1 Introduction

Crowdfunding has revolutionized entrepreneurial fundraising by offering a digital alternative to traditional fundraising methods (Hossain and Oparaocha 2017). It has experienced a major growth following the 2008 economic crisis, when the supply of capital to small and medium-sized companies was substantially reduced (Lehner et al. 2015). In addition, crowdfunding has the potential to promote innovation, not only by facilitating the flow of funds to innovative startups that would otherwise face credit constraints, but also by allowing the crowd of investors to participate in the entrepreneurial idea and thus contribute to its improvement and development (Hervé and Schwienbacher 2018).

However, the rise of this fundraising method has significantly disrupted the existing dynamics, interactions, and networks that characterize traditional fundraising processes between entrepreneurs and potential investors. In fact, the digital nature of financial transactions mediated by crowdfunding exacerbates the existing information asymmetries between fundseeking entrepreneurs and potential investors or funders (Agrawal et al. 2014; Courtney et al. 2017). Therefore, it is necessary to thoroughly investigate the information disclosure strategies that generate trust and encourage potential backers to contribute funds to support a particular entrepreneurial idea (Butticè and Vismara 2022).

Signaling theory has been used extensively in crowdfunding research to understand which signals from fund-seeking entrepreneurs incentivize potential backers to invest (Ahlers et al. 2015; Piva and Rossi-Lamastra 2018; Vismara 2018). A study by Shafi (2021) on equity crowdfunding suggests that crowd investors may lack the knowledge and experience needed to make complex investment decisions, and therefore focus their attention on factors (i.e., signals) that are easier to assess. Indeed, the democratization of fundraising made possible by crowdfunding has not only provided entrepreneurs with an alternative means of raising capital but has also allowed less experienced individuals who would not have been able to invest in entrepreneurial projects through traditional channels to do so with small amounts of capital through crowdfunding (Mollick and Robb 2016).

The aim of this research is to analyze the signaling power of traditional visual cues and reporting presence in social networks in mitigating the information asymmetries that characterize equity crowdfunding campaigns, distinguishing between information disclosure strategies conducive to crowdfunding success and overfunding, which is operationalized here as raising 10% or more of the target funding (Sendra-Pons et al. 2023). To this end, qualitative comparative analysis (QCA) is used to analyze two models, one aimed at examining which combinations of traditional visual cues and social networks presence are conducive to equity crowdfunding success (Model 1) and the other to overfunding (Model 2). The data were collected from the Spanish equity crowdfunding platform StartupxploreFootnote 1.

The configurational methodology allows delving into the underlying complexity of information disclosure strategies that lead to success and overfunding in equity crowdfunding. Previous research has employed QCA to study sponsor satisfaction (Xu et al. 2016) and project quality and entrepreneurs’ credibility (Huang et al. 2022) in reward-based crowdfunding. It has also been employed to study the various factors that influence the success and failure of donation-based (Peng et al. 2022) and equity crowdfunding campaigns (De Crescenzo et al. 2020), and to examine investor’s confidence in crowdlending environments (Ferrer et al. 2022). However, the uniqueness of this research lies in providing updated insights concerning two specific information elements, i.e., traditional visual cues and the presence of social networks, in signaling the entrepreneurial project quality, building trust and incentivizing investment. Interestingly, by juxtaposing both signals, this research aims to contribute to a future research avenue on signal complementarity identified by Mochkabadi & Volkman (2020) and even test possible substitutability between signals. Furthermore, the equifinality (i.e., multiple combinations of traditional visual cues and social networks presence can lead to crowdfunding success and overfunding) and multifinality (i.e., the same combinations can lead to different outcomes) underlying QCA allows for a simple yet comprehensive approach to the often compound nature of entrepreneurial fundraising dynamics. However, QCA is limited by the number of conditions it considers.

The remainder of the paper is organized as follows. First, a theoretical framework is developed, focusing on the dynamics of equity crowdfunding, the theories of asymmetric information and signaling, and the role of traditional visual cues and the presence of social networks as signals. The propositions of the study are then formulated. This is followed by an overview of the data considered in the study and the methodology employed. The results are then presented and the information disclosure strategies obtained are discussed, both in terms of those that are conducive to crowdfunding success and those that are conducive to overfunding. The paper ends with conclusions, theoretical contributions, practical implications, limitations and future lines of research.

2 Theoretical framework

The previous literature has highlighted the difficulties faced by entrepreneurs in raising the financial resources needed to pursue their nascent business ideas (Butticè and Vismara 2022). These difficulties are even more pronounced during economic and financial crises when entrepreneurs have to face severe credit constraints and tighter credit conditions (Papaoikonomou et al. 2012). In fact, an early report by the Organization for Economic Co-operation and Development (OECD) on the impact of the 2008 global crisis on SME and entrepreneurship finance already discussed tighter bank lending policies, both in terms of guarantees and loan sizes (OECD 2009).

Crowdfunding can be defined as a fundraising method used by the entrepreneurial ecosystem —and by individuals who do not intend to develop a business idea— in which funds are raised from a crowd of potential investors/funders who contribute small amounts (Hossain and Oparaocha 2017). As reported by Pichler and Tezza (2016), its growth and rapid scaling can be attributed to two main drivers. On the one hand, the popularization of Web 2.0 made it possible to mediate fundraising through a digital environment easily accessible through the Internet. On the other hand, there was a latent need to supplement the shrinking flow of credit during the global financial crisis of 2008.

The different types of crowdfunding can be classified according to (i) the rewards offered to backers and (ii) the dynamics of the fundraising process. The rewards offered allow distinguishing between investment and non-investment crowdfunding models. In the former, backers receive a monetary reward for their contribution. This is the case of crowdlending, which is similar to a traditional loan in which a larger number of lenders, rather than a single financial institution, contribute small amounts in order to receive the principal plus a financial interest (fixed or variable) as agreed in advance. It also includes royalty-based crowdfunding, where contributions are rewarded with royalties, or equity crowdfunding, where backers become owners of the startup in proportion to their contribution. In turn, non-investment crowdfunding models include donation-based crowdfunding, a philanthropically driven form of fundraising that does not consider a financial reward, and reward-based crowdfunding, which offers backers units of the product or service being marketed (Belleflamme and Lambert 2016; Shneor 2020).

In terms of fundraising dynamics, crowdfunding can follow an ‘all-or-nothing’ or ‘keep-it-all’ mode (Cumming et al. 2020). In an ‘all-or-nothing’ campaign, the entrepreneur(s) can only access the funds contributed by the backers if the funding target set at the beginning of the campaign is met, thus returning the funds if the target is not met. These campaigns are known to be riskier, both for the entrepreneur(s), who will not have access to the contributed funds if the target is not met, and for the backers, who may incur an opportunity cost for the time their funds were withheld if the campaign fails. In contrast, a ‘keep-it-all’ campaign allows the entrepreneur(s) to access the funds even if the funding goal is not met. However, alternative sources of funding need to be explored to complement crowdfunding, which may increase the risk to backers in terms of the entrepreneur’s ability to deliver the entrepreneurial project as promised.

This research focuses on equity crowdfunding campaigns that follow an ‘all-or-nothing’ fundraising dynamic. These campaigns are developed in the following section, emphasizing the existing information asymmetries and the signaling process that takes place between fund-seeking entrepreneurs and potential investors.

2.1 Equity crowdfunding: information asymmetries and signaling

As previously mentioned, equity crowdfunding consists of a crowd of individuals investing in an often entrepreneurial project over the Internet in exchange for equity. Among the family of crowdfunding types, equity crowdfunding can be considered the most complex because the acquisition of equity by investors makes the contracting process more challenging and often requires an extensive due diligence process (Vulkan et al. 2016). Gierczak et al. (2016) also note that the complexity associated with “the provision of capital and the resulting returns” is the highest in equity crowdfunding, when compared to other models. In this digital investment platform, there are pronounced information asymmetries between fund-seeking entrepreneurs and potential investors (Courtney et al. 2017). Moreover, the democratization that equity crowdfunding has brought in terms of individuals being able to invest small amounts of money has meant that, as Ahlers et al. (2015) point out, these platforms are targeted at small investors whose ability to evaluate the different projects is limited. All this makes the need for signals that entrepreneurs can use to demonstrate to the crowd of potential investors the suitability of investing in their projects even more pressing.

As Mochkabadi and Volkmann (2020) point out, equity crowdfunding research has recently gained prominence, with one of the lines of research being the study of signals and how different signals complement each other. This line of research has been addressed on numerous occasions through signaling theory (Spence 1973). In financial environments characterized by large information asymmetries between economic agents (Akerlof 1970), as is the case with crowdfunding, this theory suggests that the more informed party (here, the fund-seeking entrepreneurs) must send signals to the less informed party (here, the crowd of potential investors) in order to generate confidence in the suitability of investing in a given project. In this way, the information gap between the two agents is narrowed and the crowd of potential investors, often inexperienced to make an in-depth evaluation of the projects, is provided with signals to make investment decisions.

A comprehensive systematic literature review by Mazzocchini and Lucarelli (2022) notes that while the study of the determinants of success and failure in equity crowdfunding is “fast growing”, fragmentation still dominates, with studies adopting different theoretical perspectives. The authors acknowledge that, although visual cues (e.g., the use of pitch videos or images) and the presence of social networks have been used as explanatory variables, the study of images and pitch videos “cannot be overlooked” in the future research agenda.

This research focuses on two specific sets of signals: first, traditional visual cues, operationalized as the number of images and the length of videos, and second, social networks presence, specifically on Instagram, Facebook, and Twitter. This empirical paper uses qualitative comparative analysis (QCA) to explore the interplay of traditional visual cues and social networks presence in leading to both equity crowdfunding success and overfunding. In doing so, the study aims to identify relevant information disclosure strategies and the complementarity/substitutability of the different cues considered. In this regard, it aims to inform fund-seeking entrepreneurs and equity crowdfunding intermediary platforms about information disclosure strategies associated with equity crowdfunding success and overfunding. The next section develops the previous literature and formulates the propositions.

2.2 Traditional visual cues: images and videos

It should be noted that, as discussed by Hossain and Oparaocha (2017), visual communication is central to crowdfunding campaigns. Through the use of visual content, such as images or videos, important information is conveyed from entrepreneurs, as project initiators, to potential backers, thereby building trust and commitment among the various actors involved by reducing information gaps.

The use of images and videos as signals in crowdfunding campaigns has been studied extensively. Previous evidence has shown how visual cues such as images and videos influence backers’ decision-making in different types of crowdfunding campaigns (e.g., Courtney et al. (2017), for reward-based crowdfunding, Li et al. (2016), for equity crowdfunding, or Kim et al. (2022), for donation-based crowdfunding). In this way, images and videos help bridge information asymmetries between fund seekers and backers, incentivizing investment.

As summarized by Xu (2018), videos and images have been identified as powerful psychological cues when it comes to persuading potential backers. In this sense, images and videos help to provide a more complete story around the entrepreneurial idea, which can better engage the potential investor. However, for reward-based crowdfunding campaigns, Yang et al. (2020) found that images and videos can reduce the effectiveness of the text length in improving funding performance (overshadowing effect).

For videos, Kolbe et al. (2022) summed up a number of findings: Mollick (2014), Courtney et al. (2017), Cumming et al. (2017), Li et al. (2016), Anglin et al. (2018), and Johan and Zhang (2020), using a dummy variable for whether the campaign had a video, found a positive correlation between the release of a video and the success of the campaign. Complementarily, Scheaf et al. (2018) found that video quality, treated as a continuous variable, has a positive correlation with crowdfunding success; Thapa (2020) identified a curvilinear relationship between video length and funding success; Parhankangas and Renko (2017) demonstrated the importance of a language style that is concrete, precise, and interactive; and Kim et al. (2016) also found that a video’s language style affects success. Furthermore, Hu and Ma (2021) delved specifically into the tone of videos and found that passionate and warm videos were positively correlated with funding success. Additionally, Troise et al. (2020) found no effect of video length on funding raised.

Overall, as Carradini and Fleischmann (2023) have recently noted, successful campaigns have a higher number of images and typically include a project video, in addition to links and gifs. However, only images, videos, and links were found to have a positive impact on success. Data were collected from Kickstarter, a rewards-based crowdfunding platform, suggesting that these elements are relevant in attracting investors interested in rewards. For video content in particular, Wasiuzzaman and Suhili (2023) specifically found that its length is highly significant in positively influencing success, meaning that a longer video attracts more backers. The authors suggest that this is due to the reduction of uncertainty and information asymmetries.

2.3 Social networks: Instagram, Facebook, and Twitter

As pointed out by Aral et al. (2013), social media have vastly transformed the way people communicate with each other and, consequently, the way people consume. Unlike images and videos, which are static information elements (once published in the campaign, they do not change), links to the entrepreneur’s or startup’s social networks can be considered dynamic information elements, since the interaction on these platforms is constantly updated. In this sense, the interaction with potential investors through Instagram, Facebook and TwitterFootnote 2 provides an environment for the development of online word-of-mouth, where entrepreneurs can not only provide updated information to potential investors but engage more deeply with the entrepreneurial idea (Borst et al. 2018).

In the context of equity crowdfunding, information asymmetries can also be mitigated by entrepreneurs’ social connections. Thus, social networks are useful for reducing uncertainty, attracting attention, and increasing the chances of a campaign’s success (Vismara 2016). Indeed, as stated by Nitani et al. (2019), prior evidence shows that the richness of a firm’s or entrepreneur’s social networks sends positive signals in terms of trust and project quality. In a reward-based crowdfunding context, the number of social media accounts for an entrepreneurial project increases both the number of investors and the amount of money pledged (Clauss et al. 2020).

Recent findings by Wasti and Amed (2023) confirm the central role of social networks as a source of word-of-mouth that contributes to a campaign having a greater impact on potential backers. In this sense, the authors found a positive and significant relationship between social forums and overfunding. Similarly, Graziano et al. (2023) show that a platform manager’s social connections are relevant to crowdfunding success, linking social activity to equity crowdfunding performance. They also point out that online interaction with investors, which may be enabled in part by social networks, facilitates a feedback mechanism that leads to product and service innovation. This is consistent with the importance of social network ties in crowdfunding identified by Mollick (2014).

Based on previous evidence, it is expected that visual content and social networks disclosure play an important role in crowdfunding success and overfunding in equity crowdfunding campaigns (Proposition1), and that different combinations of these information cues, i.e., strategies, are behind both outcomes (Proposition2).

Proposition 1

Information disclosure about visual content (i.e., images and videos) and presence on social networks (i.e., Instagram, Facebook and Twitter) are relevant to campaign success and overfunding in equity crowdfunding, but no condition is necessary for such phenomena.

Proposition 2

Success and overfunding in equity crowdfunding can be achieved through heterogeneous information disclosure strategies where equifinality and multifinality are observed.

3 Data and method

3.1 Data

The data were hand-collected using information provided by the Spanish equity crowdfunding platform Startupxplore. Under the legal name of Startupxplore PFP, S.L., this is a platform authorized by the Spanish National Securities Market Commission in accordance with Law 5/2015 on the Promotion of Business Financing. According to information provided by Startupxplore, in June 2016 it became the second largest community in Europe, with more than 11,000 companies from 80 countries. A total of 43 campaigns raising more than €8 million between 2017 and 2021 were analyzed. The sample included successful projects, i.e., projects that had reached the funding goal before the end of the campaign, and unsuccessful projects.

All of the data that make up the conditions used in this study to run the Qualitative Comparative Analysis (QCA) model refer to information elements that are available to potential backers when they decide on which campaign to invest in. That is, images, videos, and presence on social networks are information displayed on the campaign microsite. In this sense, although LinkedIn is a very useful social network for business, only Instagram, Facebook and Twitter presence are used in the analysis, as these are the sole social networks reported by the fund-seeking entrepreneurs according to the data.

In addition to the data on the various conditions regarding visual content and social networks disclosure, information was collected on the amount of investment raised by the various projects in order to compile the two outcomes, SUCC and OVERF, the former referring to the success of the campaign and the latter to whether the campaign was overfunded, i.e., raised 10% or more over the target. Overfunding is characterized as a situation in which the funding target set by the entrepreneur seeking funding is exceeded by 10% or more, to avoid categorizing as such situations in which the funding target is marginally exceeded. In this sense, a campaign that minimally exceeds the threshold of success would not represent the overfunded campaigns we are interested in studying.

The use of two outcomes, i.e., SUCC and OVERF, is motivated as follows. Success, defined as reaching the funding target within the campaign timeframe, has been studied repeatedly in the crowdfunding field (see Mochkabadi and Volkmann (2020) for a list of studies on the determinants of campaign success). Indeed, the ultimate goal of the fundraising process through alternative crowd-based funding methods is to achieve the funding goal. Otherwise, if an all-or-nothing model is followed, there is no transfer of funds from the crowd to the fund-seeking entrepreneur. However, despite its popularity in crowdfunding research, it is so important to the development of an efficient alternative financing crowdfunding environment that it requires greater scrutiny. Alternatively, overfunding, which has not been addressed as extensively in the previous literature (Adamska-Mieruszewska et al., 2019), raises a number of issues or challenges that merit investigation into the determinants of the phenomenon. According to Li et al. (2022), some campaigns may fail to reach their funding target by losing prominence to overfunded campaigns. In addition, overfunded campaigns face difficulties in efficiently managing the unplanned surplus of funds. In the authors’ words, overfunding leads to market inefficiency in crowdfunding as it generates a “suboptimal allocation of scarce monetary resources” (Li et al. 2022, p. 2). In this sense, the appropriateness of studying overfunding lies in exploring those information disclosure strategies, in this case related to visual content and the disclosure of social networks presence in the campaign, that lead to this phenomenon with notorious perverse effects on crowdfunding. Both outcomes being analyzed, i.e., success [SUCC] and overfunding [OVERF], are operationalized as dichotomous or crisp values. For SUCC, a value of 1 means that a particular campaign was successful, while a value of 0 means that it did not reach the target funding. For OVERF, a value of 1 means that a particular campaign was overfunded, while a value of 0 means it was not.

Among the conditions included in the models analyzed, two of them are continuous or fuzzy values, namely image content [IMAGE] and video content [VIDEO], while the other three are dichotomous or crisp values, namely the presence of an Instagram account [INSTA], Facebook account [FACEB] and Twitter account [TWITT]. Image content is operationalized as the number of images following De Crescenzo et al. (2020), thus extending the coding method of Usman et al. (2019), which assigns different scores depending on whether there are videos and images, only one, or none. Going beyond a dichotomous characterization (see Li et al. 2016 and Mamonov and Malaga 2019), video content is operationalized as the length (in seconds) of the reported video to better capture the amount of information conveyed. Finally, as in Balboni et al. (2014) and Beier and Wagner (2015), we only consider the disclosure of social network accounts. This is owing to two reasons: first, it is impossible to have a static number of followers in each social network, since they vary throughout the fundraising campaign; and, second, the data on users in these social networks that we had available were historical, and there may be a mismatch between these data and the data at the beginning of the campaign. A summary of the operationalization of the outcome and conditions is presented in Table 1.

Table 1 Outcome and conditions used in the study
Full size table

3.2 Method

This paper adopts a configurational perspective through a qualitative comparative analysis (QCA). Often, social and economic events exhibit configurational multiplicity, where different paths can lead to a given outcome (Park et al. 2020). QCA is an appropriate method to study the success and overfunding in equity crowdfunding given the causal complexity surrounding this fundraising method. In fact, QCA allows for multiple conjunctural causation, equifinality —i.e., combinations of different elements of information disclosure can explain success and overfunding— and multifinality —i.e., the same elements of information disclosure can lead to different outcomes (Ragin 1987; Marx et al. 2014; Gerrits and Pagliarin 2021).

For this reason, the configurational approach offered by the QCA methodology, through the analysis of the interaction between the conditions of the model, enables going deeper into the reasons that produce a given phenomenon. Thus, this methodology is case-oriented and uses Boolean logic to obtain as solutions specific configurational patterns to achieve the outcome under analysis. It is suitable for small-N designs but has also been used for intermediate- and large-N designs (Berg-Schlosser et al. 2009). Through a configurational approach using QCA, both necessary and sufficient conditions leading to the outcome under study are revealed (Roig-Tierno et al. 2017). A condition is said to be necessary if it is always present when the outcome under analysis occurs. Alternatively, a condition is said to be sufficient for a certain outcome when it occurs in the presence of such a condition (Ragin 2008a; Schneider and Wagemann 2012).

Similar studies that investigate the crowdfunding phenomenon, its dynamics and determinants, and employ QCA include, for example, Xu et al. (2016), De Crescenzo et al. (2020), Ferrer et al. (2022), Huang et al. (2022), and Peng et al. (2022). As mentioned earlier, the equifinality and multifinality considered in such a configurational approach makes this method of particular interest in this research topic. By embracing causal complexity, QCA facilitates the derivation of causal recipes that contribute to a more explicit understanding of the not-so-obvious dynamics of the signaling process.

According to the theoretical framework developed, this paper aims to delve into those conditions related to visual content (i.e., the volume of image and video content) and social networks (i.e., the entrepreneur’s reported presence on Instagram, Facebook and Twitter) that lead to the success of an equity crowdfunding campaign (Model 1) and to overfunding (Model 2). The models are formulated as follows:

Model 1 \(SUCC=f(IMAGE,\,VIDEO,\,INSTA,\,FACEB,\,TWITT)\)

Model 2 \(OVERF=f(IMAGE,\,VIDEO,\,INSTA,\,FACEB,\,TWITT)\)

where “SUCC” means crowdfunding success and “OVERF”, overfunding.

Data calibration for fuzzy values was performed according to Berné-Martínez et al. (2021). The full non-membership point was set at 50% below the sample mean; the maximum ambiguity point was set at the sample mean; and the full membership point was set at 20% above the sample mean (see Table 2). Crisp values do not require calibration. Once the data were calibrated, the analysis of necessary conditions was performed first and then the logically feasible configurations of conditions for Models 1 and 2 were obtained. The results of these analyses are presented in the next section.

Table 2 Calibration for outcome and conditions
Full size table

4 Results

First, the analysis of necessary conditions is performed. For a condition to be necessary, its consistency must be greater than or equal to 0.9 (Schneider and Wagemann 2012). According to this analysis, no condition, either in presence or in absence, either when the outcome is crowdfunding success [SUCC] or when it is overfunding [OVERF], obtains a consistency equal to or greater than 0.9. It follows that no condition can be considered necessary. For SUCC, the presence of a Facebook account (0.594595) and the absence of a Twitter account (0.594595) are the conditions with the highest consistency values. For OVERF, the presence of images (i.e., a large number of images, 0.58000) is the condition with the highest level of consistency. At this stage, it is worth clarifying that the absence (or negation)Footnote 3 of a crisp condition implies that this condition is missing (e.g., “~INSTA” means that an Instagram account is not disclosed) while the absence (or negation) of a fuzzy condition indicates a low level of this condition and not its total absence. For example, “~IMAGE” means that few images are disclosed, while “~VIDEO” means that a short video is revealed. The results of the analysis of the necessary conditions are shown in Table 3.

Table 3 Analysis of necessary conditions
Full size table

Among the three solutions reported by fsQCA, namely the parsimonious, the complex and the intermediate solution, the first is the one discussed in this paper. The parsimonious solution includes simplifying assumptions introduced by researchers. Alternatively, the complex solution considers neither easy nor difficult counterfactuals, while the intermediate solution considers simplifying assumptions based on easy counterfactuals (Ragin 2008a). Table 4 shows the parsimonious solution for Model 1, where the outcome is crowdfunding success [SUCC]. Six different causal configurations emerge, for which the raw coverage, that indicates how much of the outcome is explained by a given solution, and the unique coverage, which indicates the proportion of the outcome explained by each individual condition in a configuration, are reported (Ragin 2008b). Consistency refers to “the percentage of causal configurations of similar composition which result in the same outcome value” (Roig-Tierno et al., 2017, p.17). Consistency is above 0.8 in all causal configurations, exceeding the threshold above which goodness of fit is considered to apply (Ragin 2008a). Indeed, consistency reaches its maximum value, i.e., 1, in three out of six reported causal configurations. Overall, the parsimonious solution for Model 1 explains about 74% of the empirical cases, as evidenced by a solution coverage of 0.742703.

Table 4 Parsimonious solution for Model 1
Full size table

The description of the causal configurations derived from the parsimonious solution of Model 1 is presented below. Thus, the success of a crowdfunding campaign can be explained by:

Causal configuration [1]. The disclosure of a large number of images despite the absence of a Twitter account.

Causal configuration [2]. The disclosure of a Twitter account despite the disclosure of few images and the absence of an Instagram account.

Causal configuration [3]. The disclosure of a large number of images and an Instagram account despite the disclosure of a short video.

Causal configuration [4]. The disclosure of a long video despite the absence of an Instagram account.

Causal configuration [5]. The disclosure of a Twitter account despite the absence of a Facebook account.

Causal configuration [6]. The disclosure of an Instagram and a Facebook account despite the absence of a Twitter account.

On the other hand, when testing the outcome overfunding [OVERF], i.e., raising 10% or more funds above target, four causal configurations are reached, as can be seen in the parsimonious solution for Model 2 reported in Table 5. In this case, the solution coverage is 0.444333, which means that approximately 44% of the empirical cases are explained by the four causal configurations mentioned. Again, consistency exceeds 0.8 and reaches its maximum value for one causal configuration.

Table 5 Parsimonious solution for Model 2
Full size table

The description of the causal configurations associated with the parsimonious solution for Model 2, explaining campaign overfunding, is as follows:

Causal configuration [7]. The disclosure of a long video despite the absence of an Instagram account.

Causal configuration [8]. The disclosure of a large number of images despite the absence of both an Instagram and a Twitter account.

Causal configuration [9]. The disclosure of a Twitter account despite the absence of a Facebook account.

Causal configuration [10]. The disclosure of both a long video and a Facebook account despite the disclosure of few images and the absence of a Twitter account.

5 Discussion

5.1 Information disclosure for successful campaigns

Figure 1 shows the different configurations of conditions that lead to crowdfunding success, distinguishing between conditions related to (i) visual content, i.e. images and videos, and (ii) social networks, i.e., Instagram, Facebook, Twitter. Section A shows the conditions related to visual content and Section B the conditions related to social networks. The configurations where the conditions are connected by a continuous line are referred to as substitute strategies, i.e., where the absence of social networks reporting is complemented by the presence of visual content, [C1, C4]. The other configurations, where the conditions are connected by dashed lines are either mixed strategies [C2, C3], i.e., strategies where the presence and absence of visual content are combined indistinctly, or unique strategies [C5, C6], i.e., strategies where the conditions are related only to social networks.

Fig. 1
figure 1

Configurations leading to crowdfunding success. Note: Conditions connected by continuous lines show configurations of substitute strategies; those connected by dashed lines show mixed and unique strategies. The symbol “~” indicates the absence (or negation) of a condition

Full size image

Substitute and unique strategies [C1, C4, C5, C6] slightly dominate the success of equity crowdfunding campaigns in terms of raw coverage. This suggests that the information disclosure strategies behind the success of an equity crowdfunding campaign either (i) compensate for the lack of reporting on social networks (i.e., Instagram, Facebook, Twitter) by reporting traditional information elements related to visual aspects (i.e., images and videos), or (ii) focus only on social networks. The substitutive nature of the conditions on visual content and social networks is explained below.

5.2 Information disclosure for overfunded campaigns

Figure 2 shows the different configurations of conditions that lead to overfunding, i.e., a campaign that raises at least of 10% more than the target funding, again distinguishing between (i) visual content and (ii) social networks. In two configurations, substitutability between visual content and social networks is observed, i.e., substitute strategies [C7, C8]. On the other hand, a unique strategy [C9] and a mixed strategy [C10] are found.

Fig. 2
figure 2

Configurations leading to overfunding. Note: Conditions connected by continuous lines show configurations of substitute strategies; those connected by dashed lines show mixed and unique strategies. The symbol “~” indicates the absence (or negation) of a condition

Full size image

Substitute strategies dominate [C7, C8], in terms of raw coverage and number of configurations. The presence of traditional visual content plays an important role both in crowdfunding success and overfunding, where the lack of social networks reporting is compensated by the presence of visual content in all substitute strategies.

5.3 Substitute, mixed and unique strategies

Four of the ten condition configurations are substitute strategies [C1, C4, C7, C8], two for crowdfunding success and two for overfunding. The lack of information disclosure on social networks can be compensated by the presence of traditional visual content. In this sense, there is a one-way substitutability. This has practical implications for the design of successful equity crowdfunding campaigns. Those entrepreneurs with less presence on social networks should make an effort to communicate the benefits of their projects through visual content, i.e. images and videos.

In turn, three condition configurations are mixed strategies [C2, C3, C10], two of which are conducive to crowdfunding success and the other to overfunding. Mixed strategies always include combinations of conditions (both in their absence and presence) regarding visual content and social networks, and no clear substitution effect can be found. The different substitute and mixed strategies can be seen in the 3 × 3 matrix in Fig. 3. Only three strategies [C5, C6, C9] are unique strategies considering only the social networks conditions.

Fig. 3
figure 3

Substitute and mixed strategies. Note: Boxed configurations with thin borders are information disclosure strategies that favor crowdfunding success; those with thick borders favor overfunding

Full size image

Overall, the substitute strategies are characterized by a clear pattern of combining the presence of visual content (i.e., a large number of images and/or a long video) and the absence of social networks [C1, C4, C7, C8]. The reverse pattern, i.e., the absence of visual content − only a few images and/or a short video − and the presence of social networks, does not occur in any case. This reinforces the central role of images and videos over social networks disclosure. Then, unique strategies also play an important role [C5, C6, C9], as they only include conditions related to social reporting, highlighting among their conditions the presence of a Twitter account and the absence of a Facebook account. Finally, mixed strategies highlight the heterogeneity inherent in the information disclosure strategies underlying the success and overfunding in the equity crowdfunding phenomenon [C2, C3, C10].

Following Fiss (2011), in the summary of sufficient conditions for crowdfunding success (Panel A) and overfunding (Panel B) shown in Table 6, a black circle (●) indicates the presence of a condition and a white circle (○) indicates the absence of a condition. In cases where neither the presence nor the absence of a given condition is reported, this means that this condition is irrelevant for the configuration in question.

Table 6 Summary of sufficient conditions
Full size table

6 Conclusions, theoretical and practical contributions, limitations and further research

6.1 Conclusions

This study aims to identify which elements of information disclosure related to (i) visual content, i.e., images and videos, and (ii) social networks, i.e., Instagram, Facebook, Twitter, are conducive to crowdfunding success and an overfunding of equity crowdfunding campaigns. The results of the Qualitative Comparative Analysis (QCA) performed allow deriving a total of ten information disclosure strategies, six of which are conducive to crowdfunding success and four of which are conducive to overfunding. In turn, these strategies are classified into: (i) substitute strategies, in which the absence of social networks is compensated by the presence of images and videos [C1, C4, C7, C8]; (ii) unique strategies, in which only the presence and absence of social networks are considered [C5, C6, C9]; and (iii) mixed strategies, combining the absence and presence of both groups of conditions [C2, C3, C10].

The results add new value by shedding light on the substitutability/complementarity of traditional visual content and social networks as information cues. Although the importance of information cues as signalers is consistent with the previous literature highlighting the central role of both images, videos, and social networks disclosure in influencing the crowd decision-making (e.g., Courtney et al. 2017; Li et al. 2016; or Kim et al. 2022), this research extends previous knowledge by examining in detail the heterogeneity of strategies that characterize both success and overfunding in an equity crowdfunding campaign. Through the use of QCA, multiple pathways leading to crowdfunding success and overfunding are explored, going beyond the study of unidirectional relationships.

6.2 Theoretical and practical contributions

At the theoretical level, this research contributes to a deeper understanding of those aspects of information disclosure that are likely to reduce the existing information asymmetries in digital financial environments characterized by risk and uncertainty and, consequently, that are conducive to crowdfunding success and overfunding. Thus, it extends the empirical evidence in this research area by (i) adopting a conceptual model that focuses on the visual content and social networks as signals in equity crowdfunding and (ii) using a configurational methodology. By using data from a relatively unexplored platform, despite its manifest importance in terms of volume of money intermediated, it enriches the results of the previous literature. In addition, the configurational approach of the study through Qualitative Comparative Analysis (QCA) enables access to specific nuances, since this methodology is based on equifinality and multifinality, so that the seemingly complex process of raising capital through equity crowdfunding can be approached more comprehensively.

On a practical level, a number of practical guidelines are suggested to fund-seeking entrepreneurs who want to use equity crowdfunding as a fundraising method. First, traditional visual content (images and videos) seems to have the ability to convey information about the characteristics of the campaign, thus reducing existing information asymmetries and bringing the entrepreneurial project closer to potential investors. In addition, this visual content has the ability to compensate for the lack of disclosure of social networks. Thus, images and videos remain central to the success and overfunding of equity crowdfunding campaigns.

Second, a number of information disclosure strategies suggest that the presence of social networks has sufficient signaling power to lead to the success and overfunding of equity crowdfunding campaigns. Finally, some strategies benefit from the interaction between the presence and absence of both sets of conditions (visual content and social networks).

In short, images, videos and social networks have an important signaling power when it comes to ensuring the success or overfunding of equity crowdfunding campaigns. This study highlights how these information cues can be used and examines their interplay.

6.3 Limitations

However, this study is limited by its sample size and its single data source, which restricts the generalizability of the results, as well as by its focus on visual content and social networks as information cues, while others may be relevant and have not been explored in this study. Regarding the former, the use of a single data source is due to the fact that the information disclosure process between the fund seeker and the crowd varies across platforms, requiring focusing on one of them (i.e., Startupxplore). However, future studies should delve into larger samples and confirm the findings in different platforms and contexts (e.g., geographic, normative, cultural). Regarding the information cues considered, this study does not pretend to include all the cues that may influence the crowd decision-making. In this sense, there may be other information elements used by the crowd to support its decision-making. The present study, based on the previous literature, focuses on the amount of visual content displayed and reported presence in social networks as relevant cues in determining success and overfunding in equity crowdfunding, as well as the interplay between both sets of cues. In addition, the study is limited by the current operationalization of the conditions, which emphasizes the amount of visual content and the presence of social networks but does not account for their specific characteristics. Despite these limitations, the study contributes to a better understanding of information disclosure strategies equity crowdfunding that lead to success and overfunding, where visual content and social networks are used as signals to reduce information asymmetries.

6.4 Further research

Further research should delve into the specifics of the different signals. If this study considers the volume of visual content as the number of images and seconds of video, future studies can examine the type of images, their size, and even what specific information is conveyed, whether it is information from the entrepreneurial team or an outline of the project, as well as the instrumental characteristics of the videos. This is an invitation to further explore the field of crowdfunding research through multidisciplinary approaches, from psychology to neuroscience, and to collect data through relatively new and unexplored techniques in this research field such as eye tracking (Buttice et al. 2022), speech analysis (Allison et al. 2022), or experiments (Bapna and Ganco 2021). Furthermore, future studies should analyze other dependent variables beyond the purely financial success variables, i.e., non-financial success indicators such as satisfaction, self-efficacy, ego-boost or brand awareness (Shneor and Vik 2020).

And beyond the scope of this paper, as suggested by Mazzocchini and Lucarelli (2022), research on equity crowdfunding should extend to different phases of the fundraising process, such as the choice of equity crowdfunding as a fundraising instrument by the entrepreneur(s), the suitability screening of campaigns by a given platform before they are made available to the public, or the future development of the crowdfunded initiatives and their long-term performance. Additionally, future research should address the micro, mezzo and macro levels of analysis in a balanced manner, reversing the trend that seems to prevail towards studies focused on the former (Shneor and Vik 2020).

Finally, research on crowdfunding has the potential to contribute to a wide range of research areas given the multidisciplinary nature of the implications derived from its findings. All this points to a promising field of study where fintech, entrepreneurship, innovation, digitization and the power of crowds converge.