads much closer to the completions status than
ever. In short, CRM and Marketing Automation together should leads to reduced costs
and increased revenue.
5.4.6 For whom is Marketing Automation most relevant?
Marketing Automation is best suited to companies with predictable communication that
could be converted into a process with a large volume of users, subscribers or customers
(Taylor, 2015). The more contact the company has, the more potential impact the
software has. A company business model will influence the type of relationship between
a customer and the merchant. Retailers with direct customer contact will have greater
opportunity to collect user data compared to wholesaler. With few or no direct contact
with end consumer due to the nature of its distribution model the wholesaler will
traditionally focus on mass media. As a result, online retailers are in general more
interested in Marketing Automation than companies without an online shop (Heimbach,
2015).
5.5 Marketing Automation theoretical models
The aim of this chapter is to introduce to the readers a fundamental theoretical concept,
which influenced the development of Marketing Automation workflow, and depict the
fundamental processes used for it. The model we propose is inspired from 3 already
existing model as following:
1. The Lead To revenue Management Model by Forrester
2. The SWAT Iteration Framework of Carter
3. Heimbach’s model
27. 27
To start, we will introduce the “Lead to revenue management” (L2RM) concept
developed by Forrester Research which represent two parallel closed loop journeys.
Marketing Automation was originally used and designed for Business to business
industries but is today largely adopted in business to consumer’s context because of the
explosion of customer touch points and contacts. We will use it as a leading model. The
concept is particularly interesting because it puts in parallel the journey of the buyers
and the seller, aligning the marketing tasks (attraction) and the selling tasks (deliver).
To enrich the Forrester model to Marketing Automation, we have selected the very first
step of the “SWAT Iteration Framework” of Carter. Carter in his book “Actionable
Intelligence, a Guide to delivering business results with big data fast” outlines the
importance of the definition of a strategic business question before processing the data,
visualize segments and to take action. Without clear business, questions to solve in
mind, the implementation of Marketing Automation rules may lead to poor results.
Finally, we have chosen to present a third concept taken out of the paper “Marketing
Automation” of Irina Heimbach, which attempt to depict the general framework of
Marketing Automation. This recent paper brings to our attention six important
elements, part of a workflow specific to Marketing Automation. Taking the best of each
of the three existing model, we will design the “Enhanced New Marketing Automation
Model”.
5.5.1 Lead to revenue management by Forrester
Initial lead-to-revenue management automation solutions were developed to bridge a
gap between marketing or lead generation activities (e.g., website, online forms, trade
shows, direct mail, telemarketing, and email campaigns) and selling activities that were
managed by a CRM system (e.g., closing the deal). The opportunity to calibrate
marketing spend to revenue generation was a significant driver of L2RM. The below
figures show Lead-to-revenue management is a set of disciplines that can be strongly
supported by Marketing Automation, but marketers need to focus on the below
processes to make L2RM automation initiatives a success.
28. 28
Illustration 5-5 – The L2RM Process (Source: Forrester Research, Inc.)
This info graphic, based on independent research from Forrester, gives a visual
representation of how marketers can make the most of their lead to management
investments by developing a Marketing Automation road map that meet buyers journey.
The process displays two major processes. One is on the buyer’s side defined as the
buyer’s journey and the second one is on the merchant side, the Lead to revenue
management (L2RM) process. For each five phases of the buyers ‘journey the figures
show a corresponding merchant process.
5.5.1.1 The buyers’ journey
Buyers have different interest at different steps of the buyer’s journey. It is crucial to get
the buyers question that needs to be addressed at each stage. It is also important to
understand what will trigger a buyer to move to next stage and identify the barrier for
the progression. Here is the buyer’s journey phases explained:
Discover: At this stage, the buyers is asking himself, what outcome am I going to
achieve? What need or pain am I trying to fix or improve? Typically, the buyers
29. 29
determines the need to solve a problem. The budget for the solution is determined and
some approaches to solving the issue are addressed.
Explore: Buyers will start exploring and identifying possible solutions, product or
services, choosing an approach, consider risks and start comparing alternatives. At this
stage the buyer will be starting select some vendors and check for references or read
review for example.
Buy: A this stage, the buyers define a shortlist of vendors that are invited to bid, vendors
submit offers, solution is acquired.
Engage: This phase corresponds to the after sales. After the product acquisition, the
customer may get in touch with the vendor, will ask for support, and possibly upgrade
its product. The moment is crucial because this is at that moment that he will adopt the
product if its experience and results are satisfying.
Advocate: If the buyers has adopted the product, he had a positive experience of the
product or services; he might turn into an advocate. Advocate are the most wanted
brand influencers. They are the customer that advocate the brand because they are
passionate about the product or services and respect the company.
5.5.1.2 The Lead to Revenue Management Process (sell)
The closed-loop can be spited in two major phases. Attract and deliver. Initial lead-to-
revenue management automation solutions were developed to bridge a gap between
these two phases. Attract corresponds to the typical marketing activities with brings in
qualified leads which are mature enough that they can be approached by the selling
team in order to close the deal. This phase is structured in two sub-phases called
capturing and nurturing:
Capture: The capture phase encompasses all marketing activities that will help you
clarify who your best customers are, identify what they need and understand how to
connect with them.
Lead nurturing: Is the process of building effective relationships with potential
customers throughout the buying journey (Rothman, 2015). Note that Marketing
30. 30
Automation software editor Marketo distinguishes lead nurturing which is an evolved
version of drip marketing. Drip marketing has in common the send out of
communication such as email, direct mail or phone call, but it does not take into
consideration the activity of behaviour of users, because it is static and non-adaptive.
For example, drip marketing does not take into account personal preferences and
actions and cannot deliver the same value as lead nurturing, because it is adaptive and
personalized. The second phase is the deliver status, which starts with:
Sell: This covers the closing phase. It includes trial periods, bidding, price negotiations,
signing of contracts, and delivery of the product or service being sold.
Build advocacy: A customer advocacy policy encompasses all aspects of customer
contact, including products, services, sales and complaints. It is a total commitment
towards customer satisfaction. The idea is that if a customer is happy with the company,
they will pay more for the service. As mentioned above, if the customer had a positive
experience of the product or services, he will turn into an advocate.
5.5.2 The strategic business question according to Carter’s model
Business discovery is the critical step in transforming the unstructured mass of data
available into actionable intelligence (Carter, 2014). Carter is bringing up an interesting
element and invites us to think about the strategic questions that we are trying to solve
before implementing any automated marketing tasks. Merchants today are collecting
large volumes of data across many channels leading – sometimes to confusions and
complexity. Every business has its priorities and overall strategy. Direct your effort
depending on your overall strategic priorities.
31. 31
Illustration 5-6 - The strategic business questions from Carter's model
Therefore, we should not try to collect all the data, but only focus on the “burning
platform” (Carter, 2014). The strategic business questions – highlighted in blue in the
below chart - should be a guide to ensure that the priority of the project remains high.
Selecting on question to focus on allows the rapid and iterative development to occur
as the business and technology team can focus. Clarity of purpose allows sponsors and
supporters to get behind the project because they will clearly see the opportunity to
deliver tangible results. For Carter, while many people have been advocating the mass
collection of data, actionable intelligence suggests you to choose a different approach.
The data collection needs to be focused solely on the strategic question (Carter, 2014).
This enables the acquisition team to collect and review quality and relevant data. In
other words leading data processing method answers the questions: “What kind of data
do we really need to answer the strategic questions? Where can I find it? What is the IT
infrastructure that we need? Who are the people with the required skills to analyse it?
How can we capture the data in a cost effective manner? Focussing on the required data
should facilitate the acquisition, lower cost and increase feasibility. Therefore, we
propose to add the strategic business question to the Forrester model to complete it. A
revised version of it will be proposed in the chapter 5.6.
32. 32
5.5.3 Irina Heimbach’s model
Irina Heimbach with her “General framework of Marketing Automation” as presented
below, introduces some interesting element that will complement the L2RM model as
well:
Illustration 5-7 - General Framework of Marketing Automation
Current and stored information. Here Heimbach introduces the notion of stored and
current information. As seen in the chapter above, availability of data is a condition in
the context of Marketing Automation. Data remain an important element beyond the
analysis; since all automated marketing actions are direct responses to existing,
incoming or changing customer/user information (Heimbach, 2015). These data may
originate from a customer database, but may as well stem back to tracked user journeys
or clickstream data on the website in real-time. We assume she refers to synchronous
and asynchronous, because both tracked user journey or clickstream data can be stored
into a database.
Monitoring interfaces and rules. This is the place where marketer controls the
performance and creates the rules. If for example several options are possible, the
Marketing Automation software will be able to apply these options based on particular
events based on contextual or behavioural user information. This optimization process
33. 33
and the performance of the above-mentioned rules can be monitored and adjusted by
the manager at any time. The literature also talks about Campaign Management or
marketing orchestration.
Set of rules. Once the marketing owns some user or customer data, stored or collected
in real time, the marketer may create marketing rules. These rules are in other words
some marketing processes, which are pre-defined in the campaign management system.
Campaign Management will be an important element added to the original Forrester
model presented later and further developed in the chapter 5.6.
5.6 Enhanced New Marketing Automation Model
As mentioned earlier, we would like to propose below the “Enhanced New Marketing
Automation Model” which derived from the Forrester L2RM model, enriched with the
Carter and Heimbach models presented earlier. In this enhanced model, we have
added two additional elements: the business questions and the campaign
management component.
Illustration 5-8- The Enhanced New Marketing Automation Model
34. 34
The business question is a critical element and the starting point of all Marketing
Automation model as exposed in the chapter 5.5.2 above. The campaign Management
system is situated at the heart of all Marketing Automation; this is where rules and flows
are orchestrated and managed to activate customers.
5.6.1 The campaign management
Today, Marketers are using a wider range of communication channels to reach prospects
and customers with relevant messages on their preferred devices. Paid and organic
advertising campaigns are becoming an increasingly important feature for campaign
management within their Marketing Automation platform. As results, the Marketing
Automation systems now includes two core mechanisms: the “Data management
platform or DMP” and a “set of rules” like Heimbach models suggests earlier in chapter
5.5.3. We have grouped them under “Campaign management” and can be defined as
follow: campaign management is the technology-enabled application of data-driven
strategies to select customers or prospects for customized communications and offers
that vary at every stage of the customer lifecycle and buyer readiness (Buttle, Customer
Relationship Management: concepts and technologies, 2015). Campaign management
lays in the heart of the Marketing Automation process, because it put together the
planning, the implementation and the measuring of communication programs towards
targeted prospect or customers. In our newly created model, set of rules, data
management platform and analytics are the core elements of campaign management.
5.6.2 Marketing Automation Rules
This is the orchestrating element of Marketing Automation process. Before running a
campaign, it has to be designed and planned. Workflows established in the “Set of Rules”
set clear order in which tasks have to be performed. If we want to pick a representative
image of Marketing Automation, it would be one of these flow-like diagrams, e.g if a
customer visits one particular page, then send him this email — if he is a qualified lead
send him that promotion — else wait 2 days and send them this other email — if we
35. 35
know his email address — else cookie them for display advertising retargeting (Taylor,
2015): Here are the key elements of the workflow:
Illustration - 1 Marketing Automation flow chart from Oracle Eloqua
The trigger: What causes the workflow to start? Event based marketing also called
trigger marketing is a form of marketing that identifies key events in the customer and
business lifecycle which trigger a communication of offer. When an event occurs, a
customer specific marketing activity is undertaken (Ramshaw, 2015). The cause may be
an IP address (for location-based marketing), the usage of a particular browser, a device
or the time.
Timing: This is the time in between different steps of the flow. For example, a reminder
is sent 12 hours after a user has abandoned his cart.
Conditions: This defines what happen if a condition X is true or if the condition X is false.
For example if a user abandons a shopping cart before payment, this triggers a follow-
up reminder email aimed at converting the lapse browser into a purchase.
36. 36
Activation: What should be activated when the condition is true or false? This could
activate the use of the customer-preferred medium. Campaign execution happens when
the message is delivered through the selected communication channels.
Mediums: Campaigns can be run across many different mediums independently,
consecutively or simultaneously, such as display advertising, email, search, social media,
website, outbound or inbound phone, text messages or mobile. Personalization in
Marketing Automation context means that customers are known as individuals rather
than demographic stereotypes. Personalization can happen in real time upon
individual’s preferences or behaviours.
Workflows or rules allows manager to plan, design, manage and monitor specific
automated marketing campaigns, which are often complex and follow event based next-
step rules. The most basic campaign management tools enable campaign workflow,
audience segmentation and targeting and campaign execution.
5.6.3 Data Management platforms
Since (Little, 2001) first formulation, the core motivation for implementing Marketing
Automation has not changed: the lack of appropriate models while facing huge amount
of data automatically collected by online companies. Digital channels today provides to
companies new and valuable source of customer data such as historical, financial and
behavioural data across multiple devices. Nevertheless, the multiplication of channels
and data volume makes it difficult to gather, store, structure, analyse and action in a
simple way. Data remain an important element beyond the analysis, since all automated
actions are a direct response to existing, incoming or changing customer / user
information (Irina Heimbach, 2015). Data management becomes crucial for companies
when at least one of these three conditions are met:
First, their customers have a complex customer journey and they are using several
mediums to engage their brand such as email, web, mobile, store, customer support etc.
Second, the company has customer data stored in different silos with various origins
37. 37
such as online and offline, leading to a poor customer single view. Finally, the company
has response time issue when engaging which customer that it would like to improve.
Illustration 5-9 - Functional aspects of DMP (Source: Comprendre les DMP, 2015)
Data management offer the following fundamental benefits. It helps create a single
customer view, providing ways to identify users, enrich their profile and action these
across the digital landscape. It helps find new and high value audiences and monetise
them. Drive efficiencies across multiple channels and devices. In fact, a DMP has the
ability to help across the whole customer journey. Help marketers, and their media
agencies, secure greater efficiencies in targeting display advertising (Bay, 2015). Data
Management Platforms are especially good at optimizing the media buying process such
as display, Search Engine Marketing, video advertising or social media and support
greater personalization of and customization across the various touch points. The Data
Management Platform represented above is concretely a software platform that
enables the collection and the centralization of prospects and customers data. The
platform can enrich the user records in re-assembling user data, in segmenting or
scoring the value of each record and finally by activate the data to reach out users,
prospects or customers. The three phase are described in details in the below illustration
and the following chapters:
38. 38
5.6.4 Data collection
Today, marketing people are primarily and in many cases exclusively engaging online
and are closely aligning their effort with the key information found in their online
behaviour. By analysing and understanding prospect online behaviour such as email
responses, web pages view, social engagement and other core attribute, marketer have
a wealthy of insights to personally guide prospects through the buying process. The DMP
manages the process of taking structured data from a number of different sources and
organizing it at the customer level or at the cookie level if the person is unknown.
Various source of data may be aggregated in order to enrich the user records or profile,
such as purchase history, socio-demographics, behavioural data such as website
customer journey (tag management, analytics), web forms, explicit data such as
demographics, social media data preferences, geo-localizations.
Today, experts talk about three categories of data. First party data are the data collected
from digital platforms such as websites, apps, data from CRM systems and from
customers and their behaviours. Amazon, for example, uses its first-party data to show
users products it thinks they might buy on its homepage. Second party data are data
from partners who share their first-party data with other partner companies. For
example, a large advertiser such as P&G might make a deal with a large publisher to gain
access to its audience information. As far as P&G is concerned, that information isn’t
“first-party” data because it didn’t collect it itself. But it isn’t third-party data, either,
which is typically gleaned from a variety of places. Third party data are data from other
sources such as websites, online newsletter or blogs which have anonymous behavioural
data. For example, a third-party data provider might pay publishers (magazine, online
newspaper) to let it collect information about their visitors, and use it to piece together
detailed profiles about users’ tastes and behaviours as they move around the Web. This
information can then be sold to advertisers to help them target their ad buys. Finally
note that data reconciliation means also the process of assigning first; second and third
party data that belong to a single user to a single record.
39. 39
5.6.5 Data processing
In this chapter, we would like to introduce key methods to process the data. These
tasks are normally performed into the Data Management Platform.
Data normalization. Before to use the data, it is crucial to normalize the data. Data
normalization is a process in which data attributes within a data model are organized to
increase the cohesion of entity types. In other words, the goal of data normalization is
to reduce and even eliminate data redundancy, an important consideration for
application developers because it is incredibly difficult to stores objects in a relational
database that maintains the same information in several places (Wembler, 2015).
Progressive profiling. A typical data collection supported by Marketing Automation is
the “progressive profiling”, enabling marketers to ask for information incrementally
instead of all at once. Over time, leads will become more qualified because of their
interaction with a website (and other digital properties) and will likewise deliver useful
information to sales (Boush, 2011).
Data reconciliation. Once the user record is normalized, the Data Management Platform
can for example reunify two different records in one, for example when a unique user
has visited a website browsing first a mobile then a desktop. Two main methods are in
use: Deterministic is the method that is used when the users are registered, we call it
also explicit as opposed to implicit. The Data Management platform knows the person
has used the two mediums because he was identified. Probabilistic is the method that
uses algorithm to determine whether the users have similar behaviour. Probability are
based on e.g. IP address, screen resolution, time of connection, interest etc. The Benefit
is to be able to provide users a seamless customer experience across devices and
channels and for the merchant, to track user across channels and devices to understand
its preferences.
Data segmentation. A data management platform must enable the creation of
segmented contact lists based on field values (e.g. title, level, department) and inferred
data on location and activity (obtained through interactions with forms, company Web
pages, emails, etc.) for use in automated programs and reporting (Decisions, 2015).
40. 40
These segments can be kept in a static list, reflecting their status when the segmentation
was made, or automatically updated as contact data and activities change.
Data scoring is a form of segmentation. Lead scoring is a shared sales and marketing
methodology for ranking leads in order to determine their sales or activity readiness.
The score leads is based on the interest a customer show for a company, their current
place in the buying cycle and their fit in regards to the business (Maria Pergolino, 2015).
Data scoring fundamentals distinguishes two major statuses: explicit and implicit data.
Explicit scoring is based on information the prospect tells the company or otherwise
directly identifiable information. Implicit scoring is based on information that the
company observes or infers about the prospect, such as their online behaviours (Maria
Pergolino, 2015). Another important basic concept about customer scoring is the
distinction between Active vs. Latent buying behaviour. The benefits come from
adjusting your scoring accordingly. Active buying behaviour identifies “hot” leads based
on activities that demonstrate sales readiness and current interest. Latent buying
behaviour, on the other hand, involves lower engagement activity. The scoring concept
in illustration 5.11 shows sixteen different scoring segments based on purchase
frequency and activity level. For each segment, marketer can develop specific programs
Illustration 5-10 - Scoring concept (Source: EC4U, Delphine Arvangas)
41. 41
to bring that particular group of customer to the next segment level. Either up to a
higher Recency-Frequency-Monetary score or up to a higher activity level. The principle
is to use this scoring model to permanently track customer evolution across the scoring
model and propose him the right activation or incentive to develop the relationship
positively. Scoring model is essential to develop Marketing Automation programs. The
following illustration 5.11 shows another model with five different segmentations based
on scoring and their possible actions, focus and benefits.
Illustration 5-11 - Media optimization and scoring (Source: Converteo, 2015)
Predictive modelling is a commonly used statistical technique to predict future
behaviour. Predictive modelling solutions are a form of data-mining technology that
work by analysing historical and current data and generating a model to help predict
future outcomes. In predictive modelling, data is collected, a statistical model is
formulated, predictions are made, and the model is validated (or revised) as additional
data becomes available (Pliptop, 2015).
Lookalike modelling is an ad tech technics that bring together automation and
programmatic buying. It is somehow very near from predictive modelling. It is a
42. 42
methodology used by marketers to define what users they should target with the
highest propensity to buy (Hayter, 2013). They use first party data left over by existing
customers to find people who behave in the same way, but who haven not bought a
product yet. Let us say an electronic device manufacturer is having a sale and wants to
encourage further online purchases. He places a pixel on the sale confirmation page and
analyse the behaviour that purchasers have undertaken elsewhere on the web using for
example third party data, which are completely anonymous. This group is analysed in
order to reveal online behaviours that rank most highly amongst people with a
propensity to buy certain products.
Analytics and attribution modelling. All the results of a campaign are assessed and
measured whether the original objectives have been achieved. Several techniques are
used. Modelling is the process of interpreting the campaign results statistically, so that
future campaigns can be based on statistical insight into what works and what does not.
It is the process of identifying a set of user actions or events that contribute in some
manners to a desired outcome assigning a value to each of these events. Marketing
attribution provides a level of understanding of what combination of events in what
particular order influence individuals to engage in a desired behaviour, typically referred
to as a conversion. Often, the attribution of the sale is given to the last touch point, e.g
the website (last click wins model). The most attribution modelling used in the practice
are the “U model” where the first and last interaction are over proportionally valued.
Some other model are: the “progressive model” where each click receive an increase
value, the “linear model” where all interactions are valued at same level, the “digressive
model” where each interaction loose value with time (Fétique, 2015).
Reporting. The campaign results are computed and delivered in standard or customized
management reports to relevant parties.
5.6.6 Activation
Once the data is processed, Data Management Platforms can suggest segments that
may be activated. There are three major activation possibilities shown in the illustration
5.9. Paid media covers all paid advertising medium such as display banners, search,
43. 43
mobile and social units for example. Depending on the analysed data, the campaign
management system will activate a campaign using one or the other paid medium. The
second category is owned media, which covers email, mobile or website. Owned media
is a channel a company control. There is fully owned media like a website and partially
owned media like Facebook fan page or Twitter account. Often, owned media are
activated in combination with personalization. This means a newsletter or a website-
landing page will be personalized with the user preference and based on its past
purchase or behavioural history, pre-analysed in the data management platform. Finally,
offline medium may also be activated through call centre, in store communication or
why not printed direct mailing.
5.6.7 Application example
For a concrete application of Marketing Automation flows, we have structured the
exemplary rules following the principle of the customer life cycle represented in the
illustration below, which is another view of the “Enhanced New Marketing Automation
Model” introduced in chapter 5.3. The main difference lays in the fact that the
illustration shows the specific activities the company could activate for each stage of the
customer life cycle. As opposed to mass media activities, the advantage of Marketing
Automation is its ability to identify micros opportunities to engage with small amounts
of prospects or customers. The interest lays in the succession of marketing rules that
can be designed in advance and that are automatically triggered when a particular
customer event occurs. The charts conceptually show the major key opportunities to
engage with a prospect or customer at different stage of the customer life. We have
selected six key different phases. Each of them describes a particular moment of the
customer life cycle, which deserves some specific engagement tactics.
The table 5.12 below is an attempt to describe the above Marketing Automation flows
for each of the mentioned phases from 1 to 6. As seen in the chapter Marketing
Automation Rules 5.6.2, the rules are composed of triggers, conditions, timing, and
activations. To make the rule more real, we have added a communication message. We
also have mentioned the main technology or functionality used to execute the rule.
45. Illustration 5-13 - Examples of Marketing Automation flows
# Life Cycle Event Condition Timing Activation Message Main used
technology
1 Capture The prospect is visiting the
company website
He is browsing the
company site for > 3
minutes and belongs to
the hot lead segment
3 minutes Newsletter sign up
form is automatically
displayed as a pop-up
message
Invitation to receive a
newsletter with
incentive
User progressive
profiling
2 Nurture The prospect signs-up the
company newsletter (or
create an account)
The prospect isnt' a
customer
Nurturing
program over
6 weeks
Newsletter
introductory offers
Welcome program of 5
email messages
presenting interesting
facts about the
company
Nurturing
3 Remarketing A prospect is visiting a
particular webpage 3 times
consecutively withing the
same week
The prospect has left
the webiste without
ordering
Realtime Display or serach
remarketing campaign
is launched
Promotion Remarketing
4 First Sell The propsect's session is
unsusuall long lasting > 5
minutes
The prospect isnt' yet a
customer
5 minutes A web chat is launched The customer agent is
proposing some help
Customer support
5 Build
Advocacy
A customer has birthday soon The custome is a top
customer
10 days ahead The prospect receives a
discout promo-code as
a birthday gift
Happy birthday.
Celebrate with us.
Scoring
6 Re-assess A customer has not ordered
since 12 months
The customer is part of
the top customer
segment
After 12
months
A special promotion is
sent per email
Long time no see? here
is a special gift for you if
you order
Scoring
46. 5.6.8 Summary
With the introduction of marketing automation, we have shown the growing importance
of the discipline across the late 1990 influenced by the change in customers’ behaviour
on the Internet, the 2008 crisis and the raise of the cloud computing (see chapter 5.1).
Marketing Automation is not exclusively used by sales people anymore but by a growing
number of business to consumer companies and marketing teams essentially to
manager their huge volume of customer interactions in a more efficient manners. The
Marketing Automation industry is however quite small with only about 10% of the larger
companies which are are using it, and about 3% of small companies.
We have also selected most representative Marketing Automation definitions and tried
to give the reader the most precise definition. We have kept the following lead definition
in mind while writing the thesis: Marketing Automation is a category of technology that
allows companies to streamline, automate, and measure marketing tasks and
workflows, so they can increase operational efficiency and grow revenue faster (Jon
Miller, 2013). Beyond that, we have introduced forms of business automation such as
sales automation, service automation and Advertising Automation (see chapter 5.3 and
5.4) in order to give the reader the most comprehensive view of the discipline.
Finally, we have proposed the “Enhanced New Marketing Automation Model”
presented in chapter 5.6 which was designed based on the following existing marketing
automation model: the Forester model (see chapter 5.5.1), the Heimbach Model (5.5.2)
and the Carter’s model (5.5.3). Our newly enhanced model is highlighting two major key
component of Marketing Automation: The Data Management Platform and the
Campaign Management System which we hope gives the reader a better understanding
of its core functional scope.
The following chapter will be taking the reader across the research questions that we
have designed based on the “Enhanced New Marketing Automation model” in mind and
used for the interview of the qualitative content analysis.
47. 47
6 Research
6.1 Study design
The figure below shows the study design used to write this paper. The study design is
divided into three main phases. The first phase refers to the desk research and the
second phase refers to the qualitative research study. The third to the objectives of the
research, in particular it show the question that the content research analysis has
answer. We discuss the study design more in details in the following sub-chapters:
Illustration 6-1 - Study design for qualitative content research
6.1.1 Desk research
In order to gain a greater understanding of the research field and to build a base for the
qualitative research, the current state of research has been explored. The extracted
information forms the base for the qualitative research study.
48. 48
6.1.2 Qualitative Content Analysis
For the qualitative content research (QCA), a study semi-structured expert interview has
been chosen as the main method. According to Schreier (Schreier, 2012), qualitative
content analysis is a method for systematically describing the meaning of qualitative
material. It is done by classifying material as instances of the categories of a coding
frame. For our qualitative research, we have organized the work around these 10 steps
(Leicester, 2015).
Deciding on you’re a research question
Selecting the material
Copy and read the transcript - make brief notes in the margin when interesting
or relevant information is found.
Go through the notes made in the margins and list the different types of
information found
Read through the list and categorize each item in a way that offers a description
of what it is about.
Identify whether or not the categories can be linked any way and list them as
major categories (or themes) and / or minor categories (or themes).
Repeat the first five stages again for each transcript.
When we have done the above with all of the transcripts, collect all of the
categories or themes and examine each in detail and consider if it fits and its
relevance.
Review all of the categories and ascertain whether some categories can be
merged or if some need to them be sub-categorized
Finally the interpreting and presenting of findings.
Based on the research question and the literature research, an interview guide has been
written which can be found in appendix 7.1. The interview guide is characterized by open
questions, which allows and encourages personal answers to be given by the
interviewees (Mayer, 2013). We have taken notes during interviews (transcribing) or
observations and take a recording so that we can concentrate and listen and respond
better. Due to data protection of the interviewed companies, some recordings are
49. 49
treated confidentially. The transcriptions have been sent to each of the interviewees for
validation and can be found in appendices 7.1.
6.1.3 The researched questions
The objective of the thesis is to answer the below questions designed around the
development of the “Enhanced New Marketing Automation Model”, presented earlier
in chapter 5.6, the lead question being: “What are the required means, skills, tools and
processes for successful Marketing Automation, leading to marketing efficiency and
more revenue generation?”
Research question 1: What are the main reasons for implementing Marketing
Automation and the benefits? (see findings in chapter 6.2.1):
Research Question 2: What information and technology infrastructure is
required for successful Marketing Automation operations? (see findings in
chapter 6.2.2)
Research Question 3: How is the data stored and processed (analysed,
structured)? From what sources? (see findings in chapter 6.2.3)
Research Question 4: Who is in charge / organizational chart / sponsor for
Marketing Automation? (see findings in chapter 6.2.4)
Research Question 5: What are the most efficient Marketing Automation rules
or flows? (see findings in chapter 6.2.5)
Research Question 6: What are the success factors for operating Marketing
Automation? (see findings in chapter 6.2.6)
Research Question 7: What are the challenges and obstacles when
implementing Marketing Automation strategies? (see findings in chapter 6.2.7)
Research Question 8: What do you believe is up next for Marketing
Automation? (see findings in chapter 6.2.8)
6.1.4 Sampling
Part of the data used in this paper will been collected in performing interviews with
experts in automation in various industries. The principle is to represent diversity so that
50. 50
broad material of data can be captured from the interviewed professionals. Online
retailers are in general more interested in Marketing Automation than companies
without an online shop (Heimbach, 2015), therefore the focus was primarily companies
with e-commerce capabilities. The interviewers were selected based on their experience
of the topic and readiness to participate the survey. The questionnaire used for the
qualitative content research was designed according to the “Enhanced New Marketing
Automation Model” proposed in the chapter 5.6. The aim of the interview is to meet
various business-to-consumers (B2C) and business-to-business (B2B) companies’
professionals using Marketing Automation across industries of different sizes and
countries such as CRM, Retail e-commerce or marketing departments.
Illustration 6-2 – 12 interviewers’ sample list
51. 51
As the illustration 6.2 shows above, three companies are in Business to Business and
nine companies are qualified as Business to Consumers. One interview was conducted
with Marketing Automation consulting agency and one with a software editor.
6.2 Research findings
In this chapter we have used the content of all the twelve interviews and structured
the respondent’s answerers around our research questions and the coding used for
the qualitative content analysis. Detailed coded interviews can be found in appendix
1.1.
6.2.1 The root causes and benefits for implementing Marketing Automation
In this chapter, we have grouped all most cited reasons for implementing Marketing
Automation. Here was the researched question: “What root cause made you
implementing Marketing Automation and to answer what business question? For what
benefit?” We found out the following reasons that we have structured by topics:
Marketing Automation should be considered as fundamental component of any
customer centric strategy. It is most suitable for companies with large volume of
customer interactions across complex digital customer journeys and to improve
marketing efficiency processes. Increasing customer satisfaction, increasing incremental
sales to avoid storing data in silos was systematically quoted. “The idea behind
Marketing Automation is the marketing orchestration, in other words to combine all
communication channels in a consistent framework, so that you can interact in a
consistent way and really accompany your customers across channels with a consistent
message and at the end of the day this will be generating revenue (Interviewee_1,
2016)”. The customer journey being extremely complex today, it is important to be able
to offer the customer a unified and consistent shopping experience across medium. For
multichannel merchants, preferring online or offline is not the question but offering
both online and offline channels in one unified experience is the most important aspect.
At the end, the objective is the personalization of advertising and content. The right
52. 52
message to the right customer at the right moment in the right context is at the end the
objective.
Integration with sales is important for many of the respondents as well. Making sales
team happy in providing qualified traffic is certainly a fundamental motivation for
Marketing Automation operators, especially in the Business to Business area. “A reason
for implementing the Marketing Automation was the integration with the sales CRM and
sales force tool in order to manage and automatize more. So every day we can send the
top leads to our sales departments and they can see how hot they are. That is a really
great function that sales team really love because you are basically not shooting them
in the dark (Interviewee_9, 2016)”.
Data management and particularly the data quality management was mentioned as a
critical issue that should be taken as a strategic business question. The opportunity to
get first hand customer information or first party data was named as a key advantage,
even if data privacy is a growing concern. “The customer data storage might be threaten
by the legislator in the future, to protect data privacy more. The change in the safe
harbour laws is a concrete signal for it (Interviewee_9, 2016)”.
High efficiency and measurability of the marketing campaign is a key decision factor for
implementing Marketing Automation, in particular the ease of measuring digital return
on investment and understand what prospect of customer do. Company wants to be
capable to save money and resources by automating repetitive tasks in order to
reallocate money to most successful marketing campaigns. The repurchase rate was
quoted as important KPI to measure success, especially when merchant are operating in
the retail e-commerce business and need their website to be generating traffic or when
the buying process is particularly long.
Companies are searching to improve the ease of managing complexity such as multiple
layers of co-existing automated flows involving several thousand of contact. They want
to diminish the risks of errors. “At the end, Marketing Automation tries to capture
customers data into one system and set up an infrastructure to track behaviour and send
them some communication without doing too much manual work” (Interviewee_9,
53. 53
2016). “There are company that are launching several products, several lines in several
countries with limited resources and the only solution is to automate things”
(Interviewee_8, 2016).
Merchants also wants to educate customers to complex products or services which need
on-going education. “A TV 20 sec copy isn’t the appropriate medium for educating
customer about a product as opposed to a nurturing program, which can be activated
in a more appropriate manner and in a more detailed way (Interviewee_12, 2016)”.
6.2.2 System architecture
Since Marketing Automation essentially relies on technology, we thought we would put
together information about technical architecture recommended for Marketing
Automation. Our researched question was: “What information and technology
infrastructure is required for successful Marketing Automation operations?” This what
we found out:
Most interviewee agree to say that the primarily role of Marketing Automation is to do
the marketing campaign orchestration. “The CRM will be managing the customer
records and sales, the data warehouse will be the place where you will be doing your
data analysis, the data discovery, and consolidate data from different channels
(Interviewee_1, 2016)”. To have a solid customer referential is another very important
aspect. This can be a CRM, a data warehouse or even the e-commerce itself. The system
architecture should be divided in three layers: first the data layer on the down side,
which is relying on the Data Management Platform. Data management platform is used
essentially to reconcile different sources of data and for advertising in order to get the
most qualified placement. Data Management Platform is part of the automation tool
suite. It consolidates external and internal data such as first, second and third party data.
The orchestration layer would be the second layer which is the place where the business
people design more sophisticated campaigns and manage channels, timing, conditions
and messages to customers. Marketing Automation can offer value at every single steps
of the customer life cycle, consolidating new consumer data, building the relationship
with existing customers, re-engaging lapsing customers, reengaging dormant or
54. 54
removing unprofitable customers. Finally, the third layer is more a customer
communication layer or the visible part for the prospect such as website, CMS, e-
commerce, mobile etc. Some tools such as Maximizer1
allow the merchant to
personalize the content based on the scoring model. Finally, interviewees draws our
attention to the difficulty to manage big volume of data, leading to possible lower data
load performance.
6.2.3 Data Management
Marketing Automation works essentially with data. Data is the main asset to make
Marketing Automation work. Therefore we have designed the following research
question: “How do you store, structure and analyse the data? From what sources?” here
is what we found out:
The most challenging element is to find the right action for the collected data. “Data
collection itself is not the problem; data is available from any source of platform. It
makes analytics more efficient, and figure out what action to take is the challenge
(Interviewee_5, 2016)”. Here are some possible ideas to help. First it is recommended
to choose the right data model that meets the business requirement. This is a key
success factor that confirms our recommendation to focus first on the business question
seen in the chapter 5.5.2 before starting collecting data. Data model is like mapping
guidelines or data dictionary which help to design the database, facilitate its usage and
allows to segment or target with more accuracy during the campaign design phase. We
have observed that the definition of a customer lead varies from company to company.
Therefore the data mapping definition is important before starting the Marketing
Automation project. For example, some companies calls a lead a customer question and
some other calls it a customer. The data dictionary helps to avoid confusion. One ID for
each record should be defined in the data model in order to avoid duplications. A good
practice is to use the email address as unique ID. The centralization of the data is another
1
Maximizer: CRM Software: http://www.maximizer.com/
55. 55
very important success factor in building the data architecture. By centralization,
merchants mean connecting different data layer such as Data Management Platform,
CRM, Marketing Automation and Data Warehouse. This should help do better marketing
because the business owns more data and can get a 360° view of the consumer. It also
enables companies to enrich customer profile by putting together first, second and third
party data (see chapter 5.6.5) for detailed explanation. “In order to improve the data
quality and comprehensiveness, progressive profiling techniques can help companies in
getting the data in a non-aggressive way (Interviewee_1, 2016)”. Another innovating
method used to enrich first party data used by US interviewees is the integration of third
Party data to First party data. The idea is to enrich the company first party data in adding
third party data from data brokers.
Scoring is a key fundamental which Marketing Automation uses to segment dynamically
and in real time the customer base. The scoring is generally based on two or three
dimensions such as demographic, purchase history and overall online or offline activity.
The recency-frequency-monetary (RFM) method plays a major role in the way managers
look at the financial performance of a particular customer. That way, the business could
qualify customers in different segments based on the probability that they will buy the
product or engage with the brand. At the end, the business wants to be able to rate who
are the customers with the highest potential value. Not all interviewees are using scoring
methods. Some still use very basic segmentation methods. “In our company we tend to
take the whole database and divide it into 2-3 segments for campaigning. E.G Half for
male and female with two messages. We always try to reach out 100% of it. Our partners
in our sales organization are concerned they would lose opportunities when they
address their campaign to too narrow segments (Interviewee_4, 2016)”.
6.2.4 Organization & Processes
Marketing Automation is not just technology and a piece of software. Strategy, business
questions, processes, resources and people are needed on top to make it work. How
exactly? What is the best possible organization and processes to make it function? To
answer the question, we have asked interviewees the following question: “Who is in
56. 56
charge / organizational chart / sponsor for Marketing Automation?” We found out the
following:
The involvement of the company management is key. The role of the CEO in building the
digital customer centric vision is very important. “Finding the right sponsor in the
company is the essential challenge, someone that understand the value of Marketing
Automation (Interviewee_1, 2016)”. In fact, merchants said to be able to develop
Marketing Automation skills with an agency or develop the knowledge internally. At the
end, what is important is to be able to build a learning team, to keep themselves
informed about the evolution of Marketing Automation, keep up with the pace of
change. The software editor is often a critical team member in the sense that it can
support the business actively, in both training and consulting. For larger organization,
sharing the same Marketing Automation tool across organization and consequently
expertise is essential in order to find economy of scales and develop best practices
sharing. The collaboration across teams was cited as very important. This is indeed often
a very iterative process, which requires a lot of goes and returns, trials and errors until
the rule is adopted and works well. Often there is no formal process when it comes to
create automation flows or rules, but collaboration across functions and teams is
essential. New Marketing Automation opportunity should ideally be identified by
business analysts, which carefully analyse the data and make recommendations how to
setup the campaigns. “We test a lot thanks to our good project management. We avoid
big bangs. We develop in small steps which we can carefully monitor (Interviewee_4,
2016)”. Sometime leadership in in the hands of IT, sometime in marketing or even sales.
Teams involved in the Marketing Automation processes are generally structured around
several functions such as Digital Marketing (Analytics, social media, SEO experts,
webmaster, producers), Branded Marketing (Planning, media, creative, business
analyst), IT (Data base engineers, programmers, information architecture, ERP, Data
Warehouse) and Sales or Retail (CRM manager, Key account management, Retail
manager). In larger international organization, they are mirroring functions at both
corporate and local levels.