1 Introduction

1.1 Nature of engagement

‘Customer engagement’ involves the existence of deep, meaningful and enduring connections between an organisation and its customers (Kumar et al. 2010). It is not the same as customer ‘commitment’ (Bowden 2009) as engagement represents a deeper and more meaningful connection between an organisation and its customers than commitment, and one that may endure over a longer period of time (Kumar et al. 2010). Bowden (2009) noted how engagement is a more powerful force linking the customer to the organisation than commitment and that it includes ‘feelings of confidence, integrity, pride, and passion’ for an organisation (p.64). Engagement embraces activities that seek to tie people emotionally to the organisation (Bell and Esingerich 2007). It is associated with a person’s strong psychological investment in a relationship, especially through participation in the relationship and through co-creation (Kumar et al. 2010). An empirical study conducted by Warrington and Shim (2000) found the constructs of involvement and commitment to be conceptually distinct. This was because commitment was frequently motivated by inertia as opposed to proactive affiliation (see Hess and Story 2005), and might not lead to loyalty as it may rely on shallow evaluations.

Engagement differs from trust, moreover, in that, according to Hess and Story (2005), trust is a necessary (but not sufficient) condition for the presence of commitment and, as such, constitutes an input to engagement. Van Doorn et al. (2010) argued that trust should be regarded as a driver of, but not conceptually the same as, engagement on the grounds that the latter typically involves passion, which is not the case with trust (Bowden 2009). Equally, engagement is conceptually distinct from involvement because engagement includes elements that extend well beyond the involvement concept (as expressed by, for instance, Zaichkowsky (1985). Involvement relates to the perceived relevance of an entity based on a person’s needs, values and interests, and/or to affective emotional ties and identification with an organisation (Laurent and Kapferer 1985). Engagement however extends to total dedication to an organisation, untiring support, pride in association, and energetic and proactive concern for the organisation’s well-being. Thus, involvement is subsumed within the broader construct of engagement (Hess and Story 2005; Bowden 2009). In summary, engagement may be regarded as a holistic characterisation of a customer’s behaviour that encompasses several sub-dimensions (see Warrington and Shim (2000), Bowden (2009) and Van Doorn et al. (2010) for further information concerning this matter).

1.2 Engagement among charity donors

Although there has been a great deal of interest in engagement among researchers concerned with the customers of commercial businesses, due largely to recognition of the shortcomings of conventional relationship management programmes (see Strauss and Neuhaus 1997; Meyer and Schwager 2007) and the observation that many relationship management projects fail (up to 33 % according to Krigsman (2009), little attention has been devoted to the application of the engagement construct to the nonprofit domain. The research reported below sought to help fill this gap in knowledge via an empirical study of the engagement feelings and behaviour of donors to fundraising charities. Following Van Doorn et al’s (2010) interpretation of client engagement with commercial organisations, donor engagement may be defined as a charity supporter’s feelings and behaviour that go well beyond the act of giving. Van Doorn et al’s characterisation of engagement is appropriate for the present study as it recognises that nurturing a donor base requires activities that (i) seek far more than repeat donations, (ii) demand a stronger state of connectedness with donors than mere ‘commitment’, and (iii) focus on multiple forms of non-transactional donor behaviours that result from several motivational drivers. Behaviour is likely to extend to word of mouth referrals, volunteering and otherwise participating in a charity’s activities, joining and interacting with communities (especially on-line communities) that support a cause, blogging, web posting, advocacy, and purchasing a charity’s merchandise.

1.2.1 Benefits of donor engagement

The engagement issue is important for charities for several reasons. An engaged donor will feel (i) intimately connected with the charity, and (ii) empowered to influence its activities (cf. Chaffey 2009). Feelings of engagement are experienced in the long term and, in the words of (Schaufeli et al. 2002 p.74), constitute ‘a persistent and pervasive affective-cognitive state that is not focused on any particular event or behaviour’. A major benefit accruing to charities with engaged supporters is that the relationship between the donor and the charity will be very close, meaning that the person will not only be willing to continue giving to a charity but will also go out of his or her way to be associated with the organisation (cf. Feather and Chun 2008). Also an engaged donor is more likely to forgive a charity for service inadequacies (cf. Woodcock 2007). Another advantage is that the presence of a strong emotional bond between a donor and the charity induces loyalty (Voyles 2007) and thus makes it harder for competing charities to induce the person to switch his or her support to a rival organisation (cf. Feather and Chun 2008). Engaged donors will be more ready than others to co-operate with a charity vis-à-vis, for example, participation in satisfaction surveys, in evaluations of service quality (cf. Woodcock 2007), and in on-line forums where people share their views and experiences. (Online interactions are increasingly important for diffusing information about a charity and/or the good cause it supports [cf. Bennett 2009].) Engaged donors might have higher donor lifetime values than others (research has established that related concepts are associated with giving behaviour [see Sargeant and Woodliffe 2007]) and may add further value indirectly via their advocacy of the charity to friends, family and associates (cf. Woodcock 2007). Additionally they have a ‘knowledge value’ that results from the feedback they provide to the organisation and a ‘co-creation value’ arising from their active involvement in a charity’s events (cf. Kumar et al. 2010).

1.2.2 Elements of donor engagement

Patterson et al. (2006) suggested that engagement could be applied to the marketing domain under four headings: vigour, dedication, absorption and interaction. Schaufeli et al. (2002) characterised engagement vigour in terms of mental enthusiasm for an organisation, a strong desire to support its activities, and willingness to invest effort in maintaining a relationship with the organisation, even in the face of difficulties. Thus, engaged donors should exhibit an ongoing sense of energetic connection with a charity; manifest in the application of a high level of energy to interactions with the organisation (Patterson et al. 2006). Dedication, according to Schaufeli et al. (2002), involves pride in being associated with an organisation plus a sense of belonging to and strong identification with the organisation. Thus, a dedicated donor to a charity will be inspired by the charity’s work, will be passionate about the cause it supports and will experience deep satisfaction when making donations or contributing to the organisation’s activities. The term absorption refers to a person’s profound interest in an organisation. An absorbed charity donor will be fascinated by the charity’s cause and engrossed in its work. Interaction concerns an individual’s inclination to participate in joint activities with an organisation. In the charity context this could involve volunteering, providing feedback, participating in focus groups and communicating with various stakeholders.

2 Literature review

In the words of Kumar et al. (2010), ‘the concept of customer engagement is novel and in its developmental phase, so differing opinions regarding its conceptualisation are bound to occur’ (p. 298). The current section seeks to construct a model of the antecedents and consequences of donor engagement, based on a priori considerations and on the limited extant literature in the field. Three genres of variables have been commonly posited to affect a person’s feelings of engagement with an organisation: (i) the individual’s experience of the organisation, (ii) the quality of a person’s relationship with the organisation (itself deriving in part from experience), and (iii) the congruence between the image of the organisation and an individual’s self-construed identity. These are discussed below in relation to charity donation behaviour.

2.1 A donor’s experience of a charity

Experiences of a charity occur at ‘touch points’ between the individual and the organisation (Meyer and Schwager 2007; Verhoef et al. 2009). The extent and nature of these touch points will be determined in part by a charity’s relationship marketing activities (Verhoef et al. 2009; Palmer 2010). A person’s interpretations of his or her experiences of touch points are likely to be affected by the individual’s prior perceptions of the organisation’s reputation (Bennett and Gabriel 2003; Van Doorn et al. 2010), and by perceptions of the quality of its services to donors (O’Malley and Prothero 2004; Bennett and Barkensjo 2005; Feather and Chun 2008).

2.1.1 Relationship marketing activities

British charities’ relationship marketing activities extend to telemarketing, database marketing, the sale of charity branded products and memorabilia, charity credit cards, and ‘bonding’ events such as open days or gala dinners (for details see Bennett and Barkensjo 2005). Relationship-building public relations events create norms and expectations concerning a charity’s operations and, critically, can develop a donor’s sense of shared values and affiliation with the organisation. Further relationship building activities completed by fundraising charities include invitations to donors to become involved in supporters’ clubs, volunteer activities and the sale of raffle tickets. Verhoef et al. (2009) observed how the delivery of pleasurable experiences to an organisation’s clientele is a crucial part of its relationship marketing activities. Pleasurable experiences, Verhoef et al. (2009) continued, encourage engagement. Excellent relationship marketing should in principle improve the overall quality of a donor’s relationship with a charity (Conway 1997; Bennett and Barkensjo 2005), which in turn will induce donors to want to engage with the organisation. The issue is worthy of testing however because not everyone welcomes an organisation’s relationship marketing activities (Ashley et al. 2011). O’Malley and Prothero (2004) documented examples of how relationship marketing used badly can actually reduce a customer’s liking for an organisation. This was due to disparities between relationship rhetoric and the realities delivered, to the employment of ‘intrusive’ marketing techniques, and to the difficulties associated with simultaneously attempting to create relationships with perhaps hundreds of thousands of individuals. Independently of this connection, relationship marketing may exert direct influences on donor behaviour (Bennett and Gabriel 1998; Brennan and Brady 1999; Bennett and Barkensjo 2005). Such considerations suggest the following hypothesis.

H1. Donors who regard a charity’s relationship marketing activities as excellent are more likely (a) to engage with the charity, (b) to perceive that their overall relationship with the charity is of high quality, and (c) to exhibit positive giving behaviour towards the charity.

2.1.2 Service quality

Examples of good service provision to donors include the efficient transmission of communications, the presence of effective facilities for responding to appeals, the availability of sound mechanisms for making donations, provision of convenient and user friendly systems for purchasing a charity’s products, and ensuring that people who participate in a charity’s events are well looked after. Sargeant and Shang (2012) found that the provision of benefits to donors made supporters feel a desire to associate with a charity. Although not everyone is sensitive to service quality, research suggests that the quality of the design and delivery of a charity’s services to its supporters constitutes a crucial aspect of donor experience (cf. O’Malley and Prothero 2004; Palmer 2010). In principle moreover the provision of sound service quality should stimulate a person’s desire to engage with the organisation (Patterson et al. 2006; Feather and Chun 2008; Palmer 2010). Empirical work by Sargeant (2001) established that the perceived quality of the service provided to donors by a charity impacted significantly on donors’ attitudes and behaviour towards the charity (see also Bennett and Barkensjo 2005). The technical calibre of a charity’s communications with its donors has been found to represent a major determinant of the latter’s perceptions of the level of the organisation’s service quality (Sargeant and Jay 2004; Sargeant et al. 2004). The effect of service quality on engagement could be direct or mediated by an improvement in relationship quality. Excellent service provision should enhance a person’s trust in an organisation’s work (see Shemwell et al. 1998 for details of the academic literature supporting this proposition), and trust is itself a major component of relationship quality (Rosen and Surprenant 1998; Roberts et al. 2003; Bennett and Barkensjo 2005). Additionally and regardless of any link between excellent service quality and relationship quality, a charity’s provision of excellent service to its donors may induce donors to behave positively towards the organisation. Hence:

H2. Donors who perceive a charity to be supplying excellent services to its supporters are more likely (a) to engage with the charity, (b) to perceive their overall relationship with the charity to be of high quality and (c) to exhibit positive giving behaviour towards the charity.

2.1.3 Reputation of the charity

It has been argued that a person’s view of the reputation of an organisation is a part of the individual’s experience of the organisation (Meyer and Schwager 2007; Feather and Chun 2008; Van Doorn et al. 2010). A substantial amount of literature (for details see Van Doorn et al. (2010 p. 257) alleges that organisations with good reputations are likely to engender high levels of engagement. Reputation, according to Meyer and Schwager (2007), has the capacity to shape a person’s experiences of an organisation by ‘embedding the fundamental value proposition in all its offerings’ (p. 3). Roberts et al. (2003) identified reputation as an important ‘psychological bond’ between an individual and an organisation that exerted a significant influence on perceptions of relationship quality (p. 174). Reputation helps an individual to place the sensations resulting from interactions with an organisation within a context (Gupta and Vajic 2000) and to categorise emotional and subjective experiences (Schmitt 1999; Verhoef et al. 2009). It serves as a value signal in situations of information overload and complexity and, in the charity context, has been found to assist an individual to gauge a charity’s merits (see Bennett and Gabriel 2003). Accordingly, reputation may be expected to affect a donor’s propensity to engage. Moreover, research has established the presence of direct and substantial links between a charity’s reputation and its ability to raise funds (Sargeant et al. 2008; Bennett and Gabriel 2003; Meijer 2009). Thus, a direct connection between reputation and donor behaviour may be anticipated. Accordingly:

H3. Donors who perceive a charity to have an excellent reputation are more likely (a) to engage with the charity, (b) to perceive that their overall relationship with the charity is of high quality, and (c) to exhibit positive giving behaviour towards the charity.

2.2 Relationship quality

Relationship quality has been defined as the ‘degree of appropriateness of a relationship to fulfil the needs of the customer’ (donor in the present context) (Henning-Thurau and Klee 1997, p. 751). Roberts et al. (2003) found that relationship quality was a distinct construct that differed significantly from service quality, and that relationship quality was a better predictor of behavioural intentions than service quality. Rosen and Surprenant (1998) similarly concluded that the attributes that people valued in a relationship did not necessarily correspond to the attributes of the service. Overall, prior research has focused on two clusters of variables allegedly associated with relationship quality, i.e., those relating to (i) trust in an organisation, and (ii) feelings of closeness to the organisation (sometimes proxied by other variables such as ‘sense of belonging’, bonding, intimacy, or emotional connection). (For details of past studies relating to these propositions see, for example, Crosby et al. (1990); Barnes (1997); Roberts et al. (2003, Table III); Bennett and Barkensjo (2005); Wong and Sohal (2006); Leahy (2011)). Trust and closeness are appropriate variables to employ in the present context as they are known to be heavily associated with a person’s satisfaction with an organisation’s work (Crosby et al. 1990; Sargeant et al. 2004; Wong and Sohal 2006), sense of involvement (Hess and Story 2005; Bowden 2009), commitment and loyalty (Oliver 1999; Sargeant et al. 2004). All these have the potential to contribute to engagement.

2.2.1 Trust

Trust in a charity involves perceptions of its integrity, honesty and sincerity, and the belief that it is genuinely concerned for the welfare of its beneficiaries (cf. Roberts et al. 2003). It is heavily connected with donors’ levels of satisfaction with a charity’s philanthropic work (cf. Crosby et al. 1990; Aldrich 2000; Sargeant and Woodliffe 2007) and with perceptions that a charity is efficient and professional (Bennett and Barkensjo 2005). Duncan and Moriarty (1998) argued that, in large part, trust in an organisation is the consequence of its relationship marketing activities in conjunction with perceptions of its reputation. Donors perceive trusted charities as organisations that may be relied upon to keep their promises vis-à-vis doing good, making a difference to beneficiaries’ lives, deploying resources wisely, etc. (Bennett and Barkensjo 2005). A wealth of literature supports the proposition that trust is a major dimension of relationship quality (for details see, for example, Roberts et al. (2003); Wong and Sohal (2006). It follows from the above that donors who trust a charity will be likely to engage with it. Independently of that connection it is predicted a priori that trust will affect donor behaviour.

H4. Donors with high levels of trust in a charity are more likely (a) to engage with the charity, and (b) to exhibit positive giving behaviour towards the charity.

2.2.2 Closeness

Patterson et al. (2006) suggested that feelings of closeness with an organisation were a direct determinant of engagement. Verhoef et al. (2010) similarly observed how engaged customers (donors in the present context) felt that they were an integral part of the organisation. Chaffey (2009, p. 4) referred to ‘intimacy in interactions’ as a cause of engagement. Sprott et al. (2009) described the link between an engaged customer and an organisation as involving close bonds and strong personal connections. Perceptions of being close to an organisation have been found to represent an important component of the quality of a person’s relationship with it (Barnes 1997; Wong and Sohal 2006), especially in situations (as in the charity donation context) where (i) ample opportunities for interaction exist (Leahy 2011), and (ii) individuals who experience a sense of intimacy are free to opt out of relational interactions whenever they wish to do so (Beetles and Harris 2010). It is relevant to note however that trust does not necessarily imply closeness. Bennett and Gabriel (2001) found that the desire for closeness with an organisation resulted in part from the organisation’s reputation. Donors who feel close to a charity will probably have the same values, ethical and moral positions and general world view as the charity (cf. Swift 1998). This, together with implied ‘contextual rules’ embedded within a close donor-charity relationship (cf. Simonin 1999 p. 466), might engender engagement. Closeness, moreover, could induce positive donor behaviour even among supporters who do not feel a sense of engagement.

H5. Donors who feel very close to a charity are more likely (a) to engage with the charity, and (b) to exhibit positive giving behaviour towards the charity.

2.3 Self-congruence

Sargeant and Woodliffe (2007, p. 284) cited a large number of studies which found that people often prefer to donate to charities that possess symbolic meanings and images congruent with the ways in which they like to see themselves. Supporting such charities can maintain and enhance a person’s self-identity, as it helps the individual demonstrate both inwardly and to the outside world that he or she is connected with certain modes of behaviour, attributes or values (cf. Graeff 1996). In particular, research has demonstrated that many people want mainly to help those they perceive as being similar to themselves. Thus for example the rich have been found to be prone to patronise the arts, education and healthcare, while avoiding ‘poor people’s’ causes such as homelessness (Ostrower 1996 p. 130). Gentile et al. (2007) observed how a person’s decision to affiliate with a particular organisation would often be the result of his or her desire inwardly to affirm certain values and beliefs. Such considerations could stimulate a person’s desire to engage with an organisation (see Van Doorn et al. (2010) for details of literature relating to this matter).

H6. Donors who perceive giving to a certain charity to be highly congruent with their self-identity are more likely (a) to engage with the charity, and (b) to exhibit positive giving behaviour towards the charity.

3 Dependent variables

Three dependent variables were included in the study: higher levels of giving (cf. Woodcock 2007; Feather and Chun 2008; Kumar et al. 2010), improved word of mouth (cf. Bijmolt et al. 2010; Kumar et al. 2010; Van Doorn et al. 2010), and future intention to continue support (cf. Woodcock 2007; Bijmolt et al. 2010). A priori the passion, dedication, interest in the charity, etc., associated with high engagement may be anticipated to encourage a donor to give greater amounts and to donate more frequently. Sargeant et al. (2008) found direct links between emotional engagement and charity giving, due to emotionally engaged supporters’ contacts with a charity being more memorable and exciting. Prior literature in the field is almost unanimous in suggesting that high engagement with an organisation is associated with positive word of mouth (see in particular Patterson et al. 2006; Bijmolt et al. 2010; Van Doorn et al. 2010). Word of mouth can be a critical factor for organisational success. Hibbert (1995) noted the importance of positive word of mouth for reassuring people about a charity’s reliability and for recruiting donors. Equally, Hibbert (1995) continued, negative word of mouth could be extremely damaging, especially for a high profile charity. Also it seems reasonable to suppose that the more engaged a donor with a currently supported charity then the higher the likelihood that support will continue in the future. Moreover, a highly engaged donor may well intend to increase his or her level of future support if and when the person’s circumstances make this possible.

Hypotheses H4 (b), H5 (b) and H6 (b) propose that the above dependent variables are positively affected by relationship quality and self-congruence. In order to complete the model it is finally posited that:

H7. Donor behaviour is influenced positively by donor engagement.

Figure 1 presents a diagrammatic representation of the suggested model. Donor experience is assumed to enhance relationship quality (Meyer and Schwager 2007) and, independently of this, to affect donor engagement. Additionally it is further hypothesised that even if a donor is not engaged and does not perceive a high degree of relationship quality, the individual’s behaviour might still be influenced by experience. Relationship quality is posited to impact both on engagement (Van Doorn et al. 2010) and on behaviour (Rosen and Surprenant 1998; Bennett and Barkensjo 2005). Self-congruence is hypothesised to affect donor engagement (Van Doorn et al. 2010) and behaviour (Sargeant and Woodliffe 2007; Bennett and Ali-Choudhury 2009). The Figure does not posit any relationships between, on the one hand, the donor experience variables (relationship marketing activities, etc.) and self-congruence on the other. Such connections might occur if a charity’s good reputation, high service quality and excellent relationship marketing induce a person to see him or her self as someone who possesses the same attributes and values as the charity. However, whilst reputation, etc., may cause an individual to hold an organisation in high regard, these variables are not necessarily associated with shared inner values and basic self-identity. Also, research has established that strong self-congruence typically leads to a desire to associate with a nonprofit regardless of service quality, etc. (Bhattacharya et al. 1995), Nevertheless, the test of the model (see below) did include a test of this possibility. The results showed statistically insignificant relationships (p < 0.2) in all cases.

Fig. 1
figure 1

A suggested model

Full size image

4 The study

A questionnaire covering the above mentioned constructs (plus a person’s age and financial circumstances) was constructed and pre-tested through (i) discussions with senior fundraising personnel in two large UK charities, and (ii) administration to 30 members of the public approached around Metro locations in central London. Apart from factual questions and items concerning engagement, all variables were measured using seven-point agree/disagree items adapted from scales employed in previous empirical studies. A person’s level of engagement was assessed using a fresh scale specially constructed for the present investigation. This new measure was based on literature in the consumer and more general engagement fields plus discussions with senior fundraisers in two large UK charities. A list of items was assembled, modified and refined via a procedure that involved 200 people of the type included in the present study but who were taking part in a separate investigation. This resulted in a 16-item scale, as shown in the Appendix section 5. Responses to the 16 items made by the participants in the current study were factor analysed and the resulting factors rotated using both orthogonal and oblique procedures. A correlated three factor solution emerged, (R > 0.73), the superiority of which when compared to orthogonal alternatives was confirmed via the AMOS 19 package. Factor one accounted for 29 % of total variation in the data (Cronbach’s alpha = 0.90). For convenience this factor is labelled ‘enthusiasm’, as all its indicators relate in some way to the zealousness of a person’s support for the supported charity. The second factor accounted for 28 % of variation (alpha = 0.90) and may be said to reflect ‘passion’ for the organisation. Factor three explained 27 % of variation (alpha = 0.92) and concerns a donor’s ‘fascination’ with the charity’s work. As the factors were substantially correlated, all 16 items were employed in a single measure of donor engagement, without sub-dividing the inventory into three sub-units. The lack of discrimination among the sub-divisions was confirmed via three tests. Firstly, average variance extracted (AVE) figures for each of the sub-constructs were compared against squared correlations among the three pairs of sub-constructs (Fornell and Larcker 1981). In all cases the AVE figures were at least 30 % lower than the relevant R-squares. Secondly, three regressions were run between pairs of sub-constructs constraining the parameter estimates to unity and comparing the outcomes to models where the parameters were freely estimated (Anderson and Gerbing 1988). All the Chi-square values for the differences between the outputs to the constrained and the unconstrained models (one degree of freedom) fell below 2.9, indicating the absence of discrimination between the sub-units. Finally the 95 % confidence intervals of the correlations between pairs of sub-units were examined to establish whether they contained unity (Baggozi et al. 1991). This occurred in all three cases. Thus, donor engagement is a conflated construct with the consequence that there is no need to have separate causal paths linking its various (interconnected) elements to other parts of the model.

The final questionnaire is summarised in the Appendix to the paper, which also lists the primary literature sources employed in its construction. Passers-by were approached (at random) in street locations and around Metro station in Greater London and were asked whether they had given more than £30 to a specific charity during the previous year and whether they had donated to this charity, by any means, on more than two occasions over the past year. (On average UK citizens give around £31 per donation to charity in general [median £11] and donate approximately five times a year [CAF/NCVO 2012]). Individuals who confirmed that this was the case were then asked to complete the questionnaire. About one in nine of the people approached agreed to participate in the investigation. Interviews took place in several districts (some prosperous, some socially deprived) in Greater London and were conducted by the author, two university employees, and post-graduate students trained for the task and paid national rates for time spent on the project. A target of 1000 responses was set, resulting in 791 completed questionnaires. Interviews took place at different times of the day and on disparate days of the week over a four month period.

4.1 Formation of variables

As previously stated, engagement was measured as a single entity without sub-divisions for enthusiasm, passion and interest. A factor analysis of the ten relationship quality items generated a two factor solution, with the five trust items loading onto the first factor (lambda = 4.4, alpha = 0.91) and the five closeness items loading onto the second (lambda = 4.0, alpha = 0.92). The two factors were not substantially correlated (R = 0.29). Discriminant validity between the two constructs was confirmed using the three tests previously mentioned in relation to the 16 item donor engagement scale. The AVEs for the two constructs were 0.70 and 0.68 respectively; the Chi-square difference value (one degree of freedom) was 4.99 and the upper limit on the 95 % confidence interval on the correlation coefficient between the two constructs was 0.65. Hence it is concluded that closeness on the one hand, and ‘trust’ on the other constitute different constructs. Whilst there was some multicollinearity among the three components of donor experience, the magnitudes of the correlations (R < 0.39) were below the threshold at which they would cause technical problems in a regression analysis (see Aiken and West 1991).

5 Results

The average amount given by the respondents in their most recent donation was £37. This was slightly above the UK national average of £31 (CAF/NCVO 2012); as expected in the country’s wealthiest city. On average the participants had given to their specified charity an average of 2.9 times in the previous year and reported having donated £105 to their chosen charity during that period. A wide range of charities was represented in the sample, with no one sector predominating. Fifty-nine per cent of the sample was female. The average age of the respondents was 40 years and their average annual income was £36,400. All the variables shown in Fig. 1 satisfied standard tests for normality, apart from relationship quality (which exhibited a right hand skew consequent to most participants trusting their selected charity and being satisfied with its work), and (ii) self-congruence, which had a left hand skew. Squaring the values of the latter variable and taking the square roots of the values of the former variable transformed these measures into the normal distributions required for structural equation modelling. Nine per cent of the sample could be said to be highly engaged with their nominated charity, in the sense that their responses fell in the top category of the composite formed to reflect this construct. Thirteen per cent of the responses fell in the next category of the composite; 34 % of the responses were in the bottom two categories.

Figure 1 was estimated using the structural equation modelling facility of the AMOS 19 package. The model was estimated five times: firstly for level of giving over a 12 month period, secondly for frequency of giving during the previous year, then for the scores recorded on the word of mouth variable and on the two future giving intention items. The simultaneously estimated regression coefficients pertaining to various sections of the model are presented in Tables 1 and 2. Summary goodness-of-fit diagnostics for the total model are appended to Table 2. It can be seen from the tables that self-congruence exerted a marginally significant (p = 0.05) impact on donor behaviour (confirming H6 [b]), but did not affect donor engagement (H6 [a]), contradicting the argument of Sprott et al. (2009) and others (see Van Doorn et al. 2010) that self-identity influences engagement behaviour. Thus the extent to which supporting a certain charity enabled a donor to affirm his or her self-proclaimed values, beliefs and self-identity did not usually impact on the degree to which the person wished actively to engage with the organisation. Self-congruence may have had a symbolic role in an individual’s financial support for a charity, but it did not result in animated engagement behaviour. The model offered good explanation of donation level, level of positive word of mouth and future giving intentions. Frequency of donating was however less well-explained by the posited variables. Table 1 shows that all three elements of donor experience had significant effects on relationship quality and on donor engagement, except that reputation did not impact significantly on closeness. Thus H1 (a) and (b), H2 (a) and (b) are confirmed, and H3 (b) is partially supported. H3 (a) is accepted. Trust (H4 [a]) and closeness (H5 [a]) had significant effects on engagement.

Table 1 Determinants of donor engagement and relationship quality
Full size table
Table 2 Determinants of donor behaviour
Full size table

The influences of the hypothesised explanatory variables on donor behaviour are listed in Table 2, from which it can be seen that all the variables significantly affected (p < 0.05) donation level, word of mouth, and intention to continue to support the charity. This was not the case for the prediction of frequency of donation or intention to increase future levels of support. Service quality, reputation and self-congruence failed to attain significance in the estimation of frequency of donation. Only four variables had significant effects in the estimation concerning future intended increases in levels of support: trust, closeness, engagement and self-congruence. Hence, while relationship marketing activities, service quality and a charity’s reputation influenced intentions to continue support, they did not impact on intentions to increase donations.

6 Conclusion

The investigation sought to develop themes arising from recent literature regarding customer engagement in the commercial sector and to apply the engagement construct to the nonprofit domain. An instrument for measuring the levels of engagement of charity donors with the organisations they had chosen to support was created and incorporated into a model of the antecedents and consequences of engagement. A test of this model showed that engagement exerted a highly significant impact on behaviour. It follows that activities likely to stimulate engagement should be financially worthwhile. Apart from self-congruence, the determinants of engagement (relationship marketing, service quality, trust, closeness, etc.) that have been proposed by much of the emerging literature in for-profit field offered sound explanation of the sample members’ degrees of engagement. Engaged donors gave more, were substantially more positive vis-à-vis their word of mouth, and were more likely to intend both to support their chosen charity and to increase their levels of giving. Clearly, therefore, the engagement construct has a great deal to offer in the fundraising management field. Fundraisers need to look beyond conventional relationship marketing and to identify communications and activities likely to engender deep, passionate and enduring connections between a donor and the charity the person supports. Managements should avoid reliance on simplistic models of trust and commitment, recognise the importance of the wider concept of engagement, and actively seek to influence the triggers of donor engagement. Communications that emphasise a charity’s trustworthiness, illustrate its successes, and which stimulate feelings of closeness to the organisation are especially important in this respect. Closeness could be encouraged via the establishment of social networks to which donors can contribute and which bring together a charity’s current and potential supporters. Within these networks donors themselves may influence each other to become more engaged, both with fellow supporters and with the charity.

A crucial implication of the findings is that since engagement-prone charity supporters are likely to have higher lifetime donor values, it is desirable to target these donors in recruitment and retention campaigns. This requires the establishment of procedures for identifying the engagement-prone; bearing in mind that not all donors will appreciate attempts at inducing them to become engaged. The creation of a generic instrument to measure an individual’s engagement-proneness in the charity donation sphere would represent an extremely valuable contribution to academic work in the engagement area. Another implication of the outcomes to the study is that charities should take ‘engagement management’ just as seriously as ‘relationship management’ and perhaps subsume the latter into the former. Many charities have relationship marketing managers. Arguably these individuals should instead become donor engagement managers. Charities must decide how exactly they want donors to engage with the organisation and then create mechanisms and systems (e.g., blogs, online communities) that make it easy for supporters to engage.

6.1 Limitations and areas for further research

A number of limitations apply to the study. The research was undertaken in a single sector (fundraising charities) in a single large city (London) in a single country (the UK). Thus, replication studies in other sectors and nations would be useful, as would the employment of a more extensive list of components of donor experience and of relationship quality. (It was not possible within the confines of an already crowded questionnaire to extend the range of sub-constructs considered by the present study.) The investigation included on exploratory attempt at operationalising an important construct, engagement, that to date has not been the subject of substantial empirical research. The paucity of empirical work in the area creates many opportunities for further studies. In particular it would be worthwhile experimenting with various configurations of elements of donor (customer) engagement given that, while representing disparate sub-constructs, many of these elements rise and fall simultaneously. Hence it is difficult to establish whether variables associated with specific elements are antecedents or outcomes. Palmer (2010) observed how, in the field of customer experience, ‘the greatest problem in developing an operationally acceptable measure of customer experience is the complexity of context specific variables’ (p. 202). The same remark could be made of customer engagement. For instance, is perceived service quality a cause of engagement (as assumed by the present study) or might it be a consequence. Is frequency of contact with an organisation an effect of engagement or is it a determining factor? Clearly there is much to discover in the customer engagement area.