Increasing customer loyalty in internet marketing

Increasing customer loyalty in internet marketing

Department of Information Management, Chaoyang University of Technology, Taichung, Taiwan

{lschen, s10114616}@cyut.edu.tw

Abstract. In recent years, with rapid development of social networking web-sites, more and more travelers make their travel and accommodation decisions by referring to online comments (electronic word of mouth). It’s especially true for the customers who live in bed and breakfast. But, due to the limited market-ing budget, a bed and breakfast (B&B) enterprise needs an effective and cheap to promote their products and services through internet marketing. Social media marketing could one of cheap and powerful internet advertising channels. How-ever, most of bed and breakfast enterprises lack sufficient human resource and time to interact to the online users of social networking websites. Moreover, there are lots of social media marketing techniques, but we don’t know which one is crucial for a bed and breakfast enterprise. Therefore, this study aims to define the key factors of social media marketing, and then use decision tree to identify the important factors for increasing customers’ loyalty. A survey of so-cial media marketing in Facebook will be provided to demonstrate the effec-tiveness of our utilized methods.

Keywords: Internet Marketing, Electronic Word of Mouth, Decision Trees, Feature Selection

1

Introduction

With rapid development of social networking websites such as Facebook, Trip Advisor, and Twitter, more and more travelers make their travel and accommodation decisions by referring to online comments (electronic word of mouth) [27]. It’s espe-cially true for the customers who live in bed and breakfast which has been of major travel trends in Taiwan according to the 2012 survey of Tourism Bureau, Taiwan. But, due to the limited marketing budget, a bed and breakfast (B&B) enterprise needs an effective and cheap to promote their products and services through internet market-ing.

In addition, social networking websites has become an important source of in-formation. Travelers will share their travel experiences to others and respond to relat-ed comments. This many-to-many interaction spread through the internet can be called electronic word of mouth (e-WOM) [6, 7]. This kind of trusted information has been produced by the trusting relationship between/among community members. In

Taiwan, 75% of consumers trust the opinions of Internet users. Before making their purchase decisions, about 80% of consumers will refer to other comments on the In-ternet and as a basis for future purchase reference [12, 16, 18]. Many companies also interact to consumers and understand their needs by social media. So, we can say that the rapidly growing popularity of social networking sites has changed the traditional marketing content [24].

According to the report of 2013 social media marketing industries [25], the most frequently used social media is Facebook (92%), then are Twitter(80%), Linkedln (70%), Blog (58%), and Youtube(56%). The current status of the use of social media marketing, 79% of marketing companies integrate social media into traditional mar-keting. The benefits about social media marketing, 89% respondents said the efforts they made on social media can generate more exposure for businesses. 75% think there is a positive effect of increased customer traffic, again providing market infor-mation (69%), the establishment of fan loyalty (65%), resulting in potential customers (61%), etc. However, sales are driven by long-time relationships. Among those who have more than three years of experience in the social media industry, only 50% said social media could assist them to increase sales. And the performance of social media marketing is proportional to the numbers of hours which weekly input by companies. Related researches have shown that travelers will choose a tourist destination by searching others’ experience via Facebook to compare features of different Bed and Breakfast. Thus, to interact with consumers through social media and on the field of tourism including B & B industry has become increasingly important [14].

However, limited by firm size, B & B industry generally lacks sufficient human resources and time for communicating to consumers, cultivating potential customers, and conduct social media marketing business. Therefore, understanding the important techniques of social media marketing practices and critical service quality can help B & B companies to make precise social media marketing decisions and obtain the voice of the customers.

Consequently, this study aims to define the factors (techniques) of social media marketing in B & B industry, and then utilize decision trees (DT) to recognize the crucial factors for building loyalty, respectively. Once we know the important factors of social media marketing, B & B enterprises can use their limited resource to do effective marketing and improve service quality. A survey of social media marketing in Facebook will be provided to demonstrate the effectiveness of our utilized meth-ods.

2

Related Works

2.1 Internet Marketing

Social media is to achieve the purposes of dissemination by interaction so that people can share information and give comments, establish circle of friends, maintain relationships, and communicating to others [11, 21, 22, 40]. Following the growth of web 2.0 websites such as Blog, Twitter (information exchanging and sharing); Youtube (vedio and photo sharing); Wikipedia (knowledge sharing), this trend has

continuously increases [11, 37]. Yu et al. [37] compared 4 types of media (blogs, forum, Twitter, conventional news) and they found social media outperforms conven-tional media.

Eduardo et al. [11] found that tourists using virtual communities to exchange views and experiences already exist for more than 10 years. Tourists will obtain in-formation, learn the experiences of others, and compare tourism-related services be-fore travelling or during the trip. Tan [32] mentioned that there are three motivating factors of searching travel-related information, the exchange of information, enter-tainment, and social networking. Because travelers use the Internet to communicate with each other and share travel-related information, they can do detailed travel plans by understanding tourism destination through others shares [8, 22, 35]. Laroche et al. [20] developed a model to explain how social media community to strengthen the relationship between the brand and the customers. Boley et al. [3] compared the dif-ference of visitors’ behaviors between the websites which did post pictures and those which didn’t post pictures, when they buy souvenirs. The discovered information allows website owners to develop strategies to stimulate the effectiveness of their marketing. In the work of Tomas & Elena [29], they utilized modified balance score card to evaluate the hotel websites. They confirmed that the hotel website can build long-term customer relationships through the hotel’s policy. Jacques et al. [15] uti-lized “hotel stars” and “hotel facilities (restaurant, business center, gym, free wireless internet)” to predict roads, destination types and etc.

To sum up, social media marketing must build a persuasive rhetoric, through the exchange of experience and quality, to influence customer loyalty and maintain long-term customer relationship [2]. However, these available literatures didn’t focus on B & B industry. In addition, they cannot tell the enterprises how to use social media marketing and which social marketing techniques are important when they don’t have enough time, budget and human resource. Therefore, it is necessary to discover the important factors which influence consumers’ behaviors for tourism enterprises.

2.2 Feature Selection

You et al. [36] mentioned that there are three types of feature selection proaches. They are filters, wrappers, and embedded approaches. The filter based ap-proaches are simple and quick. The wrappers based methods are dependent on selec-tion funcselec-tions in the feedback and learning of classifiers. Embedded approaches uti-lize machine learning methods to evaluate the selected feature subset.

Considering the purpose of being easily used and computational cost, this study uses decision trees. Decision tree is one of most popular supervised machine learning methods [17]. It belongs to embedded feature selection approach. When using deci-sion tree method to be feature selection tool, the nodes in the constructed tree will be considered as important. Other features will be removed. DT is also widely applied to travel areas. For examples, Tyrvainen et al. [30] studied the traveler’s preferences and select tourist destination. They used DT to find the factors which influence travelers’ selection to develop overall planning of resorts, to enhance eco-efficiency and sus-tainable development. Kim & Upneja [19] used decision tree (C4.5) in the restaurant which is in financial difficulties to accurately predict financial distress. Duchessi &

Lauria [10] also utilized the decision tree model (CART) to predict the internet mar-keting performance of ski resorts to make advertising strategies, to enhance brand awareness, and to attract customers. These studies have shown that the decision tree method can be successfully used for feature selection.

3

Approach

This section describes the research methods and procedure. The study flow chart shown in Figure 1 can be divided into six main steps. The implementation produce will be illustrated by following instructions.

Step 1: Define the factors of social media marketing

From the available literatures, we try to identify and define the potential social media marketing elements. Then, we modified their original definitions for B & B industry. They constitute factors investigated in this work.

Step 2: Design and pretest the questionnaire

This study used a questionnaire as a measurement tool. Based on defined factors defined in step 1, we design the questionnaire and this questionnaire will be employed to survey respondents’ views regarding the importance of every factors. Before issu-ing a formal questionnaire, in order to avoid beissu-ing misunderstood the meanissu-ing of the respondents to affect the validity of the questionnaire, a pilot questionnaire will be pretested. We use one by one to have interview with respondents who should have the experiences of living in B&B and participating in social media.

Fig. 1. The implemental procedure of employed approaches

Step 3: Collect data

The completed questionnaire will be distributed to target respondents who have lived in B&B and the B&B enterprisers. In order to select valid samples, the internet and hardcopy questionnaires sent simultaneously to the respondents who lived in different areas such as northern, central and eastern Taiwan, and different age groups.

1. Define the factors of social media mar-keting factors

2. Design and pretest the questionnaire

3. Collect data

4. Implement decision tree

Step 4: Implement feature selection

In this step, we attempt to select important factors from the collected data. We use the data of the importance items (defined social media marketing factors) as our input data (attributes) and the labels “the probability of revisiting if B&B fulfills the important factors which respondents answered in questionnaire” as our output data. Then, we can construct decision trees for selecting the important factors. The detailed implemental process could be found as following sub-steps.

Step 5.1 Data preparation

Use 10-fold cross validation experiment and construct a decision tree for each fold of data. In other words, the collected data set will be divided into 10 equal sized sets and each set is then in turn used as the test set. Beside test set, we use other 9 sets as our training set to build decision trees. Therefore, we will have 10 trees.

Step 5.2 Determine the input and output variables

Step 5.3 Construct decision trees following C5.0 algorithm for each fold data set Step 5.3.1 Create an initial rule tree

Step 5.3.2 Prune this tree

Step 5.3.3 Process the pruned tree to improve its understandability

Step 5.4 Pick a tree whose performance is the best among all constructed trees Step 5: Draw conclusions

Finally, from the results of step 4, we find the important factors of social media marketing in B&B industry. And then we can draw conclusions based on them.

4

Results

Table 1 shows the 16 defined social media marketing factors. Totally there are 1200 questionnaires have been issued. 210 examples are returned for further analysis. Among these collected samples, 17 examples are from B&B enterprises and others (193 examples) responded by customers. In all samples, males accounted for 45%, female 55%; samples were mainly aged 18 to 30 years old (51%), and then 31~40 years old (21%) and above 41 years old (25%); the majority (50%) samples’ monthly income is from 20,000 to 50,000 NTD; the source of B & B information comes from shared information from social media (49%); the main reason (38%) of living in B&B is “unique characteristics”; 54% respondents spent less than 3 hours to browse social networking sites. We separate the customers and B&B enterprises to show their back-ground in table 2. From this table, we can find some major differences between cus-tomers and B&B enterprisers. For examples, B&B enterprises think “unique charac-teristics” is only one motivation for attracting customers. But, except this motivation, customers think “pricing” is another crucial motivation. Besides, compared to cus-tomers, B&B enterprises obviously spent too little time to interact to members of social networking websites.

Table 1. The defined social media marketing factors in B&B industry No. Factors Supports No. Factors Supports Q1 Fan page [13] Q9 Aesthetics and

visual quality [9, 39] Q2 Group [5, 13] Q10 Interaction quality [34] Q3 Event [13] Q11 Functional needs [2, 32] Q4 Adver-tisement [13, 33] Q12 Psychological and hedonic [32] Q5 Market-place [13] Q13 Altruism [11, 26] Q6 Beacons and Polls [4, 9] Q14 Socialization and learning [23] Q7 Applica-tions [28] Q15 Relaxation [23, 38] Q8 Selection of fea-tures [1, 21] Q16 Novelty [31, 38]

Table 2. The background of collected data

Variable Customers B&B enterprises

Gender Male 47% 24% Female 53% 76% Age <18 2% 6% 18~30 53% 24% 31~40 20% 35% >40 24% 35% Income <20,000 NTD 34% 24% 20,000~50,000 NTD 50% 53% 50,000~100,000 NTD 14% 24% >100,000NTD 3% 0% Source of B & B information Keyword advertisements 9% 18% Newspapers & magazines 34% 24% Social media 48% 53% Others 9% 6% Motivation for living B&B Pricing 35% 18% Unique characteristics 36% 65% Word of mouth 16% 12% Natural view and

land-scapes 10% 6% Others 3% 0% Spent time in social networking websites <3hrs 51% 88% 3~6hrs 39% 12% 6~9hrs 7% 0% >9hrs 3% 0%

Table 3. Performances of 10-fold cross validation experiment

Fold #1 #2 #3 #4 #5 #6 #7 #8 #9 #10 Errors 33.3% 28.6% 52.4% 38.1% 42.9% 52.4% 33.3% 38.1% 33.3% 28.6% Rule

number 14 7 14 14 11 14 16 5 8 16

Table 4. The extracted knowledge rules for the tree with the best performance (fold #2)

No. Rules

1 IF Q8 <= 1 AND Q9 > 2 AND Q15 > 2 THEN Class=HIGH [0.750] 2 IF Q4 <= 1 AND Q8 <= 1 AND Q13 > 2 THEN Class=HIGH [0.750] 3 IF Q8 <= 1 AND Q13 <= 2 THEN Class=HIGH [0.508]

4 IF Q6 > 1 AND Q6 <= 2 AND Q8 > 2 THEN Class=MEDIUM [0.857] 5 IF Q8 > 1 AND Q8 <= 2 THEN Class= MEDIUM [0.747]

6 IF Q15 > 2 THEN Class= MEDIUM [0.710]

7 IF Q6 > 2 AND Q8 > 2 AND Q10 <= 2 AND Q14 > 1 THEN Class=LOW [0.889] Note:

(1) The value of attributes (Q1~Q16) “1,2,3,4,5” means “very important, important, neutral, unimportant, not very important”.

(2) Class = “the probability of revisiting if B&B fulfills the important factors which respond-ents answered in questionnaire”

Next, we implement 10 fold cross validation experiments. Therefore, there are 10 trees have been built. And they are listed in table 3. From this table, we can find fold #2 that has the best performance (the lowest error rate). Consequently, this tree has been picked to select features. Table 4 summarizes all extracted knowledge rules by this decision tree. There are seven rules and the attributes left in the nodes will be considered as important. We also divided the all samples into “customers” and “B&B enterprisers” to find the differences. In table 5, we can find 8 important factors from 16 candidate attributes. They are “Q4(Advertisement)”, “Q6(Beacons and Polls”, “Q8(Selection of features)”, “Q9(Aesthetics and visual quality)”, “Q10(Interaction quality)”, “Q13(Altruism)”, “Q14(Socialization and learning)”, and “Q15(Relaxation)”.

Table 5. Summary of selected important factors of social media marketing All data Customers B&B enterprisers Selected im-portant factors Q4, Q6, Q8, Q9, Q10 Q13, Q14, Q15 Q1, Q2, Q3, Q6, Q8 Q9, Q10, Q12, Q14 Q15 Q15 Best perfor-mance (Over accuracy) 71.4% 73.7% 100%

5

Conclusions

This study aims to define potential social media marketing factors for B & B in-dustry. From the defined 16 candidate factors, we use feature selection approaches to find the crucial factors. Customers think “Q4(Advertisement)”, “Q6(Beacons and Polls”, “Q8(Selection of features)”, “Q9(Aesthetics and visual quality)”, “Q10(Interaction quality)”, “Q13(Altruism)”, “Q14(Socialization and learning)”, and “Q15(Relaxation)” are the key so-cial media marketing factor for increasing customer royalty.

In the future research directions, readers can find aother new feature selection to analyze sample data, and then compare the difference between consumers and B&B enterprisers. To discover the true consumers’ thinking about social media marketing can bring financial benefits. In addition, the investigation can be conducted for differ-ent industry types and other social media.

Acknowledgement

This work was financially supported in part by National Science Council of Tai-wan (Grant No. NSC 101-2628-E-324-004-MY3). Authors express our deeply appre-ciation to the funding.

References

1. Albee, A., eMarketing Strategies for the Complex Sale, McGraw-Hill, 2010.

2. Andersen, P. H., Relationship marketing and brand involvement of professionals through web-enhanced brand communities: The case of Coloplast, Industrial Marketing Manage-ment, 2005, 34(1): 39-51.

3. Boley, B. B., Magnini, V. P., Tuten, T. L., Social media picture posting and souvenir pur-chasing behavior: Some initial findings, Tourism Management, 2013, 37: 27-30.

4. Boyd, D., Ellison, N., Social network sites: definition, history, and scholarship, Journal of Computer-Mediated Communication, 2007, 13(1): 210-230.

5. Bressler, S. E. and Grantham, C., Communities of commerce: Building internet business communities to accelerate growth, minimize risk, and increase customer loyalty, McGraw-Hill, Inc. New York, NY, USA, 2000.

6. Chen, Y. C., Shang, R. A., Li, M. J., The effects of perceived relevance of travel blogs’ content on the behavioral intention to visit a tourist destination, Computers in Human Be-havior, 2014, 30: 787-799.

7. Cantallops, A. S., Salvi, F., New consumer behavior: A review of research on eWOM and hotels, International Journal of Hospitality Management, 2014, 36: 41-51.

8. Chung, J. Y., and Buhalis, D., A study of online travel community and Web2.0: Factors af-fecting participation and attitude, In Proceedings ENTER2008, Innsbruck, Springer-Verlag, Wien, 2008, pp. 267-278.

9. Cross, R. L., Parker, A., The hidden power of social networks: Understanding how work really gets done in organizations, Harvard Business Review Press, 2004, pp. 304.

10. Duchessi, P., Lauria, E. J.M., Decision tree models for profiling ski resorts’ promotional and advertising strategies and the impact on sales, Expert Systems with Applications, 2013, 40(15): 5822-5829.

11. Eduardo, P. L., Jacques, B. G., Desiderio, G. T., Ricardo, D. A., Intentions to use social media in organizing and taking vacation trips, Computers in Human Behavior, 2011, 27(2): 640-654.

12. Hu, N., Bose, I., Koh, N. S., and Liu, L., Manipulation of online reviews: An analysis of ratings, readability, and sentiments, Decision Support Systems, 2012, 52(3): 674-684. 13. Holzner, S., Facebook marketing: Leverage social media to grow your business, QUE,

2009, pp. 288.

14. Hughes, D. J., Rowe, M., Batey, M., Lee, A., A tale of two sites: Twitter vs. Facebook and the personality predictors of social media usage, Computers in Human Behavior, 2012, 28(2): 561-569.

15. Jacques, B. G., Santiago, M. G., Beatriz, G. L. V., A social media analysis of the contribu-tion of destinacontribu-tions to client satisfaccontribu-tion with hotels, Internacontribu-tional Journal of Hospitality Management, 2013, 35: 44-47.

16. Kerstetter, D. and Cho, M. H., Prior knowledge, credibility and information search, Annals of Tourism Research, 2004, 31(4): 961-985.

17. Kirkos, E., Spathis, C., Manolopoulos, Y., Audit-Firm group appointment: An artificial in-telligence approach, Intelligent Systems in Accounting, Finance and Management, 2010, pp. 1-17.

18. Koetsier, J., Nielsen’s state of social 2012 report: more social, more mobile, more minutes, more TV, and more ads, Nielsens’ 2012 state of the social union report, 2012.

19. Kim, S. Y., Upneja, A., Predicting restaurant financial distress using decision tree and AdaBoosted decision tree models, Economic Modelling, 2014, 36: 354-362.

20. Laroche, M., Habibi, M. R., Richard, M. O., To be or not to be in social media: How brand loyalty is affected by social media? International Journal of Information Management, 2013, 33(1): 76-82.

21. O’Connor, P., Hopken, W., Gretzel, U., Information and communication technologies in Tourism 2008, Proceedings of the International Conference in Innsbruck, 2008, pp. 591. 22. Pan, B., MacLaurin, T., Crotts, J. C., Travel blogs and the implications for destination

marketing, Journal of Travel Research, 2007, 46(1): 35-45.

23. Park, D. B., Yoon, Y. S., Segmentation by motivation in rural tourism: A Korean case study, Tourism Management, 2009, 30(1): 99-108.

24. Pallis, G., Zeinalipour-Yazti, D., and Dikaiakos, M. D., Online social networks: Status and trends, New Directions in Web Data Management 1, SCI331, 2011, pp. 213-234.

25. Stelzner, M. A., How marketers are using social media to grow their businesses, 2013 So-cial Media Marketing Industry Report, 2013, pp. 2-43.

26. Sigala, M. and Marinidis, D., Exploring the transformation of tourism firms’ operations and business models through the use of web map services, European and Mediterranean Conference on Information Systems 2009 (EMCIS 2009), Brunel University, Izmir, 2009, pp. 1-13.

27. Safko, L., and Brake, D. K.., The social media bible: Tactics, Tools, and Strategies for business success, Wiley, 2009, pp. 640.

28. Treadaway, C. and Smith, M., Facebook Marketing: An Hour a Day, Sybex, 2010. 29. Tomas, E. R., Elena, C. T., An evaluation of Spanish hotel websites: Informational vs.

30. Tyrvainen, L., Uusitalo, M., Silvennoinen, H., Hasu, E., Towards sustainable growth in na-ture-based tourism destinations: Clients’ views of land use options in Finnish Lapland, Landscape and Urban Planning, 2014, 122: 1-15.

31. Terri, L. R., Anny, R., Bed and breakfasts, small inns, and the internet: The impact of technology on the globalization of small businesses, Journal of International Marketing, 2000, 8(2): 86-97.

32. Tan, W. K. and Chang, Y. C., Credibility assessment model of travel information sources: An exploratory study on travel blogs, In Proceedings of ENTER, Springer Vienna, 2011, pp. 457-469.

33. Wang, Y., Yu, Q., Fesenmaier, D. R., Defining the virtual tourist community: implications for tourism marketing, Tourism Management, 2002, 23(4): 407-417.

34. Wilson, R. F., The six simple principles of viral marketing, E-commerce consultant, 2005. 35. Xiang, Z. and Gretzel, U., Role of social media in online travel information search,

Tour-ism Management, 2010, 31(2): 179-188.

36. You, W., Yang, Z., Ji, G., Feature selection for high-dimensional multi-category data using PLS-based local recursive feature elimination, Expert Systems with Applications, 2014, 41(4): 1463-1475.

37. Yu, Y., Duan, W., Cao, Q., The impact of social and conventional media on firm equity value: A sentiment analysis approach, Decision Support Systems, 2013, 55(4): 919-926. 38. Yuksel, A., Akgul, O., Postcards as affective image makers: An idle agent in destination

marketing, Tourism Management, 2007, 28(3): 714-725.

39. Zarella, D., The social media marketing book, O'Reilly Media, 2010, pp. 232.

40. Zhou, L., Wang, T., Social media: A new vehicle for city marketing in China, Cities, 2014, 37: 27-32.