Big Data Research 2 (2015) 28–32

Contents lists available at ScienceDirect

Big Data Research

www.elsevier.com/locate/bdr

Demystifying Big Data Analytics for Business Intelligence Through

the Lens of Marketing Mix ✩
Shaokun Fan a,∗ , Raymond Y.K. Lau b , J. Leon Zhao b
a
College of Business, West Texas A&M University, Canyon, USA
b
Department of Information Systems, City University of Hong Kong, Hong Kong

a r t i c l e i n f o a b s t r a c t

Article history: Big data analytics have been embraced as a disruptive technology that will reshape business intelligence,
Received 8 November 2014 which is a domain that relies on data analytics to gain business insights for better decision-making.
Received in revised form 12 February 2015 Rooted in the recent literature, we investigate the landscape of big data analytics through the lens of a
Accepted 12 February 2015
marketing mix framework in this paper. We identify the data sources, methods, and applications related
Available online 18 February 2015
to five important marketing perspectives, namely people, product, place, price, and promotion, that lay
Keywords: the foundation for marketing intelligence. We then discuss several challenging research issues and future
Big data analytics directions of research in big data analytics and marketing related business intelligence in general.
Business intelligence © 2015 Elsevier Inc. All rights reserved.
Marketing intelligence
Marketing mix
Survey versus log data

1. Introduction to practitioners and researchers. Big data analytics is mainly con-

cerned with three types of challenges: storage, management, and
Recent technological revolutions such as social media enable processing [21]. For typical marketing intelligence tasks such as
us to generate data much faster than ever before [28]. The no- customer opinion mining, companies nowadays have many differ-
tion of big data and its application in business intelligence have
ent ways (social media data, transactional data, survey data, sensor
attracted enormous attention in recent years because of its great
network data, etc.) to collect data from a variety of information
potential in generating business impacts [12]. “Big Data” is defined
sources. Based on the characteristics of collected data, different
as “the amount of data just beyond technology’s capability to store,
manage and process efficiently” [21]. Big data can be characterized methods can be applied to discover marketing intelligence. Anal-
along three important dimensions, namely volume, velocity, and ysis models developed based on a single data source may only
variety [38]. provide limited insights, leading to potentially biased business de-
In marketing intelligence, which emphasizes the marketing- cisions. On the other hand, integrating heterogeneous information
related aspects of business intelligence, data relevant to a com- from different sources provides a holistic view of the domain and
pany’s markets is collected and processed into insights that sup- generates more accurate marketing intelligence. Unfortunately, in-
port decision-making [19]. Marketing intelligence has traditionally tegrating big data from multiple sources to generate marketing
relied on market surveys to understand consumer behavior and intelligence is not a trivial task. This prompts exploration of new
improve product design. For example, companies use consumer
methods, applications, and frameworks for effective big data man-
satisfaction surveys to study customer attitudes. With big data an-
agement in the context of marketing intelligence.
alytic technologies, key factors for strategic marketing decisions,
such as customer opinions toward a product, service, or company, We investigate different perspectives of marketing intelligence
can be automatically monitored by mining social media data [35]. and propose a framework to manage big data in this context. We
However, while accessibility to big data creates unprecedented first identify popular data sources for marketing intelligence per-
opportunities for marketing intelligence, it also brings challenges spectives. Then, we summarize the methods that are suitable for
different data sources and marketing perspectives. Finally, we give
examples of applications in different perspectives. The proposed
✩
This article belongs to Comp, bus & health sci.
framework provides guidelines for companies to select appropriate
*
Corresponding author.
data sources and methods for managing vital marketing intelli-
E-mail addresses: [email protected] (S. Fan), [email protected] (R.Y.K. Lau),
[email protected] (J.L. Zhao). gence to meet their strategic goals.

http://dx.doi.org/10.1016/j.bdr.2015.02.006
2214-5796/© 2015 Elsevier Inc. All rights reserved.
 S. Fan et al. / Big Data Research 2 (2015) 28–32 29

Fig. 1. A marketing mix framework for big data management.

2. A big data management framework terest from some or all units of a population using well-defined
concepts, methods and procedures, and compiles such informa-
The marketing mix framework is a well-known framework that tion into a useful summary form” [10]. Firms use surveys to col-
identifies the principal components of marketing decisions, and it lect data for various purposes, such as understanding customers’
has dominated marketing thought, research, and practice [6]. Bor- preferences and behaviors. For example, Apple has sent surveys
den [5] has been recognized as the first to use the term “marketing to customers who recently purchased an iPhone to gain feedback
mix” and he proposed a set of 12 elements. McCarthy [29] re- about their purchase and their experience with the product [16].
grouped Borden’s 12 elements to four elements or 4Ps, namely Log data is generated by information systems that capture trans-
product, price, promotion, and place. The 4P model has been con- actional records and user behavior [20]. For example, Walmart has
sidered to be most relevant for consumer marketing. However, it started to explore analyzing social media data to gain customer
has been criticized as being a production-oriented definition of opinions about the company or a particular product [7]. Log data
marketing, and researchers proposed a fifth P (people) [18]. We and survey data can be different in terms of size, quality, fre-
adopt the 5P model of the marketing mix framework in this paper quency, objectives, contents, and processing techniques [37]. The
because these perspectives play critical roles in developing suc- two data collection methods complement each other in various
cessful marketing strategies in the information age. business contexts. Surveys can be useful when we want to collect
In this paper, we propose a marketing mix framework to man- data on phenomena that cannot be directly observed. Log data are
age big data for marketing intelligence. This model classifies the preferred when real-time conclusions about users’ actual behavior
research in marketing intelligence into five perspectives accord- are required. The two methods can be combined when we want
ing to the marketing mix framework. Further, we identify common to study the relationship between user intention and user behav-
data, methods, and applications in each perspective and high- ior. There are advantages and disadvantages to both methods, and
light the dominating big data characteristic with respect to each we believe big data management should take both methods into
perspective. This framework provides guidelines for marketing consideration.
decision-making based on big data analytics. Fig. 1 is an overview
of the proposed big data management framework for marketing 2.2. Methods
intelligence. First, data from various sources are retrieved and uti-
lized to generate vital marketing intelligence. Second, a variety of Marketing intelligence refers to developing insights from data
analytics methods are applied to convert raw big data to actionable for marketing decision-making. Data mining techniques can help
marketing knowledge (intelligence). Finally, both data and meth- to accomplish such a goal by extracting or detecting patterns or
ods are combined to support marketing applications with respect forecasting customer behavior from large databases. According to
to each perspective of the marketing mix model. the data mining literature, common data mining methods include
association mining, classification, clustering, and regression [31].
2.1. Data We need to select appropriate data mining methods based on the
data characteristics and business problems [25].
Researchers use various methods to collect data, such as sur-
veys, interviews, focus groups, observations, and archives [2]. Note 2.3. Applications
that data collection methods are different from research methods.
For example, experiments are a widely used research method in 2.3.1. Customer segmentation and customer profiling
marketing, but researchers rely on surveys, observations, or in- For effective marketing, it is essential to identify a specific
terviews to collect experimental data [27]. Surveys and logs are group of customers who share similar preferences and respond
the two most common methods to acquire data for business in- to a specific marketing signal. Customer segmentation applications
telligence [22]. A survey is defined as “collecting information in can help identify different communities (segments) of customers
an organized and methodical manner about characteristics of in- who may share similar interests. Kim et al. [23] proposed cluster-
30 S. Fan et al. / Big Data Research 2 (2015) 28–32

ing customer groups with respect to lifecycle characteristics. Usu- both customer reviews and promotions [26]. The authors found a
ally, various clustering and classification techniques are applied to substitute relationship between the WOM volume and coupon of-
customer segmentation and user profiling. However, customer seg- ferings, but a complementary relationship between WOM volume
mentation is becoming increasingly challenging under a big data and keyword advertising. Promotional marketing analysis can also
environment. For instance, to differentiate among customer groups include factors from other perspectives, such as price and place.
for telecommunication applications, it is necessary to analyze their For example, enabled by mobile technologies and location-based
call data apart from their demographics [1]. The volume of call services, companies can use customers’ location information to im-
data is huge (e.g., the communication time between each pair of prove their promotion strategy and select targeted customers.
customers on each day), and a variety of data should be taken into To improve product awareness and promote products to poten-
account (e.g., both qualitative demographic data and quantitative tial customers, recommender systems have been widely used in
call records). In fact, for the most fine-grained targeted marketing the e-commerce context [15]. User rating-based collaborative fil-
(e.g., one-to-one marketing), we are not talking about identifying tering methods or content-based association mining methods are
groups of similar customers, but the “profiling” of each individual commonly applied to develop recommender systems. However, ex-
customer such that the most suitable products/services are mar- isting methods may not scale up to big data. For instance, given N
keted to the most appropriate individual given a steam of customer user ratings, the general computational complexity of a collabora-
service consumption data generated in real-time [1]. tive filtering method is N 2 [9]. Therefore, it is quite challenging to
scale up existing recommender systems to cope with big data (e.g.,
2.3.2. Product ontology and product reputation management N = tens of millions) and generate appropriate recommendations
To alleviate the shortcoming of retrieving limited product rep- to potential customers in real-time as expected in e-commerce
utation via survey data, Morinaga et al. [30] developed an auto- settings. This is the reason why “velocity” is one of the most chal-
matic framework to monitor the reputation of a variety of products lenging issues for the “promotion” perspective in the context of
by mining Web contents. Clustering and association mining tech- marketing intelligence.
niques are among the most common methods employed to support
reputation management applications. More recently, Di et al. [14] 2.3.4. Pricing strategy and competitor analysis
proposed a reputation management method which not only mines There has been much research on what pricing strategies man-
text-based reputation data from the Web but also considers the agers should follow under various situations. Traditionally, empir-
graphical images of products posted to the Web. Nevertheless, by ical research on pricing strategies uses survey data and regression
the time of this writing, twenty billion images have been uploaded methods. For example, researchers used a national mail survey to
to Instagram.1 Given such an extraordinary size of images archived study the determinants of pricing strategies [32]. They found dif-
online, it is extremely challenging to analyze the sheer volume of ferent pricing strategies are preferred under different marketing
images for product reputation management, not to mention the situations. The growth of e-commerce has made price informa-
variety of formats of source data (e.g., text versus images). To tion available on websites and researchers started using log data
carry out an automatic analysis of the textual comments posted to study pricing strategy in e-commerce websites. For example, a
to the Web for product reputation management, it is essential recent study uses a method to estimate demand levels from sales
to develop a rich computer-based representation of product in- rank and derive demand elasticity, variable costs, and the optimal-
formation for subsequent product reputation analysis. Recently, an ity of pricing choices directly from publicly available e-commerce
automated product ontology mining method that is underpinned data [17]. Based on the data derived from various log data sources,
by latent topic modeling has been explored to build product on- they can study the optimality of price discrimination. While re-
tologies based on textual descriptions of products extracted from gression methods are widely used for price prediction applications,
online social media [24]. The automatically constructed product association mining methods are applied to competitor analysis ap-
ontologies can be used as the basis to support product reputation plications. An automated competitor analysis application does not
management applications and other marketing intelligence appli- simply identify the potential competitors of a company; it also
cations. However, given the computational complexities involved effectively discovers the potentially competitive products and the
in automated product ontology extraction from online social me- product contexts [3]. This type of application has proven useful to
dia, new computational methods must be developed to cope with facilitate the “price” aspect of the marketing mix model. However,
the volume, velocity, and variety issues of big social media data. the sheer volume of product pricing information on the Web has
also posed new challenges to scale up existing applications with
2.3.3. Promotional marketing analysis and recommender systems big data.
In the increasingly competitive business environment, billions
of dollars are spent on promotions each year [34]. Thus, promo- 2.3.5. Location-based advertising and community dynamic analysis
tional marketing analysis has attracted a lot of attention from prac- Place is also an important dimension in marketing analysis. Re-
titioners and researchers. Effective promotional strategies are one search on place-based marketing focuses on the impact of places
of the key success factors for companies to increase their sales and on marketing strategies. For example, researchers used a survey
revenue [4]. Promotional data usually includes information about to collect customer data and study different levels of place-based
promotion types (price cut or coupons), promotion time, and pur- marketing in the form of region of origin strategies used by winer-
chase records during the promotional period. Early work related ies in their branding efforts [8].
to promotional marketing analysis mostly focused on analyzing With the widespread use of mobile technology, location-based
how different types of customers respond to different promotional services (LBS) can provide users personalized information in a spe-
strategies or how different categories of products affect the effec- cific location at a specific time. Location-based advertising has
tiveness of promotional strategies [33]. Most existing work uses been proposed as an efficient marketing strategy [13,27]. Location
regression methods to study promotions in different contexts [4]. is one of the most important solutions to meet consumers’ need
In the big data environment, more log data becomes/is available and it is a valuable source for personalized marketing informa-
for promotion analysis. A recent work studied WOM derived from tion. In location-based advertising, customers can get timely ad-
vertisements or product recommendations based on their current
position or predicted future position. Location-based advertising
1
http://instagram.com/press/#. provides a new tool for companies to attract more customers and
 S. Fan et al. / Big Data Research 2 (2015) 28–32 31

enhance brand value. One challenge for location-based advertising dia data can be used to study customer opinions toward a
is how to accurately predict customers’ locations. Both spatial and company or a product. However, data collection and analysis
temporal data should be taken into consideration (temporal mov- methods may be different due to different structure, quality,
ing pattern mining for location-based service). We need to process granularity, and objective. Further, survey data and log data
a large volume of spatial and temporal data within a short time can also be used to study the same marketing problem. How
period before customers move to new locations. Thus, the “veloci- to conduct surveys in social media and confirm the survey re-
ty” issue of big data is also one of the most challenging aspects for sult with log data in social media will become an important
location-based advertising. topic in e-commerce research and applications.
Researchers explored the log data in location-based social net- 5. Examine how to balance investments in marketing intelligence
works to uncover user profiles; these automatically discovered user techniques. Big data-enabled marketing intelligence will be-
profiles have the potential to be subsequently applied to location- come a competitive source for consumer behavior and product
based targeted marketing [36]. Regression and classification meth- planning; therefore all companies must invest in big data in-
ods are often utilized for location-based marketing applications. In frastructure including data scientists and big data platforms.
another study, Castro et al. [11] leverage the GPS traces of individ- 6. Explicate how to refine the framework as the big data tech-
uals to uncover the location-based dynamics of different commu- nology evolve continuously. Data of a variety of formats and
nities. Through analyzing the dynamics of local communities, it is qualities will continue to grow and be digitized. Even though
possible to predict their changing product/service preferences. As peta-scale data (e.g., petabytes of customer records) may be
a result, effective marketing strategies can be developed with re- considered big data now, the same volume of data may not
spect to both the place and time dynamics of a group of customers. be considered big in a few years. It is important to continu-
Nevertheless, this type of application needs to deal with both the ously refine the framework, methods, and techniques that we
“variety” and “velocity” issues of big data. For instance, both the discuss in this paper in order to meet the challenges for more
relational data among users in location-based social networks and advanced business intelligence in the next generation of big
GPS signals need to be analyzed to uncover the location-based dy- data management.
namics of a local community. In addition, since individuals may
constantly move around different places, location-based marketing Acknowledgements
applications must be able to respond quickly in order to maintain
the location sensitivity with respect to the constantly moving cus- Dr. Lau’s work is partly supported by a grant from the Science
tomers. Technology and Innovation Committee of Shenzhen Municipal-
ity – Basic Research Program (Project: JCYJ20140419115614350).
3. Future research directions Dr. Zhao’s work has been partially supported with the NSFC project
#71110107027 and the SAP Grant – CityU Project No. 9220059.
We propose to use a marketing mix framework for guiding
research in big data management for marketing intelligence. We References
identify the data sources, methods, and applications in different
[1] M.A. Abbasoğlu, B. Gedik, H. Ferhatosmanoğlu, Aggregate profile clustering for
marketing perspectives. We further discuss the challenging issues telco analytics, Proc. VLDB Endow. 6 (12) (2013) 1234–1237.
related to big data management in the context of various mar- [2] W.G. Axinn, L.D. Pearce, Mixed Method data Collection Strategies, Cambridge
keting perspectives. Based on the framework, we highlight future University Press, 2006.
research directions in big data management. [3] S. Bao, R. Li, Y. Yu, Y. Cao, Competitor mining with the web, IEEE Trans. Knowl.
Data Eng. 20 (10) (2008) 1297–1310.
[4] D.R. Bell, J. Chiang, V. Padmanabhan, The decomposition of promotional re-
1. Study how to select appropriate data sources for particular sponse: an empirical generalization, Mark. Sci. 18 (4) (1999) 504–526.
goals. The amount of available data is increasing. Current tech- [5] N.H. Borden, The concept of the marketing mix, J. Advert. Res. 4 (2) (1964) 2–7.
niques do not allow us to process all data available in a timely [6] F. Brassington, S. Pettitt, Principles of Marketing, FT Prentice Hall, 2005.
[7] E. Brown, Mining the social data stream for deeper customer insight, http://
manner. Thus, data selection is a critical decision for manag- www.zdnet.com/, 2012.
ing marketing intelligence. How to select data that can provide [8] J. Bruwer, R. Johnson, Place-based marketing and regional branding strategy
the most value to business decision-making requires future perspectives in the California wine industry, J. Consum. Mark. 27 (1) (2010)
research on the alignment between data and marketing intel- 5–16.
[9] Y. Cai, R.Y. Lau, S.S. Liao, C. Li, H.-F. Leung, L.C. Ma, Object typicality for effective
ligence goals.
Web of Things recommendations, Decis. Support Syst. 63 (2014) 52–63.
2. Analyze how to select appropriate data analysis methods. [10] Survey Methods and Practices, Statistics Canada, 2010.
There are many types of methods that can be used to pro- [11] P.S. Castro, D. Zhang, C. Chen, S. Li, G. Pan, From taxi GPS traces to social and
cess data. Given a particular data set, many methods may be community dynamics: a survey, ACM Comput. Surv. 46 (2) (2013) 17.
[12] H. Chen, R.H. Chiang, V.C. Storey, Business intelligence and analytics: from big
applicable. Regression and classification are usually used for
data to big impact, MIS Q. 36 (4) (2012) 1165–1188.
prediction, while clustering and association rule mining are [13] S. Dhar, U. Varshney, Challenges and business models for mobile location-based
used for description. Further, big data brings issues such as services and advertising, Commun. ACM 54 (5) (2011) 121–128.
imbalanced data distribution and large number of variables, [14] W. Di, N. Sundaresan, R. Piramuthu, A. Bhardwaj, Is a picture really worth a
thousand words?: on the role of images in e-commerce, in: Proceedings of the
which cannot be efficiently handled by existing data mining
7th ACM International Conference on Web Search and Data Mining, ACM, 2014,
methods. We need to improve existing methods to increase pp. 633–642.
the efficiency and accuracy. [15] M.B. Dias, D. Locher, M. Li, W. El-Deredy, P.J.G. Lisboa, The value of personalised
3. Inquire how to integrate different data sources to study com- recommender systems to e-business: a case study, in: Proceedings of the 2008
plicated marketing problems. Most existing studies use data ACM Conference on Recommender Systems, ACM, Lausanne, Switzerland, 2008,
pp. 291–294.
from one single data source. However, some complicated busi- [16] D. Etherington, Apple sends out iPhone survey, seeks feedback on Android,
ness problems require combining data from different sources. touch ID and more, http://techcrunch.com/, 2014.
For example, in order to study the impact of social media be- [17] A. Ghose, A. Sundararajan, Evaluating pricing strategy using e-commerce data:
havior on purchase behavior, we may need to combine social evidence and estimation challenges, Stat. Sci. (2006) 131–142.
[18] C.L. Goi, A review of marketing mix: 4Ps or more?, Int. J. Mark. Stud. 1 (1)
media data and transaction records.
(2009) 2.
4. Investigate how to deal with the heterogeneity among data [19] H. Hedin, I. Hirvensalo, M. Vaarnas, The Handbook of Market Intelligence: Un-
sources. For example, both customer reviews and social me- derstand, Compete and Grow in Global Markets, John Wiley & Sons, 2014.
32 S. Fan et al. / Big Data Research 2 (2015) 28–32

[20] A. Jacobs, The pathologies of big data, Commun. ACM 52 (8) (2009) 36–44. [30] S. Morinaga, K. Yamanishi, K. Tateishi, T. Fukushima, Mining product reputa-
[21] S. Kaisler, F. Armour, J.A. Espinosa, W. Money, Big data: issues and challenges tions on the web, in: Proceedings of the Eighth ACM SIGKDD International Con-
moving forward, in: 6th Hawaii International Conference on System Sciences, ference on Knowledge Discovery and Data Mining, ACM, 2002, pp. 341–349.
(HICSS), 2013, IEEE, 2013, pp. 995–1004. [31] E.W. Ngai, L. Xiu, D.C. Chau, Application of data mining techniques in customer
[22] M. Kaptein, P. Parvinen, E. Poyry, Theory vs. data-driven learning in future relationship management: a literature review and classification, Expert Syst.
e-commerce, in: 46th Hawaii International Conference on System Sciences, Appl. 36 (2) (2009) 2592–2602.
(HICSS), 2013, IEEE, 2013, pp. 2763–2772. [32] P.M. Noble, T.S. Gruca, Industrial pricing: theory and managerial practice, Mark.
[23] S.-Y. Kim, T.-S. Jung, E.-H. Suh, H.-S. Hwang, Customer segmentation and strat- Sci. 18 (3) (1999) 435–454.
egy development based on customer lifetime value: a case study, Expert Syst. [33] K. Pauwels, D.M. Hanssens, S. Siddarth, The long-term effects of price promo-
Appl. 31 (1) (2006) 101–107. tions on category incidence, brand choice, and purchase quantity, J. Mark. Res.
39 (4) (2002) 421–439.
[24] R.Y. Lau, C. Li, S.S. Liao, Social analytics: learning fuzzy product ontologies for
[34] S. Srinivasan, K. Pauwels, D.M. Hanssens, M.G. Dekimpe, Do promotions benefit
aspect-oriented sentiment analysis, Decis. Support Syst. 69 (2014) 80–94.
manufacturers, retailers, or both?, Manag. Sci. 50 (5) (2004) 617–629.
[25] S.-H. Liao, P.-H. Chu, P.-Y. Hsiao, Data mining techniques and applications–
[35] W. Tan, M.B. Blake, I. Saleh, S. Dustdar, Social-network-sourced big data analyt-
a decade review from 2000 to 2011, Expert Syst. Appl. 39 (12) (2012)
ics, IEEE Internet Comput. 17 (5) (2013) 62–69.
11303–11311.
[36] M.A. Vasconcelos, S. Ricci, J. Almeida, F. Benevenuto, V. Almeida, Tips, dones
[26] X. Lu, S. Ba, L. Huang, Y. Feng, Promotional marketing or word-of-mouth? Evi- and todos: uncovering user profiles in foursquare, in: Proceedings of the Fifth
dence from online restaurant reviews, Inf. Syst. Res. 24 (3) (2013) 596–612. ACM International Conference on Web Search and Data Mining, ACM, 2012,
[27] X. Luo, M. Andrews, Z. Fang, C.W. Phang, Mobile targeting, Manag. Sci. 60 (7) pp. 653–662.
(2013) 1738–1756. [37] J.L. Zhao, S. Fan, D. Hu, Business challenges and research directions of manage-
[28] A. McAfee, E. Brynjolfsson, T.H. Davenport, D. Patil, D. Barton, Big data, the ment analytics in the big data era, J. Manag. Anal. 1 (3) (2014) 169–174.
management revolution, Harv. Bus. Rev. 90 (10) (2012) 61–67. [38] P. Zikopoulos, C. Eaton, Understanding Big Data: Analytics for Enterprise Class
[29] E.J. McCarthy, Basic Marketing, Richard D. Irwin, Homewood, IL, 1964. Hadoop and Streaming Data, McGraw–Hill/Osborne Media, 2011.