The objective of this paper is to conduct an exploratory analysis of the driving factors that influence musicians in their use of a Spanish reward-based crowdfunding platform.
We developed a questionnaire that was distributed online, obtaining 78 observations from music promoters who used the Verkami platform as a means of financing their projects. Based on the data obtained, partial least squares with the extended Unified Theory of Acceptance and Use of Technology 2 (UTAUT2) was used to analyze the musicians’ experiences with the aim of determining the factors that could influence the behavioral intention (BI).
The main findings suggest that performance expectancy, social influence and hedonic motivation of musicians who use the platform have a significant impact on BI. Secondary findings indicate that the perception of quality and security are key drivers of the trust of musicians in the platform.
The data analyzed come from a specific reward-based crowdfunding platform, which prevents the results from being generalized to all musicians who have obtained funding through this method. Therefore, it would be valuable for future studies to evaluate the proposed model using data from other platforms and from different countries, which would increase the external validity of the findings. On the other hand, this study could be extended to platforms that develop other kinds of crowdfunding as potential research avenues (donation, lending and equity). In this way, it would be possible to compare the variables that are important for either modality.
This research can help to better understand musicians' experiences so that reward-based crowdfunding platforms can improve their dynamics and ultimately increase their usability in the musical sector. We believe that our research makes an important contribution to the knowledge of the experiences in using crowdfunding platforms from the perspective of the entrepreneurs of music projects. The results obtained present new findings on previous research, providing useful knowledge for users and managers of reward-based crowdfunding platforms. Although the study focuses on musicians, the identified factors may also be relevant to other creative sectors that use reward-based crowdfunding. Thus, this work expands the understanding of financial technology adoption in cultural contexts, providing empirical evidence useful for both scholars and platform managers.
Analyzing the factors that may influence musicians' use of rewards crowdfunding platforms has important benefits for both platform promoters and entrepreneurs who want to launch musical projects.
This paper will focus on reward-based crowdfunding, with a particular emphasis on the cultural sector, and more specifically, on music. This research aims to provide a deeper understanding of musicians' experiences, allowing platforms to enhance their functionality and, ultimately, improve their effectiveness in the music industry.
1. Introduction
Crowdfunding allows for-profit, creative, and cultural entrepreneurs to fund their projects through a great amount of little contributions from a big number of people, all through online platforms, without financial intermediaries (Agrawal et al., 2015; Mollick, 2018). Moreover, the last decade has seen crowdfunding become an increasingly relevant mechanism to support cultural productions, in parallel to a reduction in public financing and private investments in general (Bannerman, 2013; Boeuf et al., 2014; Migliavacca et al., 2016).
This alternative funding source assists musicians in raising capital and promoting their work, which is particularly beneficial for young musicians entering the labor market. In Spain, around 50 crowdfunding platforms have been established (González and Ramos, 2022), though only a few focus on music sector. Verkami stands out for its high success rate in music-related projects (Montero-Benavides, 2021).
From a musical perspective, creativity is crucial, but most creators lack entrepreneurial skills and attitudes when developing projects (Szostak and Sułkowski, 2021). Reward-based crowdfunding is thus highly valuable in cultural entrepreneurship, supporting the development of entrepreneurial initiatives (Mollick and Kuppuswamy, 2014). In reward-based crowdfunding, funders are promised goods or services in the future in exchange for their financial support, under a contract that does not penalize the creator for non-delivery (Gutiérrez and Sáez, 2018). This novel and increasingly popular financing method was originally conceived for cultural projects (Dalla Chiesa and Handke, 2020).
In addition to reward-based crowdfunding, there are other types of crowdfunding: donation, equity, or lending. In donation-based crowdfunding a type of public giving – contributors invest capital because they value intrinsic motivations and non-financial benefits (Belleflamme and Lambert, 2016). Equity crowdfunding is a financing model where entrepreneurs publicly offer a specific amount of equity or bond-like shares in their company online, aiming to attract a broad group of investors (Ahlers et al., 2015). The lending model facilitates credit agreements without intermediaries—namely, without the involvement of banks (Cholakova and Clarysse, 2014).
Crowdfunding sector in Latin America has experienced significant growth since 2009, with Brazil, Mexico, and Chile emerging as the countries with the most advanced ecosystems in the region. According to Rentería (2016), interviews conducted with participants from various platforms have yielded relevant information on the business models employed and the strategies implemented to seize opportunities and address the specific challenges of crowdfunding in Latin America.
In recent times, collective financing in the music sector has grown remarkably in the region. This phenomenon is reflected in the emergence of specialized crowdfunding platforms and the development of innovative models that respond to new local economic, cultural, and technological realities. The main driving force behind this progress has been the need for independent musicians to find alternative sources of funding outside traditional structures. A revealing example is the instance of Colombia, where crowdfunding has become an increasingly utilized tool for financing artistic and cultural projects (M2Crowd, n.d.).
A notable case of innovation is InCresc.com, a Venezuela-based platform launched in July 2022 (InCresc, 2023). Designed exclusively for the music sector, it enables users worldwide to support emerging musicians through micro-contributions starting at $5 (US dollars). By facilitating global access to funding, InCresc contributes to the artistic and professional development of young talent (Ecos Digitales, 2022).
Complementing this initiative, Casa Yask’in (2024), musical production company, offers an updated overview of available crowdfunding platforms and presents success stories from countries such as Mexico, Brazil, and Argentina.
The period between 2022 and 2025 has seen substantial advances in the Latin American music crowdfunding landscape. The emergence of regional platforms like InCresc, the increasing availability of specialized Spanish-language resources, the widespread adoption of reward-based models, and the integration of educational content reflect a broader shift toward sustainable, community-based cultural financing models tailored to regional realities.
There are many articles on the success factors of campaigns that have used crowdfunding platforms for their funding, but there is hardly any research on the use and experience determinants of the users who choose this alternative to finance their projects (Cecere et al., 2017). The topic of arts and culture in entrepreneurship is highly relevant to academics in many disciplines and to policy makers interested in the development of the creative class, and more specifically crowdfunding (Noonan, 2021). It is consequently important to know the motivations that lead cultural creators to use these platforms, and more specifically in the music sector.
The objective of this paper is to explore the factors motivating musicians to use a Spanish reward-based crowdfunding platform. Our research focuses on music projects funded through Verkami, testing hypotheses derived from the extended Unified Theory of Acceptance and Use of Technology 2 (UTAUT2) model. Additionally, we incorporate the constructs of Innovativeness (INT) and perceived trust (PT), as supported by other theories. Data were collected via an online survey of 78 musicians who used Verkami, analyzed using the partial least squares (PLS) methodology.
The remainder of the paper proceeds as follows. In the next section, we provide the literature review. We then propose our working hypotheses and describe the research methodology. Finally, we present the data analysis, results, conclusions, practical implications, limitations and future research lines.
2. Theory and hypotheses
Many authors have applied the Unified Theory of Acceptance and Use of Technology (UTAUT) model for various purposes (Escobar-Rodríguez and Carvajal-Trujillo, 2014; Raza et al., 2019; Thaker et al., 2019), including empirical studies of crowdfunding (Abdullah and Bakri, 2020; Darmansyah et al., 2020; Islam and Khan, 2021; Khizar and Siddiqui, 2021; Kim and Jeon, 2017; Li et al., 2018; Moon and Hwang, 2018; Sulaeman and Ninglasari, 2020).
Most studies have used the PLS methodology to determine which factors influence entrepreneurs’ intention to use new technologies, on which crowdfunding platforms depend.
This study applies the extended Unified Theory of Acceptance and Use of Technology (UTAUT2), which integrates new constructs and relationships to explain consumer acceptance and use of information technology (Venkatesh et al., 2012). We use UTAUT2 to examine how music project promoters adopt and use crowdfunding platforms for fundraising.
The original UTAUT was a synthesis of eight different theoretical models drawn from sociological and psychological theories used in the literature to explain past, present, and future intention to use a technology in organizational contexts (Venkatesh et al., 2003). UTAUT2, considers 7 constructs, 4 from UTAUT (performance expectancy (PE), effort expectancy (EE), social influence (SI) and facilitating conditions (FCs)) and incorporates three new constructs: hedonic motivation (HM), price value and habit. Given the characteristics of our study, we have only included HM from UTAUT2. HM has been included as a key predictor in much consumer behavior research (Holbrook and Hirschman, 1982) and prior information systems research in the consumer technology use context (Brown and Venkatesh, 2005).
Existing literature suggests that UTAUT effectively predicts crowdfunding behaviors (Gür and Özdoğan, 2021; Kim and Hall, 2020; Li et al., 2018; Moon and Hwang, 2018). The UTAUT 2 model posits that usage behavior—understood as the frequency with which an individual engages with information technology—is jointly influenced by behavioral intention (BI), which refers to the extent to which a person has consciously formulated plans to carry out or avoid a specific future action (Ramírez-Correa et al., 2019; Yuliani et al., 2024). In UTAUT2. BI is justified as a multifactorial outcome, where rational aspects (such as PE and EE), social aspects (SI), and emotional aspects (HM) converge (Venkatesh et al., 2012).
In addition, we include INT and PT, as these constructs are relevant for predicting music promoters’ behavioral intentions regarding crowdfunding platforms.
Likewise, in our study, trust in the platforms is a key aspect to attract project developers, therefore, following previous studies such as Kim et al. (2008), we have analyzed those antecedents of trust that are associated with user perception such as perceived privacy (PP), security and the quality of the information offered on the platform.
2.1 Hypotheses development for the proposed model
The constructs used in this work are defined below, and their inclusion in previous studies is analyzed. The relationships between them are then specified, and the hypotheses tested in this paper are formulated.
PE, as defined by Venkatesh et al. (2003), refers to the belief that using a system will enhance job performance. This construct has been shown to be a strong predictor of users' intention to use a system (Escobar-Rodríguez and Carvajal-Trujillo, 2014; San Martín and Herrero, 2012; Venkatesh et al., 2012), and several studies confirm that it positively influences both BI and actual use (Awuah, 2012; Lung et al., 2008). Specifically, PE plays a crucial role in influencing the intention to use financial technologies (Ahmad et al., 2014; Darmansyah et al., 2020; Raza et al., 2019). In the context of crowdfunding, PE also plays a significant role in the creation of projects (Pangaribuan and Wulandar, 2018). However, some studies (Moon and Hwang, 2018) suggest that PE may not always exert an effect, indicating the complexity of its influence on user behavior.
In relation to the above, the following hypothesis is tested:
The performance expectancy in the use of crowdfunding platforms positively affects the behavioral intention.
Building upon this, EE refers to the perceived ease of using technology (San Martín and Herrero, 2012), which similarly influences users' intention to adopt it (Aggelidis and Chatzoglou, 2009). In the crowdfunding domain, EE also affects users' intention to fund projects (Moon and Hwang, 2018) and is a key factor in the development of crowdfunding projects (Pangaribuan and Wulandar, 2018). For instance, when individuals perceive financial technology as user-friendly, it enhances their willingness to engage (Huei et al., 2018). The expectation effort of technology in crowdfunding platforms allows project promoters and their donors to connect online without too much difficulty and intervention, making both technology and crowdfunding knowledge useful to help certain projects get funded (W. B. W. Bank, 2006) However, as noted in some studies (Khizar and Siddiqui, 2021; Lung et al., 2008), the impact of EE on BI may not always be straightforward.
Based on the foregoing, the following hypothesis is proposed:
The effort expectancy in the use of crowdfunding platforms positively affects the behavioral intention.
Furthermore, SI refers to the perception that family and friends believe one should use a particular technology (Brown and Venkatesh, 2005; Venkatesh et al., 2003). It significantly impacts behavior and decision-making (Ajzen, 1991; Taylor and Todd, 1995), and several studies confirm its influence on the intention to use technology (Darmansyah et al., 2020; Khizar and Siddiqui, 2021; Li et al., 2018; Lung et al., 2008; Thaker et al., 2019)). However, some research suggests that SI may not always affect technology adoption (Raza et al., 2019; San Martín and Herrero, 2012). The support and guidance from those around an entrepreneur can be crucial in their decision to adopt new technologies (Alalwan et al., 2017).
Following the above arguments, the following hypothesis is tested:
The social influence regarding the use of crowdfunding platforms positively affects the behavioral intention.
Another critical construct is FC, which refer to the perception of available organizational and technical support that enables users to comfortably use a system (Venkatesh et al., 2003). While some studies (Tariqui and Tahir, 2019) have found a significant relationship between FC and users' intention to use crowdfunding platforms, other research (Moon and Hwang, 2018) suggests that the platform’s infrastructure, such as customer service, does not always have a substantial impact on users' behavioral intentions. In this study, FCs are defined as the users' perception of having adequate resources and knowledge to effectively use crowdfunding platforms.
Based on the above, the following hypothesis is proposed:
The facilitating conditions perceived in the use of crowdfunding platforms positively affect the behavioral intention.
In addition to these constructs, the HM introduced in UTAUT2 theory is also of importance. HM reflects the enjoyment or intrinsic satisfaction derived from using technology. Schulz et al. (2015) examined the hedonic value of crowdfunding projects and found that emotionality, amusement, and engagement increase the likelihood of funding success. HM is also important in understanding the BI to use crowdfunding for start-up founders (Khizar and Siddiqui, 2021). Kim et al. (2020) recommend that founders create fun, engaging content to build trust and encourage participation from funders.
In this section, in view of the above, the following hypothesis is tested:
The hedonic motivation experienced in the use of crowdfunding platforms positively affects the behavioral intention.
Alongside these, INT is defined as an individual’s willingness to try new technology (Agarwal and Prasad, 1998). Xu and Gupta (2009) describe personal INT as a tendency to take risks, which varies among individuals. Rogers (1995) noted that innovators actively seek information and rely less on others' opinions. While personal INT is commonly studied in e-commerce (Citrin et al., 2000; Escobar-Rodríguez and Carvajal-Trujillo, 2014; San Martín and Herrero, 2012; Lu et al., 2011), few studies focus on its influence on crowdfunding platforms and BI. Gür and Özdoğan (2021) explored this in the context of science crowdfunding, considering innovation as the ability of the scientist to create novel solutions to challenges, understanding crowdfunding as an innovative behavior to face traditional funding challenges.
Based on the arguments, the following hypothesis is proposed:
The personal innovativeness in the use of crowdfunding platforms positively affects the behavioral intention.
In addition, PT is an essential component of online financial transactions (Gefen, 2000), and its role in crowdfunding platforms has been highlighted in several studies. PT reflects users' belief in the reliability and integrity of crowdfunding platforms (Islam and Khan, 2021), and it is a crucial determinant of users' willingness to participate (Moon and Hwang, 2018). Gerber and Hui (2013) highlight that a lack of trust can hinder participation, and the level of trust is crucial for the success of crowdfunding campaigns.
Based on the above, it proposes the following hypothesis:
Trust in the use of crowdfunding platforms positively affects the behavioral intention.
Moreover, the perception of Information Quality (IQ), including transparency, has been shown to foster trust, further promoting participation in crowdfunding (Gefen, 2000). Consequently, IQ is not expected to influence trust in the platform as a whole but rather trust in individual campaign creators and the particular initiatives they propose (Moysidou and Hausberg, 2020). In crowdfunding settings, ongoing engagement and prior interactions with campaign creators are typically lacking; therefore, cognition-based trust relies on well-founded reasoning and the credibility of the information provided (McAllister, 1995). The platform should prioritize attentiveness to its users and provide them with appropriate information, thereby enhancing investors' perceptions of the quality of service delivered (Ferrer et al., 2023).
Based on the above, the following hypothesis will be tested:
Users' perception on the quality information of platforms has a positive effect on trust.
Further exploring the relationship between trust and user engagement, Perceived Security (PS) is another essential factor. Flavián and Guinalíu (2006) define PS as consumers’ subjective belief that their personal and financial data are not visible, stored, or managed by unauthorized third parties, and that this perception is shaped by technical aspects such as integrity, confidentiality, and authentication. Additionally, these authors distinguish between privacy and security: privacy is linked to legal and ethical issues regarding the use of information, while security is associated with technical guarantees. Consumers tend to perceive both concepts together, and this joint perception directly influences their trust in a website. The study concludes that a higher perception of security in the handling of personal data increases user trust, and that this trust is key to fostering consumer loyalty to the website.
According to Bonsón et al. (2015), PS depends on factors such as website expertise, third-party guarantees, security and privacy policies, and general concerns about privacy on the internet. Their study identifies IQ and Security as key indicators of trust in online transactions, particularly in the context of online travel purchases. Similarly, Ray et al. (2011) and Lee et al. (2015) have emphasized how PP and Security influence users' trust in online platforms, with higher levels of PS leading to increased participation in crowdfunding. Trust, specifically related to security, is a critical factor in fostering users' willingness to engage and invest in crowdfunding projects.
Based on the above, the following hypothesis is proposed:
Users' perception on the security protection of platforms has a positive effect on trust.
Lastly, the concern for PP is a critical issue in the context of trust in digital transactions. PP refers to an individual’s sense that external parties have restricted access to their personal information (Dinev et al., 2013). Turan (2015) highlights the dual trust dilemma in crowdfunding, where PP concerns arise both for investors, who trust entrepreneurs with their money, and for entrepreneurs, who share their ideas with investors. Doce and Selis (2020) found that users with negative past experiences in disclosing information are more concerned about privacy and perceive higher risks, leading them to accept regulatory limits on crowdfunding platforms. PP is a crucial factor influencing consumer intention to use such platforms (Escobar-Rodríguez and Carvajal-Trujillo, 2014), and trust is strongly linked to the quality of information and PP measures. Kim et al. (2011) also emphasize that privacy impacts trust. For crowdfunding users, privacy is a major concern, as it poses risks for entrepreneurs (Jaziri and Miralam, 2019). High concerns about privacy violations reduce willingness to use crowdfunding platforms (Rose et al., 1999; Jaziri and Miralam, 2019).
In relation to the above, the following hypothesis will be tested:
Users' perception on the perceived privacy of platforms has a positive effect on trust.
Table 1 collects the items for this study and the supporting literature to measure each construct.
Construct measurement
| Construct | Item | Authors |
|---|---|---|
| PE: performance expectancy | PE1: I find crowdfunding platforms very useful in the financing process | San Martín and Herrero (2012), Venkatesh et al. (2012) |
| PE2: Using crowdfunding platforms increases my chances of achieving financing | ||
| PE3: Using crowdfunding platforms helps me accomplish things more quickly in the financing process | ||
| EE: effort expectancy | EE1: Learning how to use crowdfunding platforms is easy for me | Venkatesh et al. (2012) |
| EE2: My interaction with crowdfunding platforms is clear and understandable | ||
| EE3: I find crowdfunding platforms easy to use | ||
| EE4: It is easy for me to become skillful at using crowdfunding platforms | ||
| SI: social influence | SI1: People who are important to me think that I should use crowdfunding platforms | Venkatesh et al. (2012) |
| SI2: People who influence my behavior think that I should use crowdfunding platforms | ||
| SI3: People whose opinions that I value prefer that I use crowdfunding platforms | ||
| FC: facilitating conditions | FC1: I have the resources necessary to use crowdfunding platforms | San Martín and Herrero (2012), Venkatesh et al. (2012) |
| FC2: I have the knowledge necessary to use crowdfunding platforms | ||
| FC3: I feel comfortable using crowdfunding platforms | ||
| HM: hedonic motivation | HM1: Using crowdfunding platforms is fun | Venkatesh et al. (2012) |
| HM2: Using crowdfunding platforms is enjoyable | ||
| HM3: Using crowdfunding platforms is very entertaining | ||
| INT: innovativeness | INT1: If I heard about a new information technology, I would look for ways to experiment with it | Herrero and Rodríguez Del Bosque (2008) |
| INT2: Among my peers I am usually the first to explore new information technologies | ||
| INT3: I like to experiment with new information technologies | ||
| PT: trust | PT1: Crowdfunding platforms have integrity | Kim et al. (2011) |
| PT2: Crowdfunding platforms are reliable | ||
| IQ: information quality | IQ1: Crowdfunding platforms provide accurate information I need in the financing process | Kuan et al. (2008) |
| IQ2: Crowdfunding platforms provide sufficient information needed in the financing process | ||
| IQ3: The information provided by crowdfunding platforms is helpful to me in the financing process | ||
| IQ4: The information in crowdfunding platforms is clear to me | ||
| PS: perceived security | PS1: Crowdfunding platforms implement security measures to protect users | Kim et al. (2008) |
| PS2: Crowdfunding platforms usually ensure that transactional information is protected from accidentally being altered or destroyed during a transmission on the internet | ||
| PP: perceived privacy | PP1: I am concerned that the crowdfunding platforms collect too much personal information from me | Kim et al. (2008) |
| PP2: I am concerned that the crowdfunding platforms will use my personal information for other purposes without my authorization | ||
| PP3: I am concerned that the crowdfunding platforms will share my personal information with other entities without my authorization | ||
| PP4: I am concerned that the unauthorized persons (i.e. hackers) have access to my personal information | ||
| PP5: I am concerned that the crowdfunding platforms will sell my personal information to others without my permission | ||
| BI: behavioral intention | BI1: I intend to continue using crowdfunding platforms to finance my projects in the future | San Martín and Herrero (2012), Venkatesh et al. (2012) |
| BI2: I will always try to use crowdfunding platforms to finance my projects/ | ||
| BI3: I plan to continue to use crowdfunding platforms frequently to finance my projects |
Figure 1 shows the model according with the research conceptual framework.
The flowchart starts with a circle labeled “Behavioral Intention (B I)” positioned at the top center. Seven individual arrows point to “Behavioral Intention (B I)” from circles arranged in a semicircular pattern below the central circle. The details about the circles and the arrows pointing from each are as follows: The arrow labeled “H 1” points from the circle labeled “Performance Expectancy (PE).” The arrow labeled “H 2” points from “Effort Expectancy (E E).” The arrow labeled “H 3” points from “Social Influence (S I).” The arrow labeled “H 4” points from “Facilitating Conditions (F C).” The arrow labeled “H 5” points from “Hedonic Motivation (H M).” The arrow labeled “H 7” points from “Trust (P T).” The arrow labeled “H 6” points from “Innovativeness (I N T).” Three circles are arranged in a horizontal series below “Trust (P T).” Each circle has an arrow pointing to “Trust (P T).” From left to right, the details for each circle and the labeled arrows are as follows: The arrow labeled “H 8” points from “Information Quality (I Q).” The arrow labeled “H 9” points from “Perceived Security (P S).” The arrow labeled “H 10” points from “Perceived Privacy (P P).”Research conceptual framework. Source(s): Authors’ own elaboration
The flowchart starts with a circle labeled “Behavioral Intention (B I)” positioned at the top center. Seven individual arrows point to “Behavioral Intention (B I)” from circles arranged in a semicircular pattern below the central circle. The details about the circles and the arrows pointing from each are as follows: The arrow labeled “H 1” points from the circle labeled “Performance Expectancy (PE).” The arrow labeled “H 2” points from “Effort Expectancy (E E).” The arrow labeled “H 3” points from “Social Influence (S I).” The arrow labeled “H 4” points from “Facilitating Conditions (F C).” The arrow labeled “H 5” points from “Hedonic Motivation (H M).” The arrow labeled “H 7” points from “Trust (P T).” The arrow labeled “H 6” points from “Innovativeness (I N T).” Three circles are arranged in a horizontal series below “Trust (P T).” Each circle has an arrow pointing to “Trust (P T).” From left to right, the details for each circle and the labeled arrows are as follows: The arrow labeled “H 8” points from “Information Quality (I Q).” The arrow labeled “H 9” points from “Perceived Security (P S).” The arrow labeled “H 10” points from “Perceived Privacy (P P).”Research conceptual framework. Source(s): Authors’ own elaboration
3. Methodology
3.1 Data collection, sample, and measures
To develop this exploratory study, we conducted a user survey on the Verkami platform, designing the questionnaire using constructs and items from UTAUT2-related literature. Verkami was selected due to its prominence in Spain’s cultural sector, with music representing the largest share of its fundraising activity (28%) (Figure 2).
The projects are listed vertically on the left (from top to bottom) as follows: “Publications,” “Games,” “Movie,” “Comic,” “Science and Technology,” “Art,” “Music,” “Photography,” “Design,” “Community,” “Performing Arts,” “Feeding.” The data from the bars is as follows: Publications: 24.1 percent. Games: 5.6 percent. Movie: 17.2 percent. Comic: 3.7 percent. Science and Technology: 1.1 percent. Art: 2.6 percent. Music: 28.0 percent. Photography: 3.5 percent. Design: 1.7 percent. Community: 5.5 percent. Performing Arts: 5.3 percent. Feeding: 1.7 percent. The bar representing “Music” is distinctly highlighted.Percentage of projects by category (2010–2024). Source(s): Authors’ own elaboration
The projects are listed vertically on the left (from top to bottom) as follows: “Publications,” “Games,” “Movie,” “Comic,” “Science and Technology,” “Art,” “Music,” “Photography,” “Design,” “Community,” “Performing Arts,” “Feeding.” The data from the bars is as follows: Publications: 24.1 percent. Games: 5.6 percent. Movie: 17.2 percent. Comic: 3.7 percent. Science and Technology: 1.1 percent. Art: 2.6 percent. Music: 28.0 percent. Photography: 3.5 percent. Design: 1.7 percent. Community: 5.5 percent. Performing Arts: 5.3 percent. Feeding: 1.7 percent. The bar representing “Music” is distinctly highlighted.Percentage of projects by category (2010–2024). Source(s): Authors’ own elaboration
All items were adapted to the context of crowdfunding platform use and acceptance. The questionnaire consisted of 11 constructs and 35 items, with responses measured on a seven-point Likert scale (1 = “strongly disagree” to 7 = “strongly agree”).
The validation of the questionnaire was initially carried out by several experts from the musical world, who had already participated in this type of funding and were familiar with the subject matter of the study. This validation led to a better wording or clarification in some of the questions.
The survey was distributed online via Google Forms to music project promoters on Verkami, who were randomly selected and contacted through Facebook. A total of 78 responses were collected: 78% male and 22% female. Age distribution: 39% (26–35 years), 36% (36–45 years), and 25% (under 26 or over 45 years). 75% had university education. Regarding platform use, 74% had used Verkami once, 21% twice, and 5% three or four times. Notably, 99% had not used any other similar platform.
3.2 Statistical method
Given the proposed research model and formulated hypotheses, this study employs the PLS method for data analysis. The selection of this technique is mainly justified by the nature of the indicators, constructs, and relationships established within the research framework. According to Hair et al. (2011), the PLS method has been widely employed in studies focused on predicting dependent constructs and analyzing complex interrelationships among latent variables.
PLS is particularly suitable for studies with limited sample sizes and complex models. This method does not require large samples for robust estimation and is frequently used in research with sample sizes ranging from 50 to 100 cases, especially in early-stage or exploratory research (Hair et al., 2011; Henseler et al., 2015). The “10-times rule” suggests that the minimum sample size should be ten times the maximum number of structural paths directed at a single construct in the model (Hair et al., 2011); with 78 cases, the present study meets or closely approaches this guideline, given the model’s complexity.
Moreover, the sample consists exclusively of musicians who have used the Verkami platform, representing a relevant and specialized segment. The high level of homogeneity and relevance of respondents increases the validity of insights derived from a smaller sample (Gefen et al., 2000).
4. Analysis of the measurement model and results
4.1 Measurement model assessment
The first stage is to determine the constructs' convergent and discriminate validity, as well as the individual reliability of each item.
The convergent validity of each construct is acceptable for a loading higher than 0.707. The loadings, or correlations between an item and its associated construct, indicate the individual reliability of each item. The loadings for significant each item are described in Table 2. Item EE3 has been eliminated as it has a lower value than the one considered acceptable. Item FC1 is retained to maintain the validity of the scale and because it is close to the required value.
Measurement model results
| Scale items | Standardized loadings | Cronbach´s alpha | Composite reliability | rho_A | AVE |
|---|---|---|---|---|---|
| Behavioral intention (BI) | 0.903 | 0.939 | 0.908 | 0.838 | |
| BI1 | 0.918 | ||||
| BI2 | 0.894 | ||||
| BI3 | 0.933 | ||||
| Effort expectancy (EE) | 0.806 | 0.886 | 0.824 | 0.724 | |
| EE1 | 0.913 | ||||
| EE2 | 0.884 | ||||
| EE4 | 0.746 | ||||
| Facilitating conditions (FC) | 0.716 | 0.822 | 0.877 | 0.611 | |
| FC1 | 0.640 | ||||
| FC2 | 0.795 | ||||
| FC3 | 0.890 | ||||
| Hedonic motivation (HM) | 0.883 | 0.928 | 0.886 | 0.811 | |
| HM1 | 0.926 | ||||
| HM2 | 0.901 | ||||
| HM3 | 0.874 | ||||
| Innovativeness (INT) | 0.912 | 0.944 | 0.935 | 0.849 | |
| INT1 | 0.924 | ||||
| INT2 | 0.920 | ||||
| INT3 | 0.920 | ||||
| Information quality (IQ) | 0.924 | 0.946 | 0.938 | 0.815 | |
| IQ1 | 0.905 | ||||
| IQ2 | 0.949 | ||||
| IQ3 | 0.863 | ||||
| IQ4 | 0.892 | ||||
| Performance expectancy (PE) | 0.902 | 0.939 | 0.904 | 0.837 | |
| PE1 | 0.916 | ||||
| PE2 | 0.929 | ||||
| PE3 | 0.899 | ||||
| Perceived privacy (PP) | 0.953 | 0.963 | 0.994 | 0.839 | |
| PP1 | 0.833 | ||||
| PP2 | 0.938 | ||||
| PP3 | 0.949 | ||||
| PP4 | 0.918 | ||||
| PP5 | 0.938 | ||||
| Perceived security (PS) | 0.856 | 0.933 | 0.876 | 0.874 | |
| PS1 | 0.923 | ||||
| PS2 | 0.947 | ||||
| Social influence (SI) | 0.859 | 0.913 | 0.896 | 0.777 | |
| SI1 | 0.869 | ||||
| SI2 | 0.835 | ||||
| SI3 | 0.938 | ||||
| Trust (PT) | 0.777 | 0.900 | 0.780 | 0.817 | |
| PT1 | 0.911 | ||||
| PT2 | 0.897 |
The Cronbach coefficient alpha levels are all above 0.70, which is considered acceptable for confirmatory study (Churchill, 1979). The composite reliabilities and rho_A are over the minimum acceptable limit of 0.70 (Gefen et al., 2000; Nunnally, 1978).
The common variance between the indicators and their construct is referred as the convergent validity. The average variance extracted (AVE) is used to assess validity, and the acceptable level should be more than 0.50 (Fornell and Larcker, 1981). The AVE scores obtained for each of the eleven constructs used exceed the minimum value desired in all cases.
Discriminant validity was determined by comparing an individual construct with all other constructs. The presence of discriminant validity indicates that a construct is unique and measures phenomena not represented by other constructs in the model. Table 3 presents the discriminant validity results based on the Fornell–Larcker criterion. This criterion compares the square root of the AVE of a construct with its correlations with other constructs. For satisfactory discriminant validity, the diagonal elements should be significantly higher than the off-diagonal elements in the corresponding rows and columns (Fornell and Larcker, 1981). Table 4 presents the discriminant validity assessment using the Heterotrait–Monotrait ratio of correlations (HTMT), which measures the average of Heterotrait–Heteromethod correlations (Henseler et al., 2015). An HTMT value below 0.85 (or 0.90 in certain contexts) indicates that the constructs are distinct.
Measurement model: discriminant validity. Fornell–Larcker
| BI | EE | FC | HM | INT | IQ | PE | PP | PS | SI | PT | |
|---|---|---|---|---|---|---|---|---|---|---|---|
| BI | 0.915 | ||||||||||
| EE | 0.201 | 0.851 | |||||||||
| FC | 0.284 | 0.626 | 0.782 | ||||||||
| HM | 0.585 | 0.309 | 0.455 | 0.901 | |||||||
| INT | 0.142 | 0.452 | 0.461 | 0.399 | 0.921 | ||||||
| IQ | 0.361 | 0.293 | 0.384 | 0.240 | 0.028 | 0.903 | |||||
| PE | 0.634 | 0.230 | 0.344 | 0.383 | 0.000 | 0.543 | 0.915 | ||||
| PP | 0.039 | −0.105 | −0.204 | −0.115 | 0.033 | −0.176 | −0.095 | 0.916 | |||
| PS | 0.020 | 0.110 | 0.123 | 0.151 | 0.251 | 0.342 | 0.080 | −0.112 | 0.935 | ||
| SI | 0.555 | 0.087 | 0.194 | 0.300 | −0.002 | 0.333 | 0.516 | −0.038 | 0.177 | 0.882 | |
| PT | 0.088 | 0.077 | 0.229 | 0.196 | 0.182 | 0.460 | 0.314 | −0.233 | 0.432 | 0.181 | 0.904 |
Measurement model: discriminant validity. HTMT ratio
| BI | EE | FC | HM | INT | IQ | PE | PP | PS | SI | PT | |
|---|---|---|---|---|---|---|---|---|---|---|---|
| BI | |||||||||||
| EE | 0.232 | ||||||||||
| FC | 0.305 | 0.752 | |||||||||
| HM | 0.655 | 0.362 | 0.525 | ||||||||
| INT | 0.152 | 0.540 | 0.511 | 0.439 | |||||||
| IQ | 0.400 | 0.337 | 0.500 | 0.267 | 0.089 | ||||||
| PE | 0.698 | 0.272 | 0.439 | 0.425 | 0.081 | 0.594 | |||||
| PP | 0.079 | 0.190 | 0.268 | 0.125 | 0.089 | 0.182 | 0.110 | ||||
| PS | 0.079 | 0.130 | 0.192 | 0.171 | 0.279 | 0.373 | 0.089 | 0.109 | |||
| SI | 0.607 | 0.109 | 0.248 | 0.337 | 0.059 | 0.349 | 0.566 | 0.073 | 0.198 | ||
| PT | 0.114 | 0.125 | 0.298 | 0.233 | 0.211 | 0.534 | 0.376 | 0.246 | 0.526 | 0.208 |
Next, we performed the fitting indexes of the saturated and estimated model to conduct a confirmatory analysis of the model.
The standardized root mean square residual fitting index is below 0.08 (Hu and Bentler, 1998). In addition, the exact fit tests based on bootstrapping have been carried out, all of them are within the values considered adequate to affirm that it is a quality measurement model, as shown in Table 5.
Fitting test
| Original sample(O) | Sample mean (M) | 95% | 99% | |
|---|---|---|---|---|
| SRMR | ||||
| Saturated model | 0.067 | 0.067 | 0.091 | 0.114 |
| Estimated model | 0.070 | 0.070 | 0.096 | 0.117 |
| d_ULS (unweighted least squares discrepancy) | ||||
| Saturated model | 2.860 | 2.922 | 5.230 | 8.227 |
| Estimated model | 3.095 | 3.220 | 5.764 | 8.596 |
| dG | ||||
| Saturated model | 2.149 | 3.011 | 4.731 | 6.040 |
| Estimated model | 2.173 | 3.081 | 4.672 | 5.684 |
4.2 Summary of test results for structural model
The 10 hypotheses of the model were tested with a bootstrapping. Table 6 shows that hypotheses H1, H3, H5, H8 y H9 are supported. The PE, the SI and the HM affects the BI. On the other hand, the perception of the quality and the perception of security have an influence on trust.
Structural model
| Hypothesis | Path | Standardized path coefficient | p-value | Supported | Construct | Rˆ2 |
|---|---|---|---|---|---|---|
| H1 | PE → BI | 0.431 | 0.000 | Yes | BI | 0.615 |
| H2 | EE → BI | −0.017 | 0.880 | No | ||
| H3 | SI → BI | 0.261 | 0.001 | Yes | ||
| H4 | FC → BI | −0.071 | 0.543 | No | ||
| H5 | HM → BI | 0.388 | 0.000 | Yes | ||
| H6 | INT → BI | 0.057 | 0.541 | No | ||
| H7 | PT → BI | −0.164 | 0.053 | No | ||
| H8 | IQ → PT | 0.332 | 0.004 | Yes | T | 0.316 |
| H9 | PS → PT | 0.302 | 0.007 | Yes | ||
| H10 | PP → PT | −0.140 | 0.218 | No |
On the other side, the variance value explained for the latent dependent variables is shown with R2, obtaining suitable values greater than or the same as 0.1, in accordance with Falk and Miller (1992). Hence, variables reach a satisfactory level of explanatory power. Figure 3 shows results of testing the model.
The flowchart starts with a circle labeled “Behavioral Intention (B I)” positioned at the top center. A legend at the bottom left indicates that solid arrows represent “Significant path” and dashed arrows represent “Non-significant path.” Seven individual arrows point to “Behavioral Intention (B I)” from circles arranged in a semicircular pattern below the central circle. The details about the circles and the arrows pointing from each are as follows: The arrow labeled “H 1” points from the circle labeled “Performance Expectancy (PE).” The dashed arrow labeled “H 2” points from “Effort Expectancy (E E).” The arrow labeled “H 3” points from “Social Influence (S I).” The dashed arrow labeled “H 4” points from “Facilitating Conditions (F C).” The arrow labeled “H 5” points from “Hedonic Motivation (H M).” The dashed arrow labeled “H 7” points from “Trust (P T).” The dashed arrow labeled “H 6” points from “Innovativeness (I N T).” Three circles are arranged in a horizontal series below “Trust (P T).” Each circle has an arrow pointing to “Trust (P T).” From left to right, the details for each circle and the labeled arrows are as follows: The arrow labeled “H 8” points from “Information Quality (I Q).” The arrow labeled “H 9” points from “Perceived Security (P S).” The dashed arrow labeled “H 10” points from “Perceived Privacy (P P).”Results of testing the model. Source(s): Authors’ own elaboration
The flowchart starts with a circle labeled “Behavioral Intention (B I)” positioned at the top center. A legend at the bottom left indicates that solid arrows represent “Significant path” and dashed arrows represent “Non-significant path.” Seven individual arrows point to “Behavioral Intention (B I)” from circles arranged in a semicircular pattern below the central circle. The details about the circles and the arrows pointing from each are as follows: The arrow labeled “H 1” points from the circle labeled “Performance Expectancy (PE).” The dashed arrow labeled “H 2” points from “Effort Expectancy (E E).” The arrow labeled “H 3” points from “Social Influence (S I).” The dashed arrow labeled “H 4” points from “Facilitating Conditions (F C).” The arrow labeled “H 5” points from “Hedonic Motivation (H M).” The dashed arrow labeled “H 7” points from “Trust (P T).” The dashed arrow labeled “H 6” points from “Innovativeness (I N T).” Three circles are arranged in a horizontal series below “Trust (P T).” Each circle has an arrow pointing to “Trust (P T).” From left to right, the details for each circle and the labeled arrows are as follows: The arrow labeled “H 8” points from “Information Quality (I Q).” The arrow labeled “H 9” points from “Perceived Security (P S).” The dashed arrow labeled “H 10” points from “Perceived Privacy (P P).”Results of testing the model. Source(s): Authors’ own elaboration
5. Conclusions
The objective of this paper is to explore the driving factors to the musicians in the use on a Spanish reward-based crowdfunding platform. The paper covers a literature review of variables and relationships used in extended Unified Theory of Acceptance and Use of Technology (UTAUT2), considered an extension of UTAUT.
The results of this research suggest important practical and theoretical contributions regarding musicians' behavior when using reward-based crowdfunding platforms.
In the context of musical crowdfunding, musicians perceive that using platforms such as Verkami enhances their ability to finance projects efficiently, which in turn reinforces their intention for continued use. This result aligns with previous crowdfunding studies that link PE with successful project creation, as indicated by Pangaribuan and Wulandar (2018) in technology adoption models. This finding directly supports the core principles of the UTAUT2 model, where Venkatesh et al. (2012) identify PE as a critical predictor of technology adoption.
On a broader theoretical basis, the connection between PE and crowdfunding success has also been observed in research on cultural collective financing, where the ability to reach financial goals quickly acts as a key driving force (Moon and Hwang, 2018; Kim and Hall, 2020).
The role of SI verifies the UTAUT2 framework, which emphasizes how perceptions from close social contexts (family, colleagues) affect technology adoption (Venkatesh et al., 2012). In this study, musicians show a greater propensity to use platforms when they perceive social validation, a phenomenon documented in reward-based crowdfunding where community support is key (Gerber and Hui, 2013; Li et al., 2018).
From a theoretical perspective, although some studies question the relevance of SI in non-organizational contexts (Raza et al., 2019), this work demonstrates its validity in cultural sectors, where positive social pressure acts as a catalyst for creative projects.
HM, incorporated into UTAUT2 for consumer contexts, explains how enjoyment and fun influence technology adoption (Venkatesh et al., 2012). In musical crowdfunding, musicians value the platform as a space to create creative and emotional content, reinforcing previous findings that link HM with engagement strategies in cultural projects (Schulz et al., 2015; Khizar and Siddiqui, 2021).
This result extends the work of Kim et al. (2020), who primarily associated HM with donors, by demonstrating that it is also crucial for creators when designing campaigns.
Although trust (PT) did not show a direct impact on BI, its antecedents—information quality and security—were significant. This supports cognition-based trust models, where transparency and technical guarantees are fundamental (Gefen, 2000; Flavián and Guinalíu, 2006). In crowdfunding, IQ reduces asymmetry between creators and backers, while security mitigates risk perceptions, as noted by Bonsón et al. (2015) in online transaction contexts.
Regarding theoretical implications, these finding qualify previous studies that prioritized trust as a direct predictor (Moysidou and Hausberg, 2020), suggesting that in specialized environments such as music, instrumental factors (IQ and PS) may be more relevant than general trust.
The lack of impact from personal INT partially contradicts Rogers (1995) and Agarwal and Prasad (1998), who link it to early technology adoption. However, this aligns with crowdfunding research that highlights the challenges faced by sectors whose projects often find it difficult to secure funding through conventional channels, as they are perceived by traditional financial institutions as too innovative, complex, or risky (Gür and Özdoğan, 2021). In this context, crowdfunding may emerge as a viable alternative for cultural initiatives, which frequently face similar barriers to accessing institutional financing.
FC were also not significant, contrary to findings in general crowdfunding studies (Tariqui and Tahir, 2019). This suggests that, for musicians, the technical usability of the platform is less relevant than its ability to generate tangible results (PE) or social recognition (SI).
The results reinforce the applicability of UTAUT2 in cultural contexts, where emotional (HM) and social (SI) factors complement rational ones (PE).
6. Practical implications
Analyzing the factors that may influence musicians' use of rewards crowdfunding platforms has important benefits for both platform promoters and entrepreneurs who want to launch musical projects.
This research can help to better understand musicians' experiences, so that reward-based crowdfunding platforms can improve their dynamics and ultimately increase their usability in the musical sector.
In light of the conclusions reported, we believe that our research makes an important contribution to the knowledge of the experiences in using crowdfunding platforms from the perspective of the entrepreneurs of music projects. The results obtained present new findings on previous research, providing useful knowledge for users and managers of reward-based crowdfunding platforms. It is also useful for public decision-makers in the field of culture and arts promotion, who can see the importance of this type of crowdfunding tools and thus collaborate to foster them.
Although the study focuses on musicians, the identified factors may also be relevant to other creative sectors that use reward-based crowdfunding. Thus, this work expands the understanding of financial technology adoption in cultural contexts, providing empirical evidence useful for both scholars and platform managers
7. Limitations and future research lines
This study identifies several limitations that open up new opportunities for future research.
First, the data analyzed come from a specific reward-based crowdfunding platform, which prevents the results from being generalized to all musicians who have obtained funding through this method. Therefore, it would be valuable for future studies to evaluate the proposed model using data from other platforms and from different countries, which would increase the external validity of the findings.
On the other hand, this study could be extended to platforms that develop other kinds of crowdfunding as potential research avenues (donation, lending and equity). In this way, it would be possible to compare the variables that are important for either modality.
It would also be interesting to test the model on different types of businesses within the music sector, such as soloists, orchestras or opera, among others, which would represent another promising line of research.
Future research could expand the data scope and also explore changes in crowdfunding adoption over time and not limit the analysis to cross-sectional data.
Furthermore, the study does not examine whether sociodemographic variables, such as age or gender, can influence the relationships between the variables in the model. Consequently, further research should analyze whether these characteristics act as moderators, to deepen our understanding of musicians' behavior.
Management area: Management
