The Effectiveness of Personalized E-mail Newsletters and the Role of Personal Characteristics

Abstract

Personalizing communication means creating persuasive messages that refer to aspects of a person's self. Although the use of personalization is increasing, research on its effectiveness is limited and the results are mixed. This study examined the persuasiveness of personalized e-mail newsletters in terms of increased attention, cognitive activity, evaluation, attitude, intention, and behavior by means of an experiment (n=109). Participants randomly received either a personalized or a generic newsletter advertising a sports center. Personalization triggered a more positive evaluation of the message; however, it did not influence the other effect variables. The effects were moderated by consumers' need for uniqueness, trust, and privacy concerns, suggesting that personalization is a good strategy to increase message evaluation only among individuals who have a high need for uniqueness.

Introduction

The aim of personalization is to make the message more noticeable and persuasive by making it important to an individual.¹ Recognizable aspects of a person—such as name, interests, or purchase history—are included in a message.² Despite the increased focus of marketing activities on e-mail personalization³ and growing commercial e-mail expenditures,⁴ we are lacking knowledge about personalized e-mail communication. Especially with the growing concerns about individuals' privacy, and new software and legislation being introduced to protect Internet users, it is important to understand how individuals react to personalization. Research on the effects of personalized messages is not only limited, but also yields different results. While several authors have proved personalization to be effective,^5–12 others have failed to document positive effects of personalization,^13–16 or even showed reactance to highly personalized solicitations.¹⁷ The mixed results might be due to the disregard of personal characteristics.¹⁸ Therefore, the rationale for this study was twofold: theoretical—to enhance our understanding of personalized communication; practical—to give implications for e-mail marketing. Therefore, we compared personalized communication with nonpersonalized communication, while taking personal factors into account.

Personalized communication

Personalization means creating communication in which information about the recipient is used to refer to some aspects of his or her self.¹⁹ Although personalization can be based on many concepts, the tactic that has gained prominence in current direct marketing is that of identifying the consumer. This strategy is based on acknowledging the recipient by including personalization cues, such as the recipient's name.²⁰ These cues do not change the content, but signal that the message is addressed to the individual.²

Personalization is believed to lead to persuasion effects, such as enhanced attention,⁴ involvement,⁵ memory,⁶ attitude,⁷ intention,⁸ and behavior.⁶ Because people are cognitively sensitive to personal information (e.g., they tend to react when addressed by name²¹), personalization leads to self-referencing. Once self-referencing is activated, the content is processed in the context of self, which makes the message personally relevant.² According to the elaboration likelihood model, people engage in central processing of such a message.²² Therefore, personalized messages lead to more cognitive activity: individuals pay more attention to and better memorize communication.²³ Moreover, such communication exerts more influence on a person's judgments,²⁴ increases attitude certainty, and thus strengthens the attitude–behavior relationship, which makes it more likely that behavior will be influenced.^25,26 Therefore, we hypothesized that

H1: A personalized message is more persuasive than a generic message; it attracts attention, increases cognitive activity, and leads to persuasion effects, namely, positive evaluation, attitude, intention, and behavior.

Personalization and personal characteristics

We focused on three moderating personal characteristics: consumers' need for uniqueness (CNFU), privacy concerns (PC), and trust in a company. According to the need for uniqueness theory, individuals with a high need for uniqueness aim to maintain a sense of being special in relation to others,²⁷ derive satisfaction from the perception that they are unique,²⁸ better memorize information that distinguishes “me” from “you,”²⁹ and are more likely to develop their uniqueness through exclusive consumer products.³⁰ Personalized offers and messages address each recipient as an individual, which allows recipients to differentiate themselves from others and makes each person believe that he or she is unique.³¹ Therefore, individuals with higher CNFU are expected to react more positively toward personalization.^17,32 In other words,

H2: CNFU strengthens the persuasiveness of a personalized message.

A possible downside of personalization is that consumers are becoming more concerned about their privacy³³ and are afraid that have lost control over it,³⁴ which may result in resistance to revealing personal information.³⁵ Therefore, PC may constitute a significant obstacle to successful personalization³⁶ by evoking reactance to communication.³⁷ Thus,

H3: PC weakens the persuasiveness of a personalized message.

A range of studies have shown that consumers are willing to share personal information with organizations they trust.^34,38 For this study, initial trust, which is developed after individuals' first experience with the company,³⁹ is especially important because individuals are not or are barely familiar with the company. Trust may positively influence customers' evaluation of a personalized service,⁴⁰ attitude toward the company,³⁹ and willingness to purchase products.⁴¹ Therefore,

H4: Initial trust in the company behind the message strengthens the persuasiveness of a personalized message.

Methods

Materials

Two versions of an existing e-mail newsletter that advertised the University Sports Centre (USC) were created: a generic and a personalized version. Both newsletters had the same layout, and provided the same information about sports, locations, and prices at the USC. The only difference was that in the personalized version, the recipient's first name was mentioned in three widely spaced places in the text.

Respondents

Data were collected from a sample of Dutch undergraduate students (n=109, of which 73.4% were women) between the ages of 18 and 31 years (M_age=21.20 years, SD=2.85). The data of four (3.7%) students who were already members of the USC were not used. Post-hoc power analysis with G*Power⁴² showed that the sample size was sufficient.

All individuals were randomly assigned to one of two experimental conditions: personalized (n=52) or generic (n=53). There were no differences between the participants in these two conditions in terms of demographic variables (e.g., age and gender). The newsletters were e-mailed to the participants as though they were real newsletters. At the bottom of each newsletter, participants were asked to take part in the survey regardless of their interest in the USC or sports in general. The chance of winning a 25- or 100-euro raffle encouraged them to take part in the study.

Measures

The questionnaire measured dependent variables, personal characteristics, and demographics. A behavioral measure was included at the end of the survey; individuals were given the option to go to the USC Web site to get more information.

As a manipulation check, awareness of personalization was assessed using a three-item scale answered on a five-point Likert scale (α=0.81). Attention was measured by the question, “How thoroughly did you read the newsletter?” There were four possible answers: not at all, only scanned it, read it partially, and read it all. Cognitive activity was measured with the thought listing task.⁴³ The number and the tone of participants' thoughts were coded by two coders. Intercoder reliability was assessed with intraclass correlation coefficients (all above 0.70) and alphas (all above 0.80). Evaluation of the newsletter was measured with a grade as used within the university (1=low, 10=high). Attitude toward the USC consisted of five-point semantic differentials (α=0.83). Intention was measured by two, five-point Likert-type questions: “How probable is it that you will contact the USC?” and “How probable is it that you will join the USC?”

Personal characteristics were measured using multi-item Likert scales anchored by 1 (totally disagree) and 5 (totally agree). We employed 9 items with the highest factor loadings from the 12-item Consumers' Need for Uniqueness (CNFU) Scale⁴⁴ (α=0.80), the three-item Global Information Privacy Concern Scale⁴⁵ (α=0.79), and the five-item Initial Trust Scale³⁹ (α=0.69). A principal components factor analysis with oblimin rotation proved that both convergent and discriminant validity of all scales were sufficient.

Results

The manipulation was successful. Participants in the personalized condition (M=2.96, SD=0.80) were significantly more aware of personalization, t(103)=3.47, p=0.001, than participants in the generic condition (M=2.38, SD=0.90).

Responses to personalization

There was a significant main effect of condition on evaluation of the newsletter, F(1, 103)=5.48, p=0.02, η²=0.05. Participants in the personalized condition evaluated the newsletter significantly more positively than those in the generic condition (Table 1). Analyses also showed a marginally significant main effect for condition on the mean number of positive thoughts, F(1, 103)=3.72, p=0.06, η²=0.04. Participants in the personalized condition expressed more positive thoughts than those in the generic condition.

Table 1.

Main Effects of the Condition on the Dependent Variables

	Generic (n=53)		Personalized (n=52)
Variable	M	SD	M	SD	t (103)	p
Attention	3.02	0.91	2.79	0.94	1.280	0.204
Evaluation of the newsletter	6.49	0.91	6.88	0.81	−2.342	0.021
Attitude (USC)	3.70	0.55	3.61	0.51	0.914	0.363
Intention to contact USC	2.64	1.27	2.58	1.11	0.277	0.782
Intention to join USC	2.43	1.20	2.46	1.04	−0.126	0.900
Number of thoughts	2.97	1.62	3.33	1.90	−1.033	0.304
Number of negative thoughts	0.77	1.00	0.72	0.98	0.271	0.787
Number of positive thoughts	0.65	0.74	1.06	1.34	−1.930	0.056
Information-seeking behavior	1.75	0.43	1.73	0.45	0.278	0.781

USC, University Sports Centre.

The role of personal characteristics

To test for interaction effects, we performed regression analyses with condition (dummy coded), personal characteristics (standardized), and the interaction between condition and personal characteristics as predictors of the dependent variables (Table 2). Simple slope analyses were applied to disentangle significant interactions. An analysis of covariance with estimated marginal means was employed to assess the differences between the two conditions among individuals who scored low (−1 SD) or high (+1 SD) on the moderating variables (Fig. 1).

FIG. 1.

Visualized interaction effects between the personalization condition and personal characteristics: consumers' need for uniqueness (CNFU), privacy concerns (PC), and initial trust (Trust).

Table 2.

Multiple Regression Models for the Dependent Variables

Variable	Attention	Evaluation (message)	Attitude (USC)	Intention (contact)	Intention (join)	No thoughts	No negative thoughts	No positive thoughts	Behavior
Condition	−0.12	0.23^*	−0.09	−0.03	0.01	0.10	−0.03	0.19^†	0.02
CNFU	−0.31^*	−0.41^**	0.19	0.21	0.25^†	−0.23	−0.00	−0.04	−0.00
Condition^*CNFU	0.12	0.30^*	−0.26^†	−0.02	−0.04	0.25^†	−0.00	−0.02	0.22
Condition	−0.11	0.21^*	−0.13	−0.04	−0.00	0.10	−0.04	0.19^†	0.06
Privacy	−0.02	−0.23	−0.30^*	−0.28^*	−0.26^†	0.03	−0.01	−0.09	−0.06
Condition^*Privacy	0.09	0.21	0.20	0.32^*	0.27^†	−0.03	−0.05	0.13	0.27^*
Condition	−0.12	0.24^*	−0.07	−0.02	0.02	0.10	−0.03	0.19^†	0.03
Trust	0.00	0.28^*	0.54^***	0.13	0.22	0.12	0.06	0.00	0.18
Condition^*Trust	0.12	−0.09	−0.41^**	−0.11	−0.18	−0.13	−0.19	0.09	−0.16

Note: Coefficients are standardized regressions coefficients (betas).

†

p<0.10, ^*p<0.05, ^**p<0.01, ^***p<0.00.

CNFU, consumers' need for uniqueness.

CNFU significantly moderated the effect of condition on evaluation of the newsletter (β=0.30, t=2.14, p=0.04). The simple slope was significant for the generic condition (b=− 0.41, p=0.004), but not for the personalized condition (b=−0.01, p=0.92), suggesting that the level of CNFU was associated with evaluation only for participants who received the generic condition. Participants with a high CNFU were significantly more negative toward the generic than toward the personalized condition, F(1. 101)=10.59, p=0.002.

PC significantly influenced the effect of message condition on intention to contact the USC (β=0.32, t=2.37, p=0.02) and information-seeking behavior (β=0.27, t=1.99, p=0.05). The effect of PC on intention to contact the USC was significant for the generic condition (b=−0.28, p=0.04), but not for the personalized condition (b=0.19, p=0.19), suggesting that the level of PC was associated with intention only for participants who received the generic condition. Individuals with low PC were significantly more willing to contact the USC after seeing the generic condition than after seeing the personalized condition, F(1, 101)=3.83, p=0.05.

With respect to information-seeking behavior, the effect of PC was significant for the personalized condition (b=0.33, p=0.02), but not for the generic condition (b=−0.06, p=0.65), suggesting that the level of PC was associated with behavior only for participants who received the personalized condition. However, no difference was found between the two conditions among participants with a low or a high PC.

Trust in the USC significantly moderated the effect of condition on participants' attitude toward the USC (β=−0.41, t=−3.31, p=0.001). The positive effect of trust on the attitude toward the USC was significant for the generic condition (b=0.54, p<0.001), but not for the personalized condition (b=−0.06, p=0.64), suggesting that the level of trust was associated with attitude only for participants who received the generic condition. Participants with a higher trust level expressed a more negative attitude after reading the personalized message than after reading the generic one, F(1, 101)=8.28, p=0.01.

Discussion

The present study systematically compared personalized and generic messages, while taking personal characteristics into account. The results partially support H1, namely, that a personalized message is more persuasive than a generic message. This held for the evaluation of the newsletter and the number of positive thoughts. However, we did not find any significant differences between the personalized and generic messages with respect to attention, attitude, intention, and information-seeking behavior. With respect to the personal characteristics, we could partially support H2, viz., that need for uniqueness strengthens the persuasiveness of personalization. However, we had to reject H3 and H4, which stated that PC weakens and trust strengthens the persuasiveness of personalization. In contrast to what we expected, participants with a low level of PC expressed a significantly stronger intention to contact the USC in the generic condition than in the personalized condition. We were also surprised to learn that individuals who trusted the USC expressed a more positive attitude toward the company after reading the generic newsletter.

A plausible explanation for the relatively weak effect of personalization is that the personalization cues signaled the persuasion attempt, rather than the relevance of the message.⁴⁶ This may inhibit persuasion or even cause reactance, especially if personalization is not justified.¹⁷ Moreover, when personalization is not perceived as genuine (the source announces personalization, but does not meet the individual's personal preferences), it may increase perceived liabilities, which in turn can make the attitude less positive, especially when people trust the organization.^40,47

Although our results show that personalization is not very effective, it is too soon to draw definitive conclusions. Further work is required to gain a complete picture of the effectiveness of personalization. It seems that personalization has a positive influence on attitude toward the message, but there is no spillover effect on attitude toward the company or behavior; thus, research should focus on the mechanism of personalization in terms of stages of information processing.

Footnotes

Disclosure Statement

No competing financial interests exist.

References

Sackmann

, Strüker

, Accorsi

. Personalization in privacy-aware highly dynamic systems. Communications of the ACM, 2006; 49:22–38.

Dijsktra

. Working mechanisms of computer-tailored health education: evidence from smoking cessation. Health Education Research, 2008; 20:527–539.

Jupiter Research. 2007. www.emailstatcenter.com. 2010 Oct 21.

Wunderman

. Direct marketing: leading today's digital world. Brandweek, 2010; 51:1–2

Bull

, Holt

, Kreuter

et al.

Understanding the effects of printed health education materials: which features lead to which outcomes?

Journal of Health Communication, 2001; 6:265–279.

Kreuter

, Bull

, Clark

et al. Understanding how people process health information: a comparison of tailored and untailored weight-loss materials. Health Psychology, 1999; 18:487–494.

Wogalter

, Racicot

, Kalsher

et al. Personalization of warning signs: the role of perceived relevance on behavioral compliance. International Journal of Industrial Ergonomics, 1994; 14:233–242.

Webb

, Simmons

, Brandon

. Tailored interventions for motivating smoking cessation: using placebo tailoring to examine the influence of expectancies and personalization. Health Psychology, 2005; 24:179–188.

Lin

C-P

, Huang

H-N

, Joe

S-W

et al. Learning the determinants of satisfaction and usage intention of instant messaging. CyberPsychology & Behavior, 2008; 11:262–267.

10.

Dijkstra

. Working mechanisms of computer-tailored health education: evidence from smoking cessation. Health Education Research, 2005; 20:527–539.

11.

Kreuter

, Wray

. Tailored and targeted health communication: strategies for enhancing information relevance. American Journal of Health Behavior, 2003; 27:227–232.

12.

Heerwegh

, Vanhove

, Matthijs

et al. The effect of personalization on response rates and data quality in web surveys. International Journal of Social Research Methodology, 2005; 8:85–99.

13.

Shen

, Ball

. Is personalization of services always a good thing? Exploring the role of technology-mediated personalization (TMP) in service relationships. Journal of Services Marketing, 2009; 23:80–92.

14.

, Cude

. “Hello, Mrs Sarah Jones! We recommend this product!” Consumers' perceptions about personalized advertising: comparisons across advertisements delivered via three different types of media. International Journal of Consumer Studies, 2009; 33:503–514.

15.

Porter

, Whitcomb

. The impact of contact type on web survey response rates. Public Opinion Quarterly, 2003; 67:579–588.

16.

Bull

, Kreuter

, Scharff

. Effects of tailored, personalized, and general health messages on physical activity. Patient Education and Counseling, 1999; 36:181–192.

17.

Bernett White

, Zahay

, Thorbjørnsen

et al. Getting too personal: reactance to highly personalized email solicitations. Market Letters, 2008; 19:39–50.

18.

, Davern

, Tam

. Personalization and choice behavior: the role of personality traits. The DATA BASE for Advances in Information Systems, 2008; 39:32–47.

19.

Kalyanaraman

, Olivier

, Magee

. 2010. The interplay between customization and cognition in web portals: the multiple role hypothesis explanation. Paper presented at the annual meeting of the International Communication Association, Suntec City, Singapore.

20.

Hawkins

, Kreuter

, Resnicow

et al. Understanding tailoring in communicating about health. Health Education Research, 2008; 23:454–466.

21.

Cherry

. Some experiments on the recognition of speech with one and with two ears. The Journal of the Acoustical Society of America, 1953; 25:975–979.

22.

Petty

, Cacioppo

. Issue involvement can increase or decrease persuasion by enhancing message-relevant cognitive responses. Journal of Personality and Social Psychology, 1979; 37:1915–1926.

23.

Rogers

, Kuiper

, Kirker

. Self-reference and the encoding of personal information. Journal of Personality and Social Psychology, 1997; 35:677–688.

24.

Hunt

, McDaniel

. The enigma of organization and distinctiveness. Journal of Memory and Language, 1993; 32:421–445.

25.

Cacioppo

, Kao

, Petty

et al. Central and peripheral routes to persuasion: an individual difference perspective. Journal of Personality and Social Psychology, 1986; 5l:1032–1043.

26.

Petty

, Briñol

. Psychological processes underlying persuasion: a social psychological approach. Diogenes, 2008; 55:52–67.

27.

Snyder

, Fromkin

. 1980. Uniqueness: the human pursuit of difference. New York: Plenum Press.

28.

Simson

, Nowlis

. The role of explanations and need for uniqueness in consumer decision making: unconventional choices based on reasons. Journal of Consumer Research, 2000; 27:49–68.

29.

Codol

, Jarymowicz

, Kaminska-Feldman

et al. Asymmetry in the estimation of interpersonal distance and identity affirmation. European Journal of Social Psychology, 1989; 19:11–22.

30.

Tian

, Bearden

, Hunter

. Consumers' need for uniqueness: scale development and validation. Journal of Consumer Research, 2001; 28:50–66.

31.

Kalyanaraman

, Sundar

. The psychological appeal of personalized content in web portals: does customization affect attitudes and behavior? Journal of Communication, 2006; 56:110–132.

32.

Snyder

. Product scarcity by need for uniqueness interaction: a consumer catch-22 carousel? Basic and Applied Social Psychology, 1992; 13:9–24.

33.

Langheinrich

, Nakamura

, Abe

et al. Unintrusive customization techniques for Web advertising. Computer Networks, 1999; 31:1259–1272.

34.

Nowak

, Phelps

. Direct marketing and the use of individual-level consumer information: determining how and when “privacy” matters. Journal of Direct Marketing, 1997; 11:94–108.

35.

Rubini

. Overcoming the paradox of personalization: building adoption, loyalty, and trust in digital markets. Design Management Journal, 2001; 12:49–54.

36.

Pitta

, Franzak

, Laric

. Privacy and one-to-one marketing: resolving the conflict. Journal of Consumer Marketing, 2003; 20:616–628.

37.

Winer

. A framework for customer relationship management. California Management Review, 2001; 43:89–105.

38.

Dommeyer

, Gross

. What consumers know and what they do: an investigation of consumer knowledge, awareness, and use of privacy protection strategies. Journal of Interactive Marketing, 2003; 17:34–51.

39.

Koufaris

, Hampton-Sosa

. The development of initial trust in an online company by new customers. Information & Management, 2004; 41:377–397.

40.

De Pechpeyrou

, Desmet

. 2007. The role of trust in consumers' evaluations of Website personalization. ANZMAC Conference. University of Otago: New Zealand.

41.

Kaplan

, Nieschwietz

. A Web assurance services model of trust for B2C e-commerce. International Journal of Accounting Information Systems, 2003; 4:95–114.

42.

Erdfelder

, Faul

, Buchner

. GPOWER: a general power analysis program. Behavior Research Methods, Instruments, & Computers, 1996; 28:1–11.

43.

Cacioppo

, Petty

. 1981. Social psychological procedures for cognitive response assessment: the thought listing technique. Merluzzi

, Glass

, Genest

. Cognitive assessment. New York: Guilford, 309–342.

44.

Ruvio

, Shoham

, Makovec Brenčič

. Consumers' need for uniqueness: short-form scale development and cross-cultural validation. International Marketing Review, 2008; 25:33–53.

45.

Malhotra

, Kim

, Agarwal

. Internet users' information privacy concerns (IUIPC): the construct, the scale, and a causal model information systems research, 2004; 15:336–355.

46.

Petty

, Cacioppo

. 1986. The elaboration likelihood model of persuasion. Berkowitz

. Advances in experimental social psychology. New York: Academic Press, 123–205.

47.

Riemer

, Totz

. 2003. The many faces of personalization; An integrative economic overview of mass customization and personalization. Tseng

, Piller

. The customer centric enterprise advances in mass customization and personalization. New York, Berlin: Springer, 35–51.