Sage Journals: Discover world-class research

Abstract

This article provides an overview of the state-of-knowledge on non-donor research, identifies key reasons for not donating money, and presents evidence of non-donor clusters based on representative data from the German Donation Monitor. The study collected and analyzed a sample of 1,908 non-donors. The descriptive results show that statements like “I cannot afford to donate”; “A too high percentage of the donation is used for administrative purposes”; and “I don’t like personal donation appeals” were the top three reasons for non-support. A cluster analysis identified five clusters and labeled them categorical, unsolicited, indifferent, restrictive, and convincible non-donors. Convincible non-donors had the highest intention to donate in the future and were predominantly composed of younger, well-educated men. Based on the findings, the article explains a non-donor portfolio matrix and recommends a focus on convincible as well as restricted non-donors in the future recruitment strategies of nonprofits.

Keywords

non-donors non-donor segmentation donor recruitment relationship fundraising

Introduction

Given the intense competition in donor markets worldwide (Harrison & Thornton, 2022), an increasing number of nonprofit organizations are critically reflecting on their existing fundraising approaches and strategies to appeal to a broader donor population. One approach to attract new prospects beyond the existing donor base is to focus on non-donors. The non-donor segment is generally assumed to be very large and varies across countries (Rajan et al., 2009; Rooney et al., 2020; Shehu et al., 2015). However, due to a lack of information, it is mostly described as a homogeneous segment. In Germany, Shehu et al. (2015) found that 42.96% of the population does not donate money, time, or blood. Rooney et al. (2020) reported a similar proportion of non-donors (40%) in the United States but offered a more nuanced view, arguing that this proportion may be overestimated, because “two thirds of non-donors give to nonprofit organizations at least once within eight years” (p. 729). This aligns with Rajan et al.’s (2009) finding that 12.3% of Canadians do not donate money.

Apart from size, non-donors are surprisingly under-researched (Chatzidakis et al., 2016; Lange, 2024). Most studies have analyzed the sociodemographic criteria of current donors and compared their findings with lapsed or non-donors (Keyt et al., 2002; Riecken & Yavas, 2005; Schlegelmilch et al., 1997). The most comprehensive study on why people do not donate money was conducted by Sargeant et al. (2000), who showed that financial constraints played a key role. A few studies explicitly clustered non-donors (Martín-Santana et al., 2020; Reid & Wood, 2008); however, these studies were conducted within the context of blood donation. For example, Reid and Wood (2008) categorized blood non-donors based on their intention to donate blood in the future, finding that 61.7% were low intenders and 38.3% were high intenders.

In summary, as non-donors can no longer be considered a homogeneous group, an investigation into the heterogeneity of non-donors, particularly in the context of monetary donations, is long overdue. There is a need to examine non-donor behavior in a way that aligns with the growing efforts of nonprofit organizations to explore new funding opportunities in response to the evolving demands of existing donor structures.

This study’s overarching research goal is to contribute to a better understanding of the heterogeneity of non-donors in a money donation context in Germany. Two research questions guide this study, which will be answered using representative data from the German Donation Monitor 2022: (a) What are the main reasons why non-donors in Germany do not donate money to nonprofit organizations? (b) Is it possible to identify distinct clusters of non-donors? The findings could also help nonprofit organizations develop strategies and actions to target non-donor subgroups most likely to start donating (Burnett, 2023). Therefore, the practice-oriented question guiding this study is: (c) Should nonprofit and fundraising managers proactively try to persuade non-donors to donate money to their organization?

Literature Review on Non-Donors, Definitions, and Reasons for Non-Support

First, we reviewed the current literature on non-donors (see Supplement 1). We decided against using a systematic literature review approach (e.g., Gazley, 2022) because many donor studies only mention non-donors in passing. Instead, we conducted a semi-systematic literature review (see Sauer & Seuring, 2023; Snyder, 2019). Generally, a semi-systematic review gives an overview of the state-of-knowledge on the topic of interest and highlights subthemes; moreover, the approach is designed for research areas that have been studied within diverse disciplines and contexts. This is particularly true in our case, as non-donor issues have been researched in the context of money, time, and blood donation.

We undertook a multi-step selection process, starting with searches in literature databases using keywords such as “non-donors,” “segmentation,” “donor profiles,” and “donors versus non-donors.” Next, we reviewed the reference lists of key articles on non-donors and checked the websites of nonprofit journals (e.g., NVSQ, NML, Voluntas, Journal of Nonprofit & Public Sector Marketing; Journal of Philanthropy, previously Journal of Philanthropy and Marketing) for additional studies on non-donors. We defined a study as relevant based on an intensive qualitative assessment of whether the topic of non-donors featured prominently in the article, focusing on non-donor articles that either mentioned reasons for non-support and/or included a segmentation component. We compiled a list of 26 articles on non-donors: the oldest by Schlegelmilch et al. (1997) and the most recent by Lange (2024). With the exception of a conceptual article by Lange (2024), all other studies were empirical and used non-donor data in the context of money, time, and blood donations. From the articles summarized in the online supplementary material, Supplement 1, we analyzed whether (a) a specific definition of non-donors was used, (b) the reasons for non-support were surveyed and analyzed, (c) a segmentation approach was employed and, if so, which non-donor segmentation criteria were used, and (d) whether a cluster analysis was conducted and which clusters of non-donors were identified.

A consequence and limitation of semi-systematic literature reviews is the occurrence of evidence selection bias because the review does not identify all available articles on the topic of interest (Drucker et al., 2016). For example, this review excluded a vast body of literature on donor segments (e.g., Veldhuizen et al., 2013) and donation barriers (e.g., Duboz & Cunéo, 2010; Piersma et al., 2017; Sundeen et al., 2007).

Definition of Non-Donors

Among the selected articles, 16 defined the term non-donor in more detail. For example, Riecken and Yavas (2005, p. 574) used a strictly behavioral conceptualization in their approach. In their study, current donors were defined as “those who had donated within the past 12 months”; lapsed donors as individuals “who had donated before but not within the last 12 months”; and non-donors as those “who have never donated” money to the researched project (March of Dimes project). Other authors used a broader approach and defined non-donors as those who had not “donated blood for at least 36 months” (Saltzmann & Boenigk, 2022, p. 193) or “who had not attempted to donate in the last year” (e.g., Masser et al., 2016, p. 2996).

We discussed the definitions with our cooperation partner—the German Fundraising Association (Deutscher Fundraising Verband [DFRV]). In 2022, the DFRV invited us to cooperate on the topic of non-donors and incorporate a few additional non-donor questions into their online donor survey called the German Donation Monitor. Thus, the study data were sourced from the DFRV’s annual survey. The DFRV also recommended we use “not donated in the last 12 months” to define non-donors, because it is the most common definition applied in German fundraising practice.

Reasons for Non-Support

Academic research on the reasons for non-support in the context of monetary donations is also limited. Sargeant et al. (2000, p. 323) provided a comprehensive list of reasons for non-support in the context of monetary donations. The authors used data from a survey of 980 donors and 249 U.K. non-donors and identified seven key reasons why people are unwilling to donate: “I cannot afford to offer my support to the charity” (23.3%); “charities ask for inappropriate sums” (22.5%); “the government should fund the work undertaken by charities” (19.3%); “I find charity communication inappropriate” (12%); “the quality of service provided by charities to their donors is poor” (6.8%); “in the past charities have not acknowledged my support” (4%); “I feel charities are not deserving” (2.8%); and “Other reasons” (9.3%). Lange (2024, p. 34) showed that, in addition to academic papers, market research surveys have also been conducted by foundations such as the Observatoire du Don en Confiance in France with comparable results. In the latter’s survey, respondents provided the following reasons for not donating money: “the lack of trust to use the funds” (32%); “lack of disposable income” (27%); “not feeling a sense of solidarity with the cause” (10%); “lack of knowledge” (9%); “not seeing the use” (6%); “not knowing which cause to choose” (6%); “feeling like they contribute through taxes” (5%); and “feeling like they help with other daily actions” (3%).

Non-Donor Segmentation Criteria

Next, we examined the selected articles to identify the most frequently used segmentation criteria. We differentiated three segmentation criteria: socioeconomic, psychographic, and behavioral criteria. Previous studies predominantly compared the sociodemographic characteristics of donors to those of non-donors (Supplement 1) and highlighted the following characteristics of non-donors: Gender: non-donors tend to be male (Keyt et al., 2002); income: non-donors tend to have a lower income level (Schlegelmilch et al., 1997); age: non-donors tend to be younger (Riecken & Yavas, 2005; Schlegelmilch et al., 1997); marital status: non-donors tend to be unmarried (Riecken & Yavas, 2005); household size: non-donors are less likely to have children or live in a single-person household (Riecken & Yavas, 2005); education: non-donors tend to have a lower level of education (Keyt et al., 2002; Sargeant et al., 2000); and, finally, religiosity: non-donors are less likely to be religious (Dury et al., 2015). These sociodemographic characteristics have been confirmed in studies on time donation, specifically non-volunteering (Lockstone-Binney et al., 2022; Niebuur et al., 2019; Reed & Selbee, 2000; Taniguchi, 2010).

Regarding psychographic segmentation criteria, previous studies have shown that non-donors tend to have a negative attitude (perception) toward the work of nonprofit organizations (Keyt et al., 2002; Lange, 2024; Lwin et al., 2014) and perceive their financial ability to support social causes as low (Lange, 2024; Sargeant et al., 2000); both criteria were integrated into this study. Another psychological criterion, highlighted by Saltzmann and Boenigk (2022), is consciousness of the decision to end donations. Some individuals consciously stop donating and are fully aware that they have become non-donors. Others may not have donated in the preceding 12 months for various reasons but do not perceive themselves as a non-donor. Therefore, we integrated consciousness as a psychological criterion in our analysis.

Moreover, previous studies have confirmed that non-volunteering is a useful behavioral segmentation criterion (Niebuur et al., 2019; Rajan et al., 2009; Schlegelmilch et al., 1997). Therefore, we included engagement-related questions in our study to determine whether non-donors are not donating money but are interested in other forms of social engagement, such as volunteering, in-kind donations, attending fundraising events, or buying products advertised in cause-related marketing campaigns. Finally, lifestyle characteristics, such as social media behavior, leisure time behavior, and typical physical activities in daily life, are potential behavioral characteristics that are often useful in consumer studies (Vyncke, 2002). As an additional evaluation, we report our results for the two additional analyses.

Existing Knowledge on Non-Donor Clusters

Among all 26 articles, we identified only three studies (see Supplement 1; all in the blood donation context) that have conducted an initial analysis to discover more heterogeneity within the non-donor segment. Based on an individual’s intention to donate blood again, the authors identified two non-donor segments, labeled as “high intenders” (38.3%) and “low intenders” (61.7%) (Reid & Wood, 2008). However, the data analysis of this study was based on regression and not a cluster analysis. Influencing factors were found to be subjective norms, perceived behavioral control, and time-related factors.

Martín-Santana et al. (2020) conducted a study of 2,382 non-blood donors in Spain using a latent class analysis for segmentation. The authors focused on non-donors’ motivation and barriers, differentiating six non-blood donor clusters: (a) impure altruists, (b) “I want to, but make it easy for me,” (c) free-riders, (d) reciprocal altruists, (e) “I can’t because I’m scared,” and (f) “I want to, but I can’t.” They concluded that segmenting non-donors is an important precondition to address the right non-donor groups with targeted marketing actions. Finally, Boenigk and Leipnitz (2016) clustered blood non-donors in big cities in Germany and identified seven urban blood donor segments and labeled them enthusiastic, easy-living, corporate, event, high-quality, informational, and uninterested clusters. Consequently, previous studies on non-donor clusters in a money donation context were not identified.

Methodological Approach

Data Collection

This study’s data were collected within the scope of the German Donation Monitor, an annual representative online survey published by the DFRV (2022). This comprehensive survey provides information on the donation behaviors of German citizens between the ages of 14 and 70 and thus contains representative information on both donors and non-donors in Germany. Until 2021, the questionnaire only contained detailed questions for donors. In 2022, the DFRV invited the authors of this study to cooperatively investigate the topic of non-donors and incorporate several questions on non-donors into the DFRV’s survey instrument.

The data collection process took place in November 2022 and was administered by an external market research institute. This yielded a total sample of 5,059 (100%) respondents: 2,696 donors (53.3%), 1,908 non-donors (37.7%), and 455 missing information (9%). The market research institute sourced participants from their existing data pool of individuals living in Germany, who had agreed in advanced to be contacted for online surveys. By implication, the population reached was confined to those with enough technology skills to participate in online surveys. The study utilized robust sampling methods to achieve representativeness for the German population. Participant selection and the subsequent weighting of responses were based on sociodemographic criteria, including gender, age, and federal household size. This ensured the survey’s representativeness of both monetary donors and non-donors in Germany.

Questionnaire and Scale

Overall, the online questionnaire of the German Donation Monitor 2022 contained 34 overarching questions structured into several parts. The first part contained multiple questions on respondents’ donation attitudes and behaviors. Only one question was relevant for our data analysis: “Have you donated money within the last 12 months?” (yes/no); the “no” answers allowed us to identify the subsample of 1,908 non-donors.

The second part of the questionnaire focused on non-donors and included questions on (a) reasons for non-support, (b) non-donors’ opinions and behaviors, (c) willingness to donate in the future, and (d) additional questions on nonprofit brand sympathy. At the end of the survey, all respondents completed the sociodemographic and lifestyle questions.

As mentioned earlier, the questionnaire was developed based on the DFRV’s (2022) existing donor monitoring questionnaire. Therefore, we were restricted to the survey’s pre-existing format, which limited our ability to make independent methodological choices, such as selecting a more detailed 5- or 7-point Likert-type scale. Thus, we used the existing 4-point scale to measure questions related to non-donors. The scale structure ranged from 1 = strongly agree to 4 = strongly disagree. Respondents were not required to reply to all the questions and also had the option to select 5 = I don’t know or 6 = no answer. As a result, the total number of answers for each question reported in our descriptive statistics (Supplement 2) varied due to missing values.

Measures

Supplement 2 gives a full overview of all the measures and descriptive statistics. The researchers translated the questionnaire’s questions from German to English and re-checked each measure with a DFRV fundraising expert. It is important to note that all items measured in our survey are represented in the literature.

Reasons for Non-Support

The respondents were asked to answer 18 statements on the overarching question, “For what reasons do you not donate?” We applied two approaches to finalize the list of reasons for non-support included in the questionnaire. First, we replicated all seven reasons for non-support measured by Sargeant et al. (2000). Second, we extended the list with 11 additional reasons for non-support based on the literature (see Supplement 2; Harrington et al., 2007; Willems & Dury, 2017). Before including them in the questionnaire, we discussed each question with a DFRV fundraising representative. The relative importance of the reasons for non-support varied, with mean values ranging between 2.06 and 3.59 on a 4-point scale (with 4 indicating strongly disagree).

Non-Donors’ Opinions and Behaviors

We also measured nine additional statements that expressed respondents’ psychological thinking (Supplement 2; items 19 to 27). For example, respondents were asked to evaluate statements like “I have the feeling that I cannot afford donations” and “I have made a conscious decision not to donate money” (adopted from Saltzmann & Boenigk, 2022). Nevertheless, it was sometimes difficult to differentiate between “reasons for non-support” and general “opinions/behaviors” and this therefore needed to be checked in our factor analysis. Mean values ranged from 2.24 to 3.56.

Future Willingness to Donate Money

Next, we asked four more questions on non-donors’ willingness to donate money in the future, which was overall not very high, with mean values ranging between 2.46 and 3.41. The measurement approach was based on Reid and Wood’s (2008) study on high (“From next year I will be there in any case and donate” and “I will most likely start donating money”) and low (“I will deliberately not donate money next year either” and “I could imagine donating money in the future”) intenders. A reliability analysis (e.g., Cronbach’s alpha) led to the retention of three items for our future donation intention scale: “I could imagine donating money in the future,” “I will most likely start donating money,” and “From next year I will be there in any case and donate.”

Additional Measures

The last sections of the non-donor questionnaire contained an additional open question on nonprofit brand sympathy (“Currently, I do not donate, but I like the following nonprofit organizations best”). This question was not the focus of our cluster analysis and therefore not analyzed in detail. However, a first descriptive evaluation shows that non-donors in Germany sympathized most with the nonprofit brand Médecins Sans Frontières (Ärzte ohne Grenzen), with 120 mentions, and Children’s Cancer Aid (Kinderkrebshilfe) with 51 mentions (Supplement 2).

Sociodemographic/Lifestyle Factors

At the very end of the survey, questions were asked about the respondents’ sociodemographic characteristics, such as gender (female; male; diverse), age (seven age groups from 14–19 years to 65+ years old), education (seven educational subcategories), occupation (eight subcategories), personal income (low = less than 1,400 euros per month; high = over 4,000 euros per month), marital status (single; married; widowed/divorced), household size (1 to 5+), and religion (5 subcategories). The results are summarized in Table 1. Respondents were asked additional questions to evaluate their engagement and social media behaviors (lifestyle) (see Supplements 2 and 3).

Table 1

Sample Characteristics of Donors and Non-Donors.

Variable	Specification	Donor sample (N = 2,696; 100%)		Non-donor sample (N = 1,908; 100%)		Group difference (p)
Gender	Female	1,342	49.8%	892	46.8%	*
	Male	1,348	50.0%	1,014	53.1%	*
	Diverse	5	0.2%	2	0.1%
Age	14–19 years	67	2.5%	102	5.3%	***
	20–29 years	434	16.1%	349	18.3%	*
	30–39 years	450	16.7%	284	14.9%
	40–49 years	432	16.0%	295	15.5%
	50–59 years	544	20.2%	368	19.3%
	60–64 years	247	9.2%	210	11.0%	*
	65+ years	521	19.3%	299	15.7%	*
Education	Without school leaving qualification	12	0.4%	10	0.5%
	Secondary modern school	156	5.8%	178	9.3%	***
	Middle school	463	17.2%	395	20.7%	*
	Vocational education	497	18.4%	453	23.8%	***
	A-levels	637	23.6%	458	24.0%
	University degree	848	31.5%	381	20.0%	***
	PhD/Doctorate	74	2.7%	26	1.4%	*
	No answer	7	0.3%	6	0.3%
Occupation	Pupil/student	172	6.4%	180	9.4%	***
	Unemployed	71	2.6%	136	7.1%	***
	Worker	169	6.3%	157	8.2%
	Employed	1399	51.9%	843	44.2%	***
	Freelancer	75	2.8%	50	2.6%
	Self-employed	120	4.5%	58	3.0%	*
	Independent farmer	6	0.2%	0	0.0%	*
	Retired	604	22.4%	415	21.8%
	No answer	79	2.9%	68	3.6%
Personal income	Low income (less than 1,500 €)	749	27.8%	788	41.3%	***
	Middle income (up to 4,000 €)	1409	52.3%	863	45.2%	***
	High income (over 4,000 €)	322	11.9%	113	5.9%
	No answer	215	8.0%	144	7.5%
Marital status	Single	798	29.6%	768	40.3%	***
	Married	1582	58.7%	829	43.4%	***
	Widowed/divorced	293	10.9%	295	15.5%	***
	No answer	22	0.8%	16	0.8%
Household size	1	470	17.4%	499	26.1%	***
	2	1022	37.9%	741	38.8%
	3	525	19.5%	340	17.8%
	4	335	12.4%	190	10.0%	*
	5+	343	12.7%	139	7.3%	***
Religion	Protestant	725	26.9%	377	19.8%	***
	Catholic	701	27.1%	364	19.1%	***
	Muslim	128	4.8%	30	1.6%	***
	Other confession	76	2.8%	53	2.8%
	No confession	924	34.3%	879	46.1%	***
	No answer	34	1.3%	22	1.1%
	Missing	107	4.0%	183	9.6%

Note. N = 1,908 (100%). Data are weighted to reflect the composition of the German population. Weighted sample sizes are reported; totals may not sum exactly due to rounding.

p = .05. **p = .01. ***p = .001.

Sample Characteristics

Table 1 displays the characteristics of the donor and non-donor samples. The donor sample served as a reference to show whether significant sociodemographic differences existed between the two segments. As previously noted, the sample is representative of the general German population aged 14 to 70 years old. In Germany, there are no age restrictions or legal barriers that prevent individuals under 18 years from donating money. Thus, the youngest age segment in the German Donation Survey is always 14 to 19 years.

Gender

We confirmed the findings of previous studies that non-donors are more likely to be male. Within our non-donor sample, 53.1% of the respondents identified as male, and 46.8% as female, whereas the ratio in the donor sample was almost equal.

Age

In our sample, non-donors were present in all age groups. In line with previous findings on non-donor characteristics (Schlegelmilch et al., 1997), we found that non-donors tended to be younger. The age groups 14 to 19 years old and 20 to 29 years old were both significantly overrepresented in the non-donor sample compared with the donor sample.

Education/Occupation

In Germany, non-donors tend to have lower levels of education. In this study, the proportion of non-donors with a university degree (20%) was significantly lower than in the donor sample (31.5%). The employment rate in the non-donor sample (44.2%) was significantly lower than in the donor sample (51.9%). The data also showed that non-donors tended to be students or the unemployed.

Income

Unsurprisingly, non-donors in our sample had lower income levels, with 41.3% earning less than 1,500 euros per month, which is a significantly larger proportion compared with the donor sample (27.8%).

Marital Status/Household Size

Non-donors were more likely to be single or unmarried. The number of 1-person households was therefore significantly higher in the non-donor sample (26.1%) than in the donor sample (17.4%).

Religion

We also found that non-donors were less likely to be religious. In the non-donor sample, 46.1% answered that they had no confession of faith, compared with only 34.3% in the donor sample. In conclusion, the descriptive results of our non-donor sample confirmed the findings of earlier studies on the typical sociodemographic characteristics of non-donors. We did not find completely new or divergent results.

Descriptive Results on Reasons for Non-support

Figure 1 provides an overview of the descriptive results for the top ten reasons for non-support. For each statement, we show the sum of the positive responses (1 = strongly agree and 2 = very much agree) and contrast it with the negative responses (3 = do not agree so much and 4 = strongly disagree). The 10 reasons are structured in descending order of the mean values, starting from the highest agreement with a mean value of 2.06 to the lowest with a mean value of 2.70. The missing percentages reflect the no choice options.

Figure 1.

Top 10 Reasons for Non-Support.

The reason with the highest ranking (M: 2.06) reflected non-donors’ perceptions of how nonprofit organizations work: 55.1% of the respondents agreed with the statement “A too high percentage of the donation is used for administrative purposes,” which confirmed assumptions often mentioned in discussions on the starvation cycle (Schubert & Boenigk, 2019). Non-donors also cited their economic situation (59.7%; M: 2.12) as a key reason for not donating money, which aligns with Sargeant et al.’s (2000) study in the United Kingdom. Regarding fundraising solicitations, 54.3% of the non-donors said they disliked direct donation appeals (M: 2.18). The economic situation at a macro level was directly linked to financial reasons at an individual level (M: 2.27). In this regard, 52.1% of the respondents agreed with the statement “I no longer donate for financial reasons, for example unemployment.” Low trust in how nonprofit organizations use donations was another key issue for non-donors (48.2%; M: 2.35). Interestingly, and as yet not identified in previous studies, non-donors also identified life disadvantages as a reason for non-support, agreeing with the statement “I also don’t get support” (48.0%; M: 2.35). Furthermore, 45.2% of non-donors agreed with the statement “The state and not the individual is responsible, to solve social problems” (M: 2.43), while 42.9% agreed with the statement “Donating is the wrong approach and that economic solutions are needed” (M: 2.49). Regarding the statement “Donating is not important for me,” 32.6% of non-donors agreed (M: 2.68), whereas 48.9% disagreed. Finally, only one-third (34.2%) of non-donors agreed with the statement “Most people donate only to get social recognition,” while 47.3% disagreed (M: 2.70). Supplement 2 summarizes the descriptive results not visualized in Figure 1, including missing values and standard deviations.

Data Analysis

Our methodological approach followed the general steps undertaken in segmentation/cluster analysis studies (Durango-Cohen et al., 2013; Jedicke et al., 2024; Kolhede & Gomez-Arias, 2022; Rupp et al., 2014). First, we conducted an exploratory factor analysis (EFA) to identify the underlying dimensions of the reasons for not donating money and to obtain a smaller, more manageable number of statistically independent variables (Clifford et al., 1995; Kolhede & Gomez-Arias, 2022). We performed the factor analysis based on the reasons for non-support (18 items) and non-donors’ opinion and behaviors (9 items), resulting in a total of 27 items (see Supplement 2). We applied principal component analysis with VARIMAX rotation. Next, we conducted a cluster analysis using factor means as inputs for the clustering procedure, which is appropriate for an explanatory research design (Hair et al., 2018). The clusters were then profiled and compared based on the mean responses of each factor. Clusters were also enriched with sociodemographic variables and examined for their intention to donate.

Exploratory Factor Analysis

We conducted an EFA to better understand whether there was any underlying factor structure in the set of 27 items. The Kaiser–Meyer–Olkin (KMO) test yielded a value of 0.899, and Bartlett’s test of sphericity was significant (p < .001). These results indicated that the dataset was appropriate for factor analysis. Next, we applied principal component analysis with VARIMAX rotation to determine the factor structure. We examined the rotated factor loadings and excluded items with factor loadings less than 0.5 or those that showed high cross-loadings, with 0.4 as the upper limit (Willems & Dury, 2017). This iterative process was repeated to refine the factor solution. Following this process, we reduced the pool from 27 to 22 items and deleted the following five items (items marked “Items deleted” in Supplement 3): No. 13 (“Fortunately, enough other people donate”), No. 15 (“I no longer donate due to family changes”), No. 16 (“I no longer donate because of negative experiences with nonprofits”), No. 18 (“People in my circle of family and friends don’t like it when I donate”), and No. 27 (“I donate money, but only when friends or family ask me to”). The final rotated factor matrix is shown in Supplement 3.

As shown in Table 2, the EFA using the final pool of items resulted in four overarching factors—(a) negative attitudes toward donating, (b) perceived financial inability, (c) past donation experiences and unconsciousness, and (d) solicitation—all with eigenvalues greater than 1, and together accounting for 58% of the explained variance. We also evaluated the measurement quality of the EFA by assessing item and construct reliability. Item reliability was measured by examining factor loadings, with the expectation that reflective items would have loadings greater than 0.7. Most items satisfied this criterion. The other items exhibited loadings of 0.6, and two items had loadings of 0.5, which were still deemed acceptable (Boenigk & Leipnitz, 2016). Construct reliability, determined using Cronbach’s alpha, ranged from excellent values of 0.92 (F1) and 0.85 (F2) to an adequate value of 0.68 (F3) and a low value of 0.52 (F4).

Table 2.

Results of the Exploratory Factor Analysis (Rotated).

Summary of the factor structure	Factor number	Eigen-values	Variance explained	Cronbach’s alpha
Factor 1: Negative attitude toward donating	12	6.8	31.1%	.92
Factor 2: Perceived financial inability	3	2.8	12.5%	.85
Factor 3: Past donation experience and unconsciousness	5	1.9	8.8%	.68
Factor 4: Solicitation	2	1.2	5.6%	.52
EFA results on a one-item basis	Factor loadings	Eigen-values	Variance (cumulative)	Cronbach’s alpha
Factor 1: Negative attitude toward donating		6.8	31.1%	.92
Donations are the wrong approach; economic solutions are needed.	.787
I don’t like personal donation appeals. They are too intrusive for me.	.759
I don’t really trust the nonprofit organizations will use my money well.	.746
Donating is not important to me.	.741
Most people donate only to get social recognition. I don’t need that.	.740
It makes no difference whether I donate or not.	.734
A too high percentage of the donation is used for administrative purposes.	.717
Nothing is given to me either.	.709
I believe that the state is responsible for solving social problems, not me.	.693
I categorically reject monetary donations.	.690
I have made a conscious decision not to donate money.	.689
There are too many negative reports in the media about nonprofit organizations.	.524
Factor 2: Perceived financial inability		2.8	12.5 (43.6%)	.85
EFA results on a one-item basis	Factor loadings	Eigen-values	Variance (cumulative)	Cronbach’s alpha
In my economic situation, I cannot afford to donate.	.903
I no longer donate for financial reasons, e.g., unemployment, lower income . . .	.874
I feel like I cannot afford to donate.	.791
Factor 3: Past donation experience and unconsciousness		1.9	8.8 (52.4%)	.68
I donate money occasionally (e.g., every 3 years).	.729
I don’t consciously donate no money, I just have not donated in a while.	.704
I used to donate money, but I don’t anymore.	.691
I donate money but only when there is a major disaster.	.596
I could certainly imagine donating, but I have not yet.	.512
Factor 4: Solicitation
I have never been asked by a nonprofit organization to donate anything.	.789	1.2	5.6 (58.0%)	.52
I don’t have enough time to think about who or what to donate to.	.700

Note. Varimax rotated principal component analysis. Statements were translated from German to English. What are your reasons for not donating? How much do you agree with the following statements? Statements were measured on a 4-point scale with 1 = strongly agree, 2 = very much agree, 3 = do not agree so much, 4 = strongly disagree. KMO = .899, Bartlett test = .000.

Negative Attitude Toward Donating

The first factor, “Negative attitude toward donating,” had an eigenvalue of 6.8 and explained 31.1% of the total variance. This factor included 12 items reflecting negative attitudes toward donating, with high internal consistency (Cronbach’s alpha = .92). However, the overall finding that non-donors exhibit a negative attitude toward donating is not new and has been confirmed in previous non-donor studies (Keyt et al., 2002; Sargeant et al., 2000).

Perceived Financial Inability

The second factor, “Perceived financial inability,” had an eigenvalue of 2.8 and explained 12.5% of the variance. This factor encompassed three statements reflecting the perception of financial inability to donate. It is important to note that these items reflect respondents’ perceptions of their financial situation rather than actual income or ability (Lwin et al., 2014; Riecken & Yavas, 2005; Sargeant et al., 2000).

Past Donation Experiences and Unconsciousness

The third factor summarized five statements on past donation experiences. The factor had an eigenvalue of 1.9 and explained 8.8% of the variance. Of interest is that, while some individuals perceived themselves as non-donors, they agreed with the statement, “I don’t consciously donate no money, I just have not donated in a while.” The other statements, for example, “I donate occasionally, every three years,” supported the impression that sometimes people do have past donation experience, although they currently do not donate money. Cronbach’s alpha value was .68, which is barely acceptable.

Solicitation

The fourth factor was the most internally discussed and re-checked factor because it summarizes only two items (“I have never been asked by a nonprofit organization to donate anything” and “I don’t have enough time to think about who or what to donate to”), which had an eigenvalue of 1.2 and still explained 5.6% of the variance. At first glance, the two items might appear to be relatively unrelated, and the Cronbach’s alpha value is low at .52 (we additionally computed the Spearman-Brown coefficient, which yielded the same value of 0.52). In contrast, both items demonstrated strong factor loadings in the EFA (.789 and .700), with no significant cross-loadings, suggesting the presence of a potentially meaningful latent construct. As noted by Furr and Bacharach (2013) and Vaske et al. (2017), such divergence between internal consistency and factor loadings is not uncommon when using limited-response ranges (like ours, 1 to 4).

With this in mind, we discussed the foundations of solicitation, nonprofit marketing (Andreoni & Rao, 2011; Bhati & McDonnell, 2025; Guo & Saxton, 2018), and selective attention theory (Driver, 2001) regarding why the two statements might be related. Our reasoning is as follows: first, solicitation in its different forms (e.g., social media posts, posters, radio commercials, TV commercials, and fundraising mailings/letters to households) is key to attracting potential donors (Bekkers & Wiepking, 2011; Schweidel & Knox, 2013). Guo and Saxton (2018), however, highlighted that nonprofit organizations often face an attention-deficit problem in regard to the general public. In today’s dynamic media world, people are saturated with information and have a very short attention span, and it is therefore very difficult for nonprofit organizations to engage existing and non-donors. We agree with this statement and argue that this holds true in particular for non-donors in Germany, because sending personalized letters or emails to non-donors (direct solicitation) is not permitted as per German law for data security reasons. Consequently, the statement “I have never been asked to donate” must hold true for most individuals except for those who are approached on the street by fundraisers (Waldner et al., 2020). Second, according to selective attention theory (Driver, 2001), not all individuals perceive those solicitation efforts equally. For those totally uninterested in donating money, these nonprofit marketing actions were barely recognized by the recipient, thereby reinforcing their perception that they had never been asked to donate. As donating is not in the non-donor’s evoked set, a statement like “I don’t have enough time to think about money donation” might be an elegant way to express non-interest. Based on the two reasonings and given that the factor structure was identified in a representative sample and the combined factor appeared to outperform the individual items in predicting relevant outcomes (we checked this with regression analysis), we ultimately decided—with some caution—to retain the two items as a single factor (Factor 4) for the purposes of this exploratory study.

Cluster Analysis

We performed a hierarchical cluster analysis using the mean scores of the four factors as cluster variables. The sample of 1,474 participants was restricted to those with complete data across all the factors’ means. As an initial step in the clustering process, we performed single linkage hierarchical clustering with the standard squared Euclidean distance metric to identify outliers and subsequently deleted 17 outliers. We selected single linkage for this preliminary analysis due to its sensitivity to distant, loosely connected points, which facilitated the detection of cases that did not align closely with the main clusters. Next, we used the hierarchical clustering algorithm with Ward’s inkage criterion to determine the number of clusters (Punj & Stewart, 1983; Reutterer & Dan, 2022). This suggested a five-cluster solution. We then refined the clustering using the non-hierarchical k-means algorithm, widely recognized for optimizing within-cluster homogeneity and maximizing between-cluster heterogeneity (Hair et al., 2018). This approach resulted in the five-cluster solution displayed in Table 3.

Cluster 1: Categorical non-donors (n = 244; 16.7%)

Cluster 2: Unsolicited non-donors (n = 178; 12.3%)

Cluster 3: Indifferent non-donors (n = 343; 23.5%)

Cluster 4: Restricted non-donors (n = 354; 24.3%)

Cluster 5: Convincible non-donors (n = 338; 23.2%).

Analyses of variance (one-way ANOVAs), followed by post hoc tests (Games-Howell) and contingency analyses, were employed to determine the quality of the cluster solution and examine inter-cluster differences.

Table 3.

Results of Cluster Analysis.

Cluster size	Total non-donor sample	Cluster 1 categorical non-donors	Cluster 2 unsolicited non-donors	Cluster 3 indifferent non-donors	Cluster 4 restricted non-donors	Cluster 5 convincible non-donors
Cluster sample sizes (N)	1,457	244	178	343	354	338
In percent (%)	100%	16.7%	12.3%	23.5%	24.3%	23.2%
Variables	Mean values						F (ANOVA)	Sig. (ANOVA)
Negative attitude toward donating	2.5	1.7^a	1.8^a	2.8^b	3.1	2.7^b	306.5	<.001
Financial inability	2.2	1.4^c,d	1.4^c,e	3.5	1.5^d,e	2.6	1121.5	<.001
Unconsciousness	3.3	3.7	3.3^f	3.5	3.2^f	3.0	82.9	<.001
Solicitation	3.2	3.8	1.8	3.6^g	3.6^g	2.6	757.7	<.001
Intention to donate	3.2	3.8	3.4^h	3.3^h	3.1	2.9	61.0	<.001

Note. Mean values are based on a 4-point scale: 1 = strongly agree; 2 = very much agree; 3 = do not agree so much; and 4 = strongly disagree. Group difference: ^{a,b,c,d,e,f,g,h}Clusters sharing the same letter are not significantly different from each other (p ≥ .05), based on the Games-Howell post hoc test. Reading help: a = Cluster 1 and Cluster 2 mean scores do not differ significantly regarding negative attitudes toward donating.

Cluster 1: Categorical Non-Donors (“I Am Strictly Against Donating”)

The statement “I am strictly against donating” best describes the opinion of categorical non-donors, who comprised 16.7% of the overall non-donor sample. Individuals in this cluster had the highest negative attitude toward donating (M: 1.7). Categorical non-donors stated that they were financially unable (M: 1.4) to donate. Moreover, they were very clear and aware of the fact that they did not donate, because the unconsciousness factor was the most pronounced among all non-donor clusters (M: 3.7). The mean value for “no solicitation” was 3.8, indicating that individuals in this cluster may have been approached by nonprofit organizations in the past and consciously chose not to donate. Not surprisingly, categorical non-donors showed very little intention to start donating in the future, with a mean value of 3.8 (the lowest possible intention score is 4.0). In other words, categorical non-donors are generally strictly opposed to donations. Regarding their demographic profile (Table 4), categorical non-donors were more likely to be female (59.8%) and aged 30 to 59 years (50%) or older (39.8%). Only 14.7% of individuals in Cluster 1 had studied at a university, compared with 19.5% to 29.6% in the other clusters.

Table 4.

Demographic Profiles of the Five Non-Donor Clusters.

Variables	Total non-donor sample (N = 1,457)		Cluster 1 categorical non-donors		Cluster 2 unsolicited non-donors		Cluster 3 indifferent non-donors		Cluster 4 restricted non-donors		Cluster 5 convincible non-donors		Chi²
Gender													81.59; p < .001
Female	686	47.1%	146	59.8%	85	47.8%	115	33.5%	211	59.6%	129	38.2%
Male	770	52.8%	98	40.2%	92	51.7%	228	66.5%	143	40.4%	209	61.8%
Diverse	1	0.1%	0	0.0%	1	0.6%	0	0.0%	0	0.0%	0	0.0%
Age													99.38; p < .001
Young (14–29 years)	344	23.6%	25	10.2%	54	30.2%	49	14.3%	108	30.5%	108	32.0%
Middle (30–59 years)	753	51.6%	122	50.0%	84	46.9%	208	60.6%	157	44.4%	182	53.8%
Old (over 60 years)	360	24.8%	97	39.8%	40	22.9%	86	25.1%	89	25.1%	48	14.2%
Education level													28.11; p = .005
Low	438	30.1%	85	34.7%	60	33.7%	92	26.8%	105	29.7%	96	28.4%
Middle	687	47.2%	124	50.6%	82	46.1%	162	47.2%	178	50.4%	141	41.7%
High	329	22.6%	36	14.7%	35	19.7%	89	25.9%	69	19.5%	100	29.6%
No answer	3	0.2%	0	0.0%	1	0.6%	0	0.0%	1	0.3%	1	0.3%
Personal income													164.60; p < .001
Low	598	41.1%	138	56.6%	79	44.4%	93	27.0%	203	57.5%	85	25.2%
Middle	683	46.9%	87	35.7%	79	44.4%	191	55.5%	116	32.9%	210	62.3%
High	88	6.0%	5	2.0%	9	5.1%	44	12.8%	7	2.0%	23	6.8%
No answer	87	6.0%	14	5.7%	11	6.2%	16	4.7%	27	7.6%	19	5.6%
Religion													22.41; p = .004
Religious	634	43.5%	99	40.5%	74	41.6%	125	36.4%	180	51.0%	156	46.2%
No confession	684	47.0%	129	52.9%	81	45.4%	184	53.7%	149	42.3%	141	41.9%
No answer	12	0.01%	1	0.4%	0	0.0%	3	1.0%	6	1.8%	2	0.6%
Missing	127	8.7%	15	6.2%	23	13.0%	32	8.9%	18	5.0%	39	11.3%

Note. Data are weighted to reflect the composition of the German population. Weighted sample sizes are reported; totals may not sum exactly due to rounding. Low education = without school and secondary school level; middle education = middle school level, vocational education; high education = university degree.

Cluster 2: Unsolicited Non-Donors (“I Don’t Want to and Have Never Been Asked”)

The unsolicited non-donors comprised the smallest group of non-donors (12.3%). Their profiles were similar to those of categorical non-donors (Cluster 1), with strong negative attitudes toward donation (M: 1.8) and strong perceived financial inability (M: 1.4). According to the Games-Howell post hoc test, these means did not differ significantly from those of Cluster 1 (p ≥ .05). The key difference between the two clusters is that individuals in Cluster 2 strongly agreed with the statement, “I’ve never been asked by a nonprofit organization to donate anything” (M: 1.8). Unsolicited non-donors also exhibited a low intention to donate in the future (M: 3.4). This cluster’s demographic profile was more balanced in terms of gender, educational level, and age, with a tendency toward younger age groups. Notably, 30.2% were in the 14 to 29 years age group, compared with 10.2% in Cluster 1. Perceived financial ability in this cluster was low, with 44.4% in the middle-income group. Similar to Cluster 1, unsolicited non-donors did not show increased involvement in other forms of engagement.

Cluster 3: Indifferent Non-Donors (“I Can, but I Don’t Want to”)

Indifferent non-donors comprised 23.5% of the overall non-donor sample. Interestingly, individuals in this segment did not have pronounced negative attitudes (M: 2.8), compared with Clusters 1 and 2. They considered themselves more or less financially able to support nonprofit organizations, as evidenced by their disagreement with statements regarding financial inability (M: 3.5), whereas financial inability was a strong component in Clusters 1 and 2 (M: 1.4). Respondents in Cluster 3 faced no significant constraints and demonstrated a relatively favorable motivational baseline. A significant proportion of this cluster fell within either the middle-income bracket (55.5%) or the high-income bracket (12.8%). This group had the highest income levels among the five clusters. The cluster had a high proportion of men (66.5%), with most of the male cluster members in the age groups 30 to 59 years old (60.6%). They were also comparatively well educated, with one quarter (25.9%) holding a university degree. Individuals were aware that they were non-donors (M: 3.5); nonetheless, their intention to start donating was, compared with Clusters 1 and 2, only slightly higher (M: 3.3). However, based on the Games-Howell post hoc test, the difference was not statistically significant when compared with Cluster 2 (p ≥ .05).

Cluster 4: Restricted Non-Donors (“I Would, but I Can’t”)

Cluster 4 was the largest cluster and represented 24.3% of the total sample. The cluster is best summarized by the statement, “I probably would donate money, but I can’t.” In this cluster, agreement with negative attitudes toward donations was the lowest compared with the other clusters (M: 3.1), indicating a more positive attitude toward donations. However, agreement on perceived financial constraints was high (M: 1.5), with almost maximum agreement on this factor. In other words, financial opportunities were perceived as more limited, especially compared with Cluster 3. Most of the respondents (57.5%) were in the low-income group, and only 2.0% indicated they had an income of more than 4,000 euros (upper-income bracket). They also had the second-highest score for intention to donate (M: 3.1).

Cluster 5: Convincible Non-Donors (“I Probably Would, but I Can’t Afford It Yet”)

Cluster 5 represented 23.2% of all the non-donor samples. Although they had not donated money in the past 12 months, they were likely to donate if their financial situation improved. Individuals were not rigid about their non-donor status, because they occasionally engaged in volunteering and other in-kind donations. Cluster 5’s mean of the unconsciousness factor was the highest among all the clusters (M: 3.0). Another finding was that the mean value for “no solicitation” was 2.6, the second most affirmative response, indicating that nonprofit organizations had not previously tried to recruit them. In terms of fundraising for nonprofit organizations, this cluster had the highest intention to donate in the future (M: 2.9). Respondents in Cluster 5 were mostly younger, well-educated men. Their educational background also suggested higher income potential in the future, making them valuable targets for fundraising campaigns.

When we compare Clusters 1 to 5 with the six non-blood donor clusters identified by Martín-Santana et al. (2020)—“Impure altruists”; “I want to, but make it easy for me”; “Free-riders”; “Reciprocal altruists”; “I can’t, because I’m scared”; and “I want to, but I can’t”—some interesting overlaps can be highlighted. On one hand, Cluster 1 (categorical non-donors; “I am strictly against donating”) is similar to their “Free-riders” cluster, where individuals were not prone to donate blood at all. This suggests that, regardless of the context, there is always a proportion of people who will never donate. On the other hand, their “I want to, but I can’t” cluster confirms that there is always a proportion of non-donors who could, in principle, be recruited (this study’s Cluster 5), subject to specific marketing actions to trigger their potentials to start donating.

Discussion

The answer to our third research question (“Should nonprofit and fundraising managers proactively try to persuade non-donors to donate money to their organization?”) is overall “Yes, but it depends on the results of each cluster studied.” Our results show that non-donors’ intention to start donating money in the future is low in all five clusters, ranging from a mean of 3.8 in Cluster 1 (categorical non-donors with nearly no willingness to donate in the future) to a mean of 2.9 in Cluster 5 (convincible non-donors with willingness that is not very pronounced, but nevertheless exists). By comparing our result regarding future willingness to donate money with results from other contexts, for example with the non-donor results of Reid and Wood (2008) in the blood donation context, it is interesting to see that our percentage of “high intenders to donate in the future” within the total sample was 13.4% (Supplement 4, future willingness to donate, item 3 “I will most likely start donating money”), compared with 38.3% for “high intenders” in a blood donation context; thus, the starting position is rather problematic.

Given the comparatively low willingness to donate in the future, it is necessary to discuss whether it makes sense to process and recruit non-donor segments. We used the logic of the somewhat old, but nevertheless helpful Boston Consulting Group portfolio matrix to assess the potential for transforming non-donors into first-time donors in the future (see Figure 2). This matrix encompasses two dimensions: financial ability (low vs. high) and attitudes toward donating (negative vs. positive). These factors were also highlighted in previous studies (Keyt et al., 2002; Sargeant et al., 2000). By examining these dimensions, nonprofit organizations can more effectively assess the likelihood of transforming non-donors into donors and customize their strategies accordingly. This study used the portfolio matrix to visualize the identified non-donor clusters and help nonprofit managers better understand and decide which cluster to invest in.

Figure 2.

Non-Donor Portfolio Matrix.

The first quadrant of the non-donor portfolio comprises categorical (C1) and unsolicited non-donors (C2) and is characterized by a rather low financial ability perceived by the individual and very negative attitudes toward donating in general. Logically, significant efforts and resources are needed to attract non-donors’ attention, with a very low success rate in recruiting them as first-time donors. We recommend a “no follow-up strategy,” although organizations should continue to track and analyze these groups and monitor whether the share of these groups increases over time. Nonprofit organizations should also explore more detailed market research to gain a better understanding of non-donors’ negative attitudes.

Quadrant 2 covers the restricted non-donors (C4) with a rather low financial ability but more positive attitudes toward donating. Here, nonprofit managers need to decide whether they think it will be possible to transform, at least partially, non-donors into donors. We are skeptical and believe that C4 should also not be processed further, as their situation is unlikely to change in the near future. Given this cluster’s relatively young age profile, one strategic option would be to increase nonprofit brand awareness. Higher brand awareness could foster positive attitudes within the non-donor group and retain their latent support until their financial circumstances improve.

Finally, the two clusters in Quadrant 3 have the potential to transform into new donors. In comparison to the other clusters, the convincible (C5) and the indifferent non-donors (C3) are characterized by a high(er) financial ability to donate money and more positive attitudes. Thus, we recommend the development of new donor-recruitment strategies and actions. Cluster 5 comprises younger, well-educated men who are particularly receptive to engagement. Research shows that factors such as age (Shehu et al., 2015), education (Rajan et al., 2009), and gender (Shehu et al., 2015) significantly influence donation behavior. A potential recruitment strategy could involve a campaign conducted through social media platforms, such as LinkedIn and Instagram, which are frequently used by non-donors in this cluster (see Supplement 5, social media channel usage among non-donors). This aligns with research suggesting that lifestyle characteristics, including social media behavior, are often regarded as effective strategies in consumer studies (Vyncke, 2002).

A different recruitment approach is required for Cluster 3 (indifferent non-donors). For example, Urselmann (2023) highlighted that those belonging to the boomer generation entering their prime fundraising age (60+ years) prefer transparent organizations and rational donor information; consequently, fundraisers should prioritize strategies that foster trust and open dialogue, ensuring that communications are tailored to address the specific concerns and preferences of this demographic segment. However, fundraisers should proceed cautiously, especially in face-to-face solicitation that may be perceived as intrusive or overwhelming (Waldner et al., 2020). Instead, a balanced approach that combines personalized outreach with digital communication channels could be effective in engaging this cluster. This could, for example, include information via tailored emails, informative webinars, or invitation-only events that create a sense of exclusivity.

In summary, future research and practitioner studies should focus on identifying concrete marketing actions to effectively engage the two promising non-donor clusters. Concurrently, our study highlights the challenge of reaching certain segments of the population that appear largely unresponsive to fundraising efforts, particularly those who are neither willing nor able to donate and those who are willing but unable to do so. Our findings on the five non-donor clusters (categorical, unsolicited, indifferent, restricted, and convincible non-donors) provide valuable insights for nonprofit research and practitioners, thereby enhancing their understanding and strategies in this domain.

Limitations and Future Research

Considering the lack of studies explicitly addressing different segments of money non-donors, this study fills the gap by providing distinct profiles of non-donors. It was long overdue to take a closer look at non-donors and challenge the assumption that non-donors constitute a large, homogeneous group. Nevertheless, this study has some limitations. First, there is the selection bias of semi-structured literature reviews. Future research could conduct a systematic review or even a meta-analysis on the topic of non-donors.

Second, although our sample is representative of Germany, the data cannot be generalized. Follow-up studies in other countries are required to confirm the cluster solution results with non-donor samples from other contexts.

Third, as mentioned earlier, another limitation is that, based on the results of the exploratory factor analysis and our internal reflections and discussions, we ultimately decided to create a fourth factor (solicitation) consisting of two items. Future non-donor studies could focus more on explanation and measurement of the selective attention of non-donors in regard to the solicitation efforts of nonprofits.

Fourth, our results show that understanding the motives of non-donors is important. We recommend focusing on strategies and recruiting actions targeting indifferent non-donors in Cluster 3 and potential donors in Cluster 5, although we were not able to provide empirical proof that these clusters are recruitable. Laboratory or field experiments could also be conducted to target recruitment efforts toward these two clusters of non-donors. Future projects could also aim to better understand the marketing resources available for recruiting non-donors in nonprofit organizations. Currently, the financial investment required to recruit a new donor ranges between 100 and 200 euros. Thus, the profitability of fundraising is threatened, as there are fewer potential donors, and recruiting and retaining them has become more expensive.

Finally, data were collected in November 2022, 9 months after Russia’s invasion of Ukraine. The war triggered a large wave of solidarity with the Ukrainian people in Germany, which led to an increase in donations. It is likely that many individuals who only donate during exceptional or disaster situations were classified as donors during the data collection period in November 2022. These individuals, who would have been included in the study as non-donors under other circumstances and who might have formed another non-donor cluster, were therefore not analyzed. Thus, additional non-donor studies are required to compare results over time.

Conclusion

For a long time, it has been widely accepted in nonprofit research and practice that non-donors are a “black box,” about whom little is known and who are considered difficult to recruit as new donors. However, the German Fundraising Association included questions on non-donors in their annual German Donation Monitor survey for the first time, which we were allowed to use for our data analysis. As a core finding, five attitudinal non-donor clusters were identified and we conclude that categorical, unsolicited, and restricted non-donors are very unlikely to donate in the future and should therefore not be targeted. In contrast, indifferent and convincible non-donors have the potential to donate in the future. Overall, our findings suggest that nonprofit organizations should analyze their own donor base from time to time to ensure that they realize their full new donor potential.

Supplemental Material

sj-docx-1-nvs-10.1177_08997640251344941 – Supplemental material for Unpacking the Non-Donor Segment: Reasons for Non-Support and Five Attitudinal Clusters Based on Data From the German Donation Monitor

Supplemental material, sj-docx-1-nvs-10.1177_08997640251344941 for Unpacking the Non-Donor Segment: Reasons for Non-Support and Five Attitudinal Clusters Based on Data From the German Donation Monitor by Silke Boenigk, Amelie Knaus, Laura Hesse, Carolin Saltzmann and Tom Neukirchen in Nonprofit and Voluntary Sector Quarterly

Footnotes

Data Availability Statement

The data of the German Donation Monitor belong to the German Fundraising Association and are not publicly available. Interested researchers can request the data from the corresponding author for scientific purposes.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

Silke Boenigk

Laura Hesse

Supplemental Material

Supplemental material for this article is available online.

Author Biographies

Silke Boenigk is a professor of nonprofit management at the Faculty of Business and Social Sciences at University of Hamburg, Germany. Her primary research focus is nonprofit leadership, donor relationship management, nonprofit marketing, and fundraising.

Amelie Knaus is a Master’s graduate from the University of Hamburg, where she already focused on the topic of non-donors. She is currently working in the fundraising and communications department of a German foundation.

Laura Hesse is junior professor for Service Management at Bauhaus University Weimar, Germany. She earned her PhD at University of Hamburg, where she joined the research project that led to this paper. Her primary research focus is on service research and the role of artificial intelligence in service interactions, particularly in addressing consumer vulnerability.

Carolin Saltzmann earned her PhD at the University of Hamburg in the field of donor relationship management and is currently head of the Security Research Department at the German Red Cross Headquarters. She conducts research in the context of civil protection, driving social, technological, and procedural innovations to address current challenges in public safety with novel approaches.

Tom Neukirchen is honorary professor and fundraising practitioner since 1997. He is member of the German Fundraising Association.

References

Andreoni

Rao

J. M.

(2011). The power of asking: How communication affects selfishness, empathy, and altruism. Journal of Public Economics, 95(7–8), 513–520.

Bekkers

Wiepking

(2011). A literature review of empirical studies of philanthropy: Eight mechanisms that drive charitable giving. Nonprofit and Voluntary Sector Quarterly, 40(5), 924–973.

Bhati

McDonnell

(2025). Message content and diversity as an effective nonprofit fundraising strategy on social media. Nonprofit and Voluntary Sector Quarterly, 54, 721–736. https://doi.org/10.1177/08997640241303920

Boenigk

Leipnitz

(2016). Acquiring potential blood donors in large cities: A preference-based donor segmentation study. Journal of Nonprofit & Public Sector Marketing, 28(4), 364–393.

Burnett

(2023). Relationship fundraising. In Breeze

Lafferty

D. D.

Wiepking

(Eds.), Fundraising reader (pp. 147–152). Routledge.

Chatzidakis

Hibbert

Winklhofer

(2016). Are consumers’ reasons for and against behaviour distinct? European Journal of Marketing, 50(1–2), 124–144.

Clifford

B. R.

Gunter

McAleer

J. L.

(1995). Television and children: Program evaluation, comprehension, and impact. Psychology Press.

Deutscher Fundraising Verband (Ed.). (2022). German donation monitor 2021 (Deutscher Spendenmonitor 2021). https://www.dfrv.de/studien-oeffentlich/#1613639891466-2a1971b0-8630

Driver

(2001). A selective review of selective attention research from the past century. British Journal of Psychology, 92(1), 53–78.

10.

Drucker

A. M.

Fleming

Chan

A.-W.

(2016). Research techniques made simple: Assessing risk of bias in systematic reviews. Journal of Investigative Dermatology, 136(11), e109–e114.

11.

Duboz

Cunéo

(2010). How barriers to blood donation differ between lapsed donors and non-donors in France. Transfusion Medicine, 20(4), 227–236.

12.

Durango-Cohen

E. J.

Torres

R. L.

Durango-Cohen

P. L.

(2013). Donor segmentation: When summary statistics don’t tell the whole story. Journal of Interactive Marketing, 27(3), 172–184.

13.

Dury

De Donder

De Witte

Buffel

Jacquet

Verté

(2015). To volunteer or not: The influence of individual characteristics, resources, and social factors on the likelihood of volunteering by older adults. Nonprofit and Voluntary Sector Quarterly, 44(6), 1107–1128.

14.

Furr

R. M.

Bacharach

V. R.

(2013). Psychometrics: An introduction (2nd ed.). Sage.

15.

Gazley

(2022). The systematic literature review: Advantages and applications in nonprofit scholarship. Voluntas, 33, 1256–1262.

16.

Guo

Saxton

G. D.

(2018). Speaking and being heard: How nonprofit advocacy organizations gain attention on social media. Nonprofit and Voluntary Sector Quarterly, 47(1), 5–26.

17.

Hair

J. F.

Black

W. C.

Babin

B. J.

Anderson

R. E.

(2018). Multivariate data analysis (8th ed.). Pearson Education.

18.

Harrington

Sweeney

M. R.

Bailie

Morris

Kennedy

Boilson

O’Riordan

Staines

(2007). What would encourage blood donation in Ireland? Vox Sanguinis, 92(4), 361–367.

19.

Harrison

Thornton

(2022). Further evidence on competition in nonprofit donor markets. Nonprofit and Voluntary Sector Quarterly, 51(4), 713–735.

20.

Jedicke

E. M.

Olk

Zogaj

Tscheulin

D. K.

Lindenmeier

(2024). Designing effective volunteering appeals: Results of choice-based conjoint and latent class segmentation analyses. Nonprofit and Voluntary Sector Quarterly, 53, 1537–1563. https://doi.org/10.1177/08997640241227166

21.

Keyt

Yavas

Riecken

(2002). Comparing donor segments to a cause-based charity: The case of the American Lung Association. Journal of Nonprofit & Public Sector Marketing, 10(2), 117–134.

22.

Kolhede

Gomez-Arias

J. T.

(2022). Segmentation of individual donors to charitable organizations. International Review on Public and Nonprofit Marketing, 19(2), 333–365.

23.

Lange

Q. E.

(2024). What are the factors that motivate nondonors? Understanding the reasons to not partake in philanthropy (HAL Id: hal-04727160). HAL Open Science Publication. https://univ-cotedazur.hal.science/hal-04727160/

24.

Lockstone-Binney

Holmes

Meijs

L. C. P. M.

Oppenheimer

Haski-Leventhal

Taplin

(2022). Growing the volunteer pool: Identifying non-volunteers most likely to volunteer. VOLUNTAS: International Journal of Voluntary and Nonprofit Organizations, 33(4), 777–794.

25.

Lwin

Phau

Lim

(2014). An investigation of the characteristics of Australian charitable donors. Journal of Nonprofit & Public Sector Marketing, 26(4), 372–389.

26.

Martín-Santana

J. D.

Beerli-Palacio

Romero-Domínguez

(2020). Recruitment strategies: Non-donor segmentation based on intrinsic and extrinsic stimuli. Vox Sanguinis, 115(1), 47–59.

27.

Masser

B. M.

France

C. R.

Himawan

L. K.

Hyde

M. K.

Smith

(2016). The impact of the context and recruitment materials on nondonors’ willingness to donate blood. Transfusion, 56(12), 2995–3003.

28.

Niebuur

Liefbroer

A. C.

Steverink

Smidt

(2019). The Dutch comparative scale assessing volunteer motivations among volunteers and non-volunteers: An adaption of the volunteer functions inventory. International Journal of Environmental Research and Public Health, 16(17), 3106.

29.

Piersma

T. W.

Bekkers

Klinkenberg

E. F.

De Kort

W. L.

Merz

E. M.

(2017). Individual, contextual and network characteristics of blood donors and non-donors: A systematic review of recent literature. Blood Transfusion, 15(5), 382–397.

30.

Punj

Stewart

D. W.

(1983). Cluster analysis in marketing research: Review and suggestions for application. Journal of Marketing Research, 20(2), 134–148.

31.

Rajan

S. S.

Pink

G. H.

Dow

W. H.

(2009). Sociodemographic and personality characteristics of Canadian donors contributing to international charity. Nonprofit and Voluntary Sector Quarterly, 38(3), 413–440.

32.

Reed

P. B.

Selbee

L. K.

(2000). Distinguishing characteristics of active volunteers in Canada. Nonprofit and Voluntary Sector Quarterly, 29(4), 571–592.

33.

Reid

Wood

(2008). An investigation into blood donation intentions among non-donors. International Journal of Nonprofit and Voluntary Sector Marketing, 13(1), 31–43.

34.

Reutterer

Dan

(2022). Cluster analysis in marketing research. In Homburg

Klarmann

Vomberg

(Eds.), Handbook of market research (pp. 221–250). Springer.

35.

Riecken

Yavas

(2005). The attitudes of donors and non-donors to the March of Dimes Charity in the United States: A case study in non-profit marketing. International Journal of Management, 22(4), 572–581.

36.

Rooney

Ottoni-Wilhelm

Wang

(2020). Dynamics of American giving: Descriptive evidence. Nonprofit and Voluntary Sector Quarterly, 50(4), 729–752.

37.

Rupp

Kern

Helmig

(2014). Segmenting nonprofit stakeholders to enable successful relationship marketing: A review. International Journal of Nonprofit and Voluntary Sector Marketing, 19(2), 76–91.

38.

Saltzmann

Boenigk

(2022). On consciousness of the decision to discontinue blood donation: Intention to return and effective recovery activities. Transfusion Medicine, 32(3), 193–209.

39.

Sargeant

Ford

West

(2000). Widening the appeal of charity. International Journal of Nonprofit and Voluntary Sector Marketing, 5, 318–332.

40.

Sauer

P. C.

Seuring

(2023). How to conduct systematic literature reviews in management research: A guide in 6 steps and 14 decisions. Review of Managerial Science, 17(5), 1899–1933.

41.

Schlegelmilch

B. B.

Diamantopoulos

Love

(1997). Characteristics affecting charitable donations: Empirical evidence from Britain. Journal of Marketing Practice: Applied Marketing Science, 3(1), 14–28.

42.

Schubert

Boenigk

(2019). The nonprofit starvation cycle: Empirical evidence from a German context. Nonprofit and Voluntary Sector Quarterly, 48(5), 467–491.

43.

Schweidel

Knox

(2013). Incorporating direct marketing activity into latent attrition models. Marketing Science, 32(3), 471–487.

44.

Shehu

Langmaack

A.-C.

Felchle

Clement

(2015). Profiling donors of blood, money, and time. Nonprofit Management and Leadership, 25(3), 269–295.

45.

Snyder

(2019). Literature review as a research methodology: An overview and guidelines. Journal of Business Research, 104, 333–339.

46.

Sundeen

R. A.

Raskoff

S. A.

Garcia

M. C.

(2007). Differences in perceived barriers in volunteering to formal organizations: Lack of time versus lack of interest. Nonprofit Management and Leadership, 17(3), 279–300.

47.

Taniguchi

(2010). Who are volunteers in Japan? Nonprofit and Voluntary Sector Quarterly, 39(1), 161–179.

48.

Urselmann

(2023). Fundraising. Professional resource acquisition for nonprofit organizations (8th ed.). Springer Gabler (in German).

49.

Vaske

J. J.

Beaman

Sponarski

C. C.

(2017). Rethinking internal consistency in Cronbach’s alpha. Leisure Sciences, 39(2), 163–173.

50.

Veldhuizen

Folléa

De Kort

(2013). Donor cycle and donor segmentation: New tools for improving blood donor management. Vox Sanguinis, 105(1), 28–37.

51.

Vyncke

(2002). Lifestyle segmentation. From attitudes, interests and opinions, to values, aesthetic styles, life visions and media preferences. European Journal of Communication, 17(4), 445–463.

52.

Waldner

C. J.

Willems

Ehmann

Gies

(2020). The impact of face-to-face street fundraising on organizational reputation. International Journal of Nonprofit and Voluntary Sector Marketing, 25(4), Article e1672. https://doi.org/10.1002/nvsm.1672

53.

Willems

Dury

(2017). Reasons for not volunteering: Overcoming boundaries to attract volunteers. The Service Industries Journal, 37(11–12), 726–745.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.06 MB