Exploring entrepreneurial characteristics, motivations and behaviours in equity crowdfunding: some evidence from Italy

Purpose – This study explores entrepreneurial decision-making in the equity crowdfunding (ECF) context, and it aims to shed some light on the relationship among three aspects: entrepreneurial characteristics (i.e. entrepreneurial alertness and entrepreneurial self-efficacy), entrepreneurial motivations (i.e. promotion, improvement of networking and acquisition of product and market knowledge) and entrepreneurial behaviours (i.e. campaign characteristics in terms of communication and offerings). Design/methodology/approach – The hypotheses testing and analysis were conducted using the partial leastsquaresapproachtostructuralequationmodelling,anddatawerecollectedfromtheoverallpopulationofItalianECFplatforms. Findings – Our results highlight that entrepreneurial characteristics may be central in ECF because of their significant impact on some motivation entrepreneurs have to adopt ECF, which in turn have an impact on meaningful campaign characteristics that can influence ECF performance. Originality/value – The current literature is mainly focused on investors ’ decisions, while a neglected perspectiveuntilnowhasbeenthatofentrepreneurs.ThisstudyisamongthefirsttofocusonentrepreneursintheECFcontext,and,tothebestofourknowledge,itisthefirststudytoinvestigatetheentrepreneurialdecision-makingprocess.Theaddedvalueofthisresearchliesintheanalysisoftheentrepreneurialaspects underlyingentrepreneurialdecisionstouseECF.


Introduction
Decision-making is central to the success, longevity and survival of entrepreneurial activities (Caputo and Pellegrini, 2019). It is a well-established topic of interest in several fields, particularly management, but there is still a paucity of studies on entrepreneurial decisionmaking (EDM) (Shepherd et al., 2015), although EDM is deeply influenced by the evolution and turbulence of the modern global landscape (e.g. new ideas coming out of Fintech, digital platforms, iot or AI) (Cohen et al., 2017;Li et al., 2018;Agostini and Nosella, 2019;Balodi, 2019;Bullini Orlandi and Pierce, 2020). Given the emphasis on making decisions in uncertain contexts (Pellegrini and Ciappei, 2015), such as crowdfunding (Xu et al., 2015), equity crowdfunding (ECF) is emerging as an intriguing research area within the sphere of entrepreneurship (Cholakova and Clarysse, 2015;McKenny et al., 2017) and in particular EDM (Bruton et al., 2015;Shepherd and Patzelt, 2017). ECF is a fundraising system for entrepreneurs (Ahlers et al., 2015;Vulkan et al., 2016) to sell a specified amount of equity to several small investors through an open call on an Internet-based platform, and it helps entrepreneurs in catching opportunities and in developing or growing their business. affect the ECF performance. Finally, we looked at how these behaviours may predict ECF performance, i.e. the outcomes of the campaign.
The paper is structured as follows. In the next section, we provide the theoretical background, then we develop the research hypotheses. In the following section, we describe the research methodology used to conduct the study, which includes a description of data and variables. We then present and discuss the findings of our research and their theoretical and practical implications. In the last section we highlight the research limitations and give some suggestions for further research.

Theoretical background
It is well known that EDM influences strategy, and it requires strategic thinking (Vermeulen and Curseu, 2008;Yagnik and Chandra, 2019). Vermeulen and Curseu (2008, p. 16) highlighted the importance of a strategic EDM process and that it leads to the choice of stepwise activity and how these steps are executed to derive a probable desired solution. In this regard, Yagnik and Chandra (2019, p. 11) argued that "decision-making in itself is a strategic activity, which counts for determining a specific course of action to reach the desired strategic goals".
The EDM process is a complex phenomenon (Shane, 2000). Entrepreneurs have a unique "entrepreneurial mindset", prompting them to search for the best opportunities to pursue (McGrath and MacMillan, 2000;Sassetti et al., 2019). Entrepreneurial characteristics play a key role in identifying and exploiting opportunities (Ucbasaran et al., 2009). Even the entrepreneurs' motivations are considered a significant aspect. In fact, entrepreneurial motivations in EDM are meaningful for business success and survival (Baum and Locke, 2004). At the same time, different entrepreneurs will look at the same situation in different ways, making different evaluations (Shane, 2000;Pellegrini and Ciappei, 2015). Entrepreneurs have to find wisdom under blurred conditions (Pellegrini and Ciappei, 2015, p. 769) when discovering, or defining, a new opportunity (Shane, 2000;Welter and Alvarez, 2015;Bayon et al., 2016) and when they have to make choices (Rosenzweig, 2013).
In recent years, crowdfunding has become a strategic choice for entrepreneurs and a significant alternative to traditional financing systems (Ahlers et al., 2015;Xu et al., 2015;Vulkan et al., 2016). ECF has three main advantages for entrepreneurs: they can make an open call for investors online, limiting an otherwise time-consuming search process (Schwienbacher and Larralde, 2012); they can choose the amount of equity offered to limit their loss of ownership and control (Ahlers et al., 2015;Vismara, 2016) and they can sell to many small investorsi.e. the crowdinstead of few and big ones (Belleflamme et al., 2014;Ahlers et al., 2015).
A growing number of entrepreneurs have started to approach ECF strategically not only to get funds but also to engage the crowd to get publicity and contacts (Macht and Weatherston, 2014;Di Pietro et al., 2018). At the same time, entrepreneurs may exploit ECF to obtain other resources or to fill knowledge gaps (e.g. specific markets or products) (Belleflamme et al., 2014;Schwienbacher, 2018).
Crowdfunding also represents a form of crowdsourcing (Lambert and Schwienbacher, 2010;Schwienbacher and Larralde, 2012;Liu et al., 2016;Di Pietro et al., 2018), where entrepreneurs obtain feedback from the crowd, which can help to develop or test new product/service ideas prior to launch. A large crowd of investors enables entrepreneurs to benefit from the "wisdom of crowds"i.e. the collective skills and knowledge of the investors (Belleflamme et al., 2014;Polzin et al., 2018) giving them useful insights.
ECF may be particularly helpful for entrepreneurs without a track record as they are less attractive to traditional investors (e.g. banks, VCs or BAs) (Collins and Pierrakis, 2012).

Exploring entrepreneurial characteristics in ECF
The crowd creates hype around a business, generates public exposure for it and its products/ services and raises awareness (Lambert and Schwienbacher, 2010;Belleflamme et al., 2014;Di Pietro et al., 2018). This may result in more exposure to more investors, facilitating future access to further capital.
In the last few years, ECF has been studied by several scholars. Most of these studies have focused on the drivers of campaigns' success in terms of equity sold and investors involved (Ahlers et al., 2015;Cholakova and Clarysse, 2015;Lukkarinen et al., 2016;Vismara, 2016Vismara, , 2018Block et al., 2018b;Polzin et al., 2018;Stevenson et al., 2019); other studies have focused on the choice of determinant campaign characteristics and how they influence investor decisions and, thus, the success of campaigns (Ahlers et al., 2015;Lukkarinen et al., 2016;Vismara, 2016;Vulkan et al., 2016;Block et al., 2018b). Feola et al. (2019) specifically explored the Italian ECF scenario and suggested that ECF investors are not similar; in fact, they found four distinct types of investors, characterised by different behaviours in their investment actions ("venture trustful", "crowdfunding technicians", "financial investors, talent scouters" and "social dreamers"). The scholars suggested that entrepreneurs should adopt a marketing perspective in raising capital for their companies and segment potential investors by leveraging specific strategies "aimed to promote investment opportunities to match the diverse investor segments" (Feola et al., 2019, p. 13).
Entrepreneurial motivations drive the choice of campaign characteristics to meet specific goals (Di Pietro et al., 2018;Estrin et al., 2018). Only a few studies have considered that ECF is more than a fundraising tool (Di Pietro et al., 2018;Estrin et al., 2018;Wald et al., 2019). ECF platforms allow the provision of resources, knowledge and services that have become accessible to hundreds of equity funders drawn from a public crowd (Di Pietro et al., 2018), while past research (Gormon and Sahlman, 1989;MacMillan et al., 1989;Macht and Robinson, 2009) shows that these elements derive from the intense private relationships between entrepreneurs and a limited number of experienced professional investors (e.g. VCs and BAs).
The entrepreneurs' perspective and their behaviour before a campaign launch are still not fully studied. Different entrepreneurs may have different characteristics and motivations. Their characteristics affect their decisions whether or not to engage in new contexts (such as an open innovation (OI) context), search for innovations or strategically approach new mechanisms to obtain further resources or input from the crowd by leveraging external knowledge and thus benefiting from external sources of innovation (Ucbasaran et al., 2008;Tang et al., 2012;Du et al., 2016;Di Pietro et al., 2018).

Hypotheses development
Entrepreneurial alertness EA is central in the EDM process. EA, according to Kirzner (1973Kirzner ( , 1979, is the entrepreneur's ability to identify opportunities usually overlooked by others. It is a key feature of the emerging entrepreneurial mindset (Obschonka et al., 2017). Entrepreneurs are open to opportunities (Renko et al., 2012), they are more alert to new entrepreneurial ventures (Busenitz, 1996;Tang et al., 2012) and they are more effective in acknowledging, and exploiting, new opportunities (Shane and Venkataraman, 2000;Maine et al., 2015;Gaglio and Winter, 2017a, b;Neneh, 2019;Sharma, 2019). This opportunity recognition (Wang et al., 2013) builds on a state of heightened alertness to new information, and entrepreneurs pay constant attention to the environment (Lumpkin and Lichtenstein, 2005;Valliere, 2013). Entrepreneurs exhibit high alertness by investigating new opportunities through a continuous search for information and a broad scanning at unconventional times and places (Kaish and Gilad, 1991;Gaglio and Katz, 2001). Ardichvili et al. (2003, p. 105) defined EA as "a necessary condition for the success of the opportunity identification triad: recognition, development, and evaluation". MD 59,5 Tang et al. (2012) suggested that alert entrepreneurs are likely to increase their companies' innovations since they perceive trends and their consequences more accurately than nonalert ones (Gaglio and Katz, 2001); they are more effective at exploiting external resources and at pursuing new opportunities (Van Den Bosch et al., 1999;Gray, 2006;Patel, 2019;Chen et al., 2020). Roundy et al. (2018) suggested that if decision-makers are entrepreneurially alert they are more ready to make strategic change decisions that lead them to subsequent entrepreneurial opportunities. The scholars argue that EA is influential because alert entrepreneurs make rapid and nimble decisions and are, therefore, more likely to earn first-mover advantages.
Furthermore, EA leads entrepreneurs to be more open and to leverage new tools (Busenitz, 1996;Tang et al., 2012). OI platforms are among the most relevant mechanisms; in fact, more and more entrepreneurs engage in OI contexts (Chesbrough, 2003;Reed et al., 2012;Hsieh et al., 2016;Gomezel and Rangus, 2018) in order to increase opportunities for their firms and their innovations (Bayon et al., 2016). In this vein, OI practices are useful mechanisms for entrepreneurs to leverage external knowledge and to accelerate internal innovation and commercialisation opportunities (Chesbrough, 2003;Du et al., 2016). Di Pietro et al. (2018) held that ECF platforms are an important example of OI platforms because entrepreneurs can get both financial resources and support from the wisdom of the crowd; at the same time, the scholars highlighted that entrepreneurs' characteristics influence their motivation to resort to an OI tool like ECF.
Moreover, several studies in the EDM literature have suggested that EA is most salient when decision-makers must respond to a specific scenario or to a disruption in their environment (Hitt et al., 2001;Roundy et al., 2018). In the last few years, ECF has represented a disruptive innovation in funding markets as it shifts the focus to the crowd (Belleflamme et al., 2014;Ahlers et al., 2015;Bruton et al., 2015). Accordingly, many entrepreneurs have started using ECF for their projects as an alternative solution to traditional funding (Kahle and Stulz, 2013;Bruton et al., 2015;Vulkan et al., 2016;Block et al., 2018a).
In line with prior studies, we suggest that EA allows for better entrepreneurial opportunities. In line with this, EA leads entrepreneurs to search for new resources from a new system like ECF and get superior performance. In particular, in ECF a high level of EA should influence the motivation to launch an online call, and it should drive entrepreneurs to search for new useful resources (not only the financial ones) by engaging the crowd. Accordingly, we define our first hypothesis as follows: H1. EA has positive effects on entrepreneurial motivations to resort to ECF.
Entrepreneurs adopt ECF not only to get financial resources but also to leverage other, more relational resources useful in seizing new opportunities (Ebbers, 2014). Pryor et al. (2016, p. 27) argued that entrepreneurs recognise actual "opportunities and find solutions by defining feedback-seeking and evaluation behaviours".
Alert entrepreneurs will participate in ECF to get several benefits by leveraging external sources from the crowd. EA will lead entrepreneurs to search for product co-creation activitiese.g. getting feedback or suggestions from the crowd to develop more customised products/services or involving investors as testersor receive market advicei.e. getting insights from the crowd about market trends, potential competitors or partners and so on (Di Pietro et al., 2018).
Alert entrepreneurs are more likely to search for more connections (Adomako et al., 2018) and exploit a crowd-network (Di Pietro et al., 2018). Active participation in the ECF mechanism gives entrepreneurs the opportunity to grow their networks since the investors will help build the company's network, creating new connections with external stakeholders (Di Pietro et al., 2018). Moreover, a high number of connections positively increase the probability of receiving outside financing (Vismara, 2016).

Exploring entrepreneurial characteristics in ECF
Finally, EA induces entrepreneurs to improve the promotion of their business and search for external resources to foster it. Thus, entrepreneurs leverage a crowd's contacts to promote their businesses and improve word of mouth. Furthermore, they enhance their company's external visibility through crowd ambassadors, who increase public awareness by acting as marketing channels (Di Pietro et al., 2018).
To understand these different effects, in this study we investigated the relationship between EA and each of the three categories of entrepreneurial motivation -PMK, NET and PRand have therefore defined the following sub-hypotheses: H1a. EA has a positive effect on searching for PMK from an ECF campaign.
H1b. EA has a positive effect on searching for NET from an ECF campaign.
H1c. EA has a positive effect on searching for PR from an ECF campaign.
Entrepreneurial self-efficacy ESE is a specific form of self-efficacy (Chen et al., 1998) affecting an entrepreneur's choice of activities, goals and performance (Zhao et al., 2005). It has been studied in many fields, particularly social psychology, to explain human behaviours, motivations and performance (Gist and Mitchell, 1992;Bandura, 1997;Mauer et al., 2017), and it plays an influential role in determining choices, level of effort and perseverance (Chen et al., 2004).
ESE has been seen as a predictor of entrepreneurial intentions and a useful element in driving toward creating new entrepreneurial activities (Zhao et al., 2005;Barbosa et al., 2007;McGee et al., 2009). Chen et al. (1998 reported that "ESE refers to the strength of a person's belief that he or she is capable of successfully performing the various roles and tasks of entrepreneurship. It consists of five factors: marketing, innovation, management, risktaking, and financial control". ESE is linked to how an entrepreneur perceives his or her own capability to accomplish a task (Bandura, 1986;Lindsley et al., 1995); it produces its effects through motivational processes (Bandura, 1977). Chen et al. (1998) suggested that ESE can be used to identify several reasons for entrepreneurial avoidance, i.e. not engaging in specific contexts or mechanisms. Forbes (2005) argued that the level of ESE affects the extent of entrepreneurs' comprehensive decision-making. Entrepreneurs with a high level of ESE do not search for new input, such as for marketing or innovation.
ESE leads to optimism, and entrepreneurs with high ESE hold strong beliefs in their entrepreneurial abilities (Krueger and Brazeal, 1994;Chen et al., 1998;Ardichvili et al., 2003;Forbes, 2005). Accordingly, we assume that high levels of ESE will have a negative impact on entrepreneurial motivation to use ECF as a way to get new (but not financial) resourcesthe entrepreneur's confidence in his capabilities will hinder the opportunity identification process and work against the search for new inputs (Zhao et al., 2005;Drnov sek et al., 2010). Accordingly, we define our second hypothesis as follows: H2. ESE has a negative impact on entrepreneurial motivations to resort to ECF.
Entrepreneurs with high ESE do not aim to leverage an OI mechanism like ECF to get crowd input since they do not consider them of primary importance and prefer to rely on their own knowledge (Ucbasaran et al., 2008). We suggest that the higher the ESE level, the lower the MD 59,5 level of ECF use by the entrepreneur to obtain new crowd input on each of the three dimensions of entrepreneurial motivation (PMK, NET and PR). A high level of ESE, in fact, could lead entrepreneurs to avoid asking the crowd for new knowledge on products or markets, to avoid leveraging funders to grow their networks and to not exploit them in promoting the company and its products/services. Accordingly, we define three subhypotheses as follows: H2a. ESE has a negative effect on searching for PMK from an ECF campaign.
H2b. ESE has a negative effect on searching for NET from an ECF campaign.
H2c. ESE has a negative effect on searching for PR from an ECF campaign.
Entrepreneurial motivations to use ECF Several studies have explored investors' motivations to use ECF (Ahlers et al., 2015;Cholakova and Clarysse, 2015;Block et al., 2018b;Polzin et al., 2018), while entrepreneurs' ones are still less investigated (Mochkabadi and Volkmann, 2018). According to several scholars (Belleflamme et al., 2014;Estrin et al., 2018), even if the main reason to use ECF is to get financial resources, there are other motivations, such as getting known, getting feedback and several other advantages to help company growth and development. These studies have shown that ECF allows entrepreneurs to test their products, develop their brand and build a loyal customer base, while turning customers into investors. Di Pietro et al. (2018) studied ECF platforms as OI platforms and found that those companies exploiting the crowd networkto get input on products, strategies and other market knowledgewere more successful than the others.
Wald et al. (2019) examined non-financial benefits of ECF campaigns and found that the crowd could offer entrepreneurs both inward benefits (investors' experiences and expertise) and outward benefits (public exposure and investor recruitment).
In this study we propose that entrepreneurs leverage ECF as a way to get access to various useful resources (Barney, 2001;Meyer, 2019). Lukkarinen et al. (2016) argued that as a campaign's characteristics can be predetermined by the entrepreneur prior to the campaign, they are related to entrepreneurs' motivations and goals. Accordingly, entrepreneurs disclose and define the campaign characteristics based on their specific motivations, so we define our third hypothesis as follows: H3. Entrepreneurial motivations affect the choice of campaign characteristics.
Consistent with the previous literature (Di Pietro et al., 2018;Estrin et al., 2018), we focus on several key parameters: the acquisition of new knowledge on products and markets (e.g. feedback on products or internationalisation strategies), the improvement of networking resources (to become more connected with other relevant stakeholders or improve the usefulness of their contacts) and more effective business promotion (leveraging the crowd as business ambassadors or as a source of positive word of mouth). We assume that entrepreneurial motivations determine ECF campaign characteristics both in terms of CCC (social networks, images and updates) and COC (equity offered and funding target).
We hypothesise that non-financial motivations negatively influence COC, as entrepreneurs set a lower target (i.e. equity offered and funding target) to engage a large number of small investors (Vismara, 2016). Accordingly, we define the following three subhypotheses: H3a. PMK motivations have a negative effect on COC.
H3b. NET motivations have a negative effect on COC.

Exploring entrepreneurial characteristics in ECF
H3c. PR motivations have a negative effect on COC.
However, we expect a positive impact for CCC, as PMK, NET and PR lead entrepreneurs to post more information, more updates and/or more images and the links to their social network profiles to deeply engage the crowd (Bretschneider and Leimeister, 2017;Block et al., 2018b). Entrepreneurs leverage several tools and disclose some specific elements useful to improve crowd knowledge (Ahlers et al., 2015;Vismara, 2016;Block et al., 2018b). Therefore, we propose the following three sub-hypotheses: H3d. PMK motivations have a positive effect on CCC.
H3e. NET motivations have a positive effect on CCC.
H3f. PR motivations have a positive effect on CCC.

Campaign characteristics
In the academic literature, several campaign characteristics have been used as predictors of ECF success (Ahlers et al., 2015;Lukkarinen et al., 2016;Vismara, 2016;Vulkan et al., 2016;Block et al., 2018b;Polzin et al., 2018). Among them, Vulkan et al. (2016) compared ECF and reward crowdfunding, while Lukkarinen et al. (2017) suggested that the crowd does not adopt the same investment decision criteria used by VCs or BAs. This latter study also showed that ECF performance is related to pre-selected crowdfunding campaign characteristics and the utilisation of private and public networks selected by entrepreneurs. The choice of campaign characteristics is a typical behaviour of entrepreneurs, and it plays a key role in increasing the probability of a successful ECF campaign (Vismara, 2016). In particular, entrepreneurs decide how much information about their business communicates and the media to adopt in order to satisfy the crowd's needs and attract them to the campaign. These choices are fundamental for entrepreneurs to communicate the quality of their business to investors and provide them with credible signals (Ahlers et al., 2015;Vismara, 2016). Choosing the right campaign characteristics helps to mitigate information asymmetries, and it improves investors' knowledge of the business and their willingness to invest in it (Ahlers et al., 2015). Accordingly, we have defined our fourth hypothesis as follows: H4. The campaign characteristics chosen by the entrepreneur affect the ECF performance.
In this study, we consider both CCC and COC. The choice of the campaign characteristics is strongly related to the decision-making process of the entrepreneur (Belleflamme et al., 2014;Lukkarinen et al., 2016). Entrepreneurs, in fact, are free to decide the type of CCC and COC to disclose as he or she tries to attract investors (Belleflamme et al., 2014;Vismara, 2016). According to Dorfleitner et al. (2018, p. 524), CCCs are typically "entrepreneurial behaviours in crowdfunding campaigns" and "this strategic communication behaviour can help investors to optimise their investment decisions".
The success of campaigns depends significantly on how the campaign is run and on the decisions of the entrepreneurs (Lukkarinen et al., 2016, p. 36). Entrepreneurs decide to send specific types of information or signals (Herbig, 1996) to investors and decide what they want to reveal to them (Belleflamme et al., 2014;Ahlers et al., 2015;Vismara, 2016). In this vein, entrepreneurs should strategically decide the factors he or she discloses within the campaigns, i.e. CCC and COC. The behaviour of decision-makers plays a key role in the online context of crowdfunding platforms (Koch and Siering, 2019). Entrepreneurs' decision-making behaviour is a highly relevant driver of ECF campaigns, able to affect their performance.
The first category, i.e. CCC, is focused on presenting the business and its products/ services, as well as on the presenting the entrepreneurs. Entrepreneurs may leverage several tools for CCC, such as social networks (e.g. Facebook and LinkedIn), images and updates, to influence investors (Mollick, 2014;Chan and Park, 2015;Block et al., 2018b).
The choice of CCCs is an entrepreneurial decision on the type and number of images or updates to post on online campaigns, as well as the type of social networks to leverage, such as LinkedIn, Facebook and so on. The choice of these parameters represents a specific development strategy. Social networks are strategic tools for crowd engagement (Sashi, 2012), and entrepreneurs have to manage several social networks in order to tap into a wide range of people (Keegan and Rowley, 2017). They are well-known elements that help entrepreneurs to spread information (Mollick, 2014;Colombo, 2015), overcome information asymmetries and obtain superior performance in ECF (Vismara, 2016). In fact, prior studies have highlighted that entrepreneurs' connections on social networks are positively related to funding collected and investors attracted.
In crowdfunding, updates are a way to let entrepreneurs disclose specific types of information during the campaign. Dorfleitner et al. (2018, p. 523) argued that entrepreneurs "can voluntarily communicate with their investors by posting updates" and "entrepreneurs use voluntary disclosure strategically". Entrepreneurs strategically engage in updates in order to inform the crowd about new business developments. Updates are an added value for investors, and they positively influence crowd participation. Mollick (2014) found that the number of updates and their timing are a proxy of campaign success.
Entrepreneurs also determine the type and the number of images to include in their initiatives or business plans to display their perspective on their products and/or services (Chan and Park, 2015). Images are useful in attracting investors and influence people's cognition and online behaviour. In fact, accessible and relevant images (e.g. products, business activities or prototypes) affect investment screening decisions and increase favourable judgments.
The second category, i.e. COC, includes both the equity offered and funding targets pre-determined by entrepreneurs. They choose the amount of equity offered, i.e. how much equity shares they intend to sell (Ahlers et al., 2015;Vismara, 2016), and the campaign's goal, i.e. how much funds they aim to raise (Lukkarinen et al., 2016;Vulkan et al., 2016). These specific parameters level is part of the entrepreneurs' strategic decision-making process and they may influence the business strategy (Francioni et al., 2015). Entrepreneurs decide if and how much to invest in their initiatives through equity retention (Leland and Pyle, 1977;Ahlers et al., 2015;Vismara, 2016). The amount of equity offered is related to their decision about the control and ownership of their companies (Macht and Weatherston, 2014). Prior studies (Ahlers et al., 2015;Vismara, 2016Vismara, , 2018Vulkan et al., 2016) have shown that a good indicator of entrepreneur's degree of commitment is the equity offered. These studies showed negative impacts of the amount of equity offered on ECF performance (Ahlers et al., 2015;Vismara, 2016Vismara, , 2018Vulkan et al., 2016), as it is usually seen as a negative signal as it highlights a lower commitment by entrepreneurs (Vismara, 2016), that they are less optimistic and are not confident about their company's capability to generate positive cash flow (Ahlers et al., 2015;Vismara, 2016). Contrariwise, equity retention is a positive signal of entrepreneurs' commitment (Leland and Pyle, 1977). As Vismara (2016, p. 583) argued, entrepreneurs signal their commitment through high ownership retention, thus "potential investors will perceive this behaviour as a quality signal." Entrepreneurs who engage in ECF also decide the funding target in online campaigns and indicate a specific amount to collect. They, in fact, set a goal for minimum target funding (Lukkarinen et al., 2016) that is related to their funding needs and their strategic decisions about their company's development.
The funding target is another indicator defined by the entrepreneur (Lukkarinen et al., 2016;Vulkan et al., 2016); the higher it gets the more difficult it is for the campaign to be Exploring entrepreneurial characteristics in ECF successful. A higher funding target is negatively related to a campaign's success (Mollick, 2014;Zheng et al., 2014); in fact, it represents a limit for several projects aiming for a high target and having more difficulties collecting the needed funding. This is an even more relevant constraint on platforms operating with the AON (all or nothing) model, like most ECF ones, where companies do not keep any of the pledged funds if they do not meet their funding goal.
In line with prior studies, we expected to find a positive relationship between CCC and ECF performance since a higher number of updates, images and social connections should predict the success of campaigns.
Contrariwiseconsistent with previous studiespre-selected COC should negatively affect ECF performance since both a larger percentage of equity offered by entrepreneurs and a higher funding target will reduce the likelihood of a campaign's success. Hence, the literature leads to our fourth hypotheses, as follows: H4a. CCC has a positive effect on ECF performance.
H4b. COC has a negative effect on ECF performance.
The following Figure 1 reports the final structure of the presented hypotheses with the constructs and the hypotheses.

Research method Context
This study focuses on ECF in Italy, the first country in Europe to define a specific regulation for ECF (Decreto Legge n. 179/2012 or Decreto Crescita Bis); moreover, Consob (Commissione Nazionale per le Societ a e la Borsa) created a national registry for ECF operators (Vismara, 2016;Rossi and Vismara, 2018). The Italian ECF market is rapidly growing, in terms of collected funding (more than 82.3 million euros (as of June 2019, 30 th ), campaigns launched (over 400) and their success rate (about 55%) (Politecnico di Milano, 2019). The Italian context is a vibrant case for ECF, and its policy-makers are vigilant in improving ECF regulations (e.g. a new regulation requires entrepreneurs to sell a minimum 5% of equity to professional investors, banks or innovative start-up incubators). Several studies have focused on ECF in Italy, such as Piva and Rossi-Lamastra (2018), who examined the effects of human capital signals on entrepreneurs' success, and Feola et al. (2019), who segmented the Italian ECF investors' market to investigate differences between segments and explore the investors' drivers when selecting investment proposals. These reasons led us to focus on Italy, a growing ecosystem rich in ECF innovation. Data collection and sample In the public national registry for ECF operators, we found 33 authorised ECF portals (Politecnico di Milano, 2019), though only 12 were eligible for this study as they were active and not specialised in real estate. We collected data on all the campaigns that ended before September 2019. The related data are shown in Table 1.
We used an online survey to collect data on entrepreneurs' EA, ESE and motivations to use ECF, using scales previously validated in the academic literature. A pilot version of the survey was administered to a small number of ECF entrepreneurs to collect their feedback on each single measure, the questions' wording and their familiarity with the survey topics (Podsakoff et al., 2012). This feedback drove us to group the motivations into three main dimensions: PMK, NET and PR.
We administered the survey using both emails and LinkedIn, and the data collection lasted eight weeks, from July to September, with two reminders sent to increase the number of participants.
In total, we surveyed 368 companies on the 12 valid ECF platforms; the questionnaire was answered by 122 entrepreneurs (33.15% answer rate), but we had to delete 25 questionnaires as they were incomplete. Thus, we had a final sample size of 97 companies (26.36% of the companies -97 out of 368). The survey sample included both successful (81%) and failed projects (19%). About 77% of companies were start-ups and 23% were small-and mediumsized enterprises (SMEs), while about 65% belonged to the service industry. We made a data check through official business register databases (e.g. Register of Innovative Start-ups/ SMEs) and the major search engine and start-up information sites (crunchbase, AngelList etc.) to better identify the type (whether a start-up or an SME) and sector (according to the Ateco classification) of each company.
The data on each campaign characteristic and ECF performance were hand-collected. In particular, we collected information about the COC (such as the amount of equity offered and target capital), the CCC (such as updates, images or links to their social network profiles) and the ECF outcomes (such as success of fundraising, the final percentage of funding collected, the number of final investors involved and the final capital collected, in euros).
To check for potential non-response bias, we conducted a wave analysis (Zollo et al., 2017(Zollo et al., , 2019 comparing the respondents in the three periods on several variables relevant in the literature and on the following performance variables: start-up status and type of company (service or not). The t-tests found no statistically significant differences between the three groups; for example, for the type of company (i.e. the proportion of service companies in the group) the p-value between the first and the second wave was 0.58, between the first and the ECF in ECF third wave it was 0.54 and between the first and the second it was 0.95. Accordingly, the nonrespondent bias was not an issue in our data sampling.

Measures
In order to measure EA and ESE, we used the scales previously validated by Tang et al. (2012) and Zhao et al. (2005), respectively. The EA scale is divided into three dimensions for a total of 13 items (six for scanning and searching for information (SS), three for association and connection (AC) and four for evaluation and judgement (EJ)) (Tang et al., 2012). According to Tang et al. (2012), SS improves the entrepreneurs' knowledge (Dimov and Shepherd, 2005) and offers them multiple possibilities (Kirzner, 1979); it favours their investigations (Busenitz, 1996). More extensive SS helps in enhancing alertness to business opportunities (Ericsson et al., 1993). AC has a key role in allowing entrepreneurs to move out of the routine and innovate (Gaglio and Katz, 2001). It is the dimension used to receive and process new information to make unique connections. EJ refers to the identification of profitable business opportunities and judgement of the potential of new information to match it with existing ideas or prototypes (Baron, 2006). The judgments of these opportunities and their related actions represent a central element of EA (McMullen and Shepherd, 2006). In this study we adopted this tri-partition and assumed that the three dimensions have a key role in the EDM process.
The ESE scale was developed by Zhao et al. (2005), identifying four central items (new business opportunities, creating new products, thinking creatively and commercialising an idea or new development). The entrepreneurial motivation construct, based on previous studies (Di Pietro et al., 2018;Estrin et al., 2018), evaluated the relative importance of several potential inputs as a valid motivation to use ECF. All the items were measured on a five-point Likert Scale. Because the original scales were in English, we followed standard translation and back-translation procedures to get the Italian versions (Saunders and Lewis, 2012).
In order to reduce the method biases (Podssakof et al., 2012), we collected the campaign characteristics directly from the ECF platforms instead of asking them of the entrepreneurs. We collected the number of updates (Mollick, 2014;Block et al., 2018b;Dorfleitner et al., 2018), the number of images (Chan and Park, 2015; Bretschneider and Leimeister, 2017) and the entrepreneurs' social networks presence (Mollick, 2014;Vismara, 2016), as indicators related to CCC. We also measured COC using the equity offered in percentage and the target amount of funding (Ahlers et al., 2015;Lukkarinen et al., 2016;Vismara, 2016Vismara, , 2018Vulkan et al., 2016).
Finally, we measured ECF performance using three commonly used proxies: the final percentage of funding collected, the final number of investors and the total funds collected (Ahlers et al., 2015;Lukkarinen et al., 2016;Vismara, 2016Vismara, , 2018. The final percentage of funding collected was a more fine-tuned measure of ECF success when the fundraising exceeded the target goal, and it may have measured its failure when the fundraising did not reach the target goal (Vismara, 2016(Vismara, , 2018. The final number of investors was a count variable and indicated the crowd participation in terms of numbers of investors involved at the end of a campaign (Ahlers et al., 2015;Vismara, 2016). This was an important measure of success, as the main logic to adopt the crowdfunding is to accumulate a large number of investors, a crowd (Vismara, 2016). The last variable, funds collected, indicated the total funding that was generated by a project (in thousands of euros) (Ahlers et al., 2015;Vismara, 2018). It is a common proxy of ECF success and measured the amount of capital raised at the end of a campaign.
The constructs and the related items are reported in Table 2.

Data analysis
We tested our model using PLS-SEM (Hair et al., 2011) with R-CRAN with the package plspm (Sanchez et al., 2017), for the model evaluation, and with the seminr package for the related MD 59,5 tests (Ray et al., 2019). PLS-SEM has already been used in several studies on strategic management (Hulland, 1999;Sarkar et al., 2001), entrepreneurial orientation (Shehu and Mahmood, 2014;Pratono and Mahmood, 2015), digital platforms (Cenamor et al., 2019) and, more specifically, crowdfunding (He et al., 2016;Zheng et al., 2017;Sahaym et al., 2019). The choice to adopt this particular approach to SEM, instead of the traditional covariancebased one, was also supported by several theoretical motivations. Our study is among the first to study how entrepreneurs' characteristics and behaviour is linked to campaigns' characteristics and success, so its purpose is mostly exploratory (Hair et al., 2014); moreover, our focus was on the relationship between the constructs, not on the best model to represent a given reality (Hair et al., 2011). Finally, in our case the PLS-SEM approach was the suggested approach as it did not rely on any distributional assumption of the measured variables Reinartz et al., 2009).
The model was tested for common method bias (CMB) with the full-collinearity approach (Kock and Lynn, 2012;Kock, 2015). According to this method, there is no significant risk of CMB if the latent VIFs are lower than the suggested limit of 3.3. In our case we found no latent VIF higher than the suggested limit, and the highest value was 2.05 for the NET construct; it followed that CMB was unlikely to be a significant threat in our case.
At the same time, the evaluation of PLS-SEM did not rely on any distributional assumption of the variables, and it did not evaluate a global fit measure to assess the model validity; so several authors (Chin, 1998;Henseler et al., 2009;Hair et al., 2016;Ravand and Baghaei, 2016) hold the need to study the data using a two-step approach: (1) quality of the

Measurement model
The assessment of the measurement model for reflective indicators in PLS was done by looking at four indicators: (1) Indicator reliability -Items' factor loading on their latent higher than 0.6 (Chin, 1998;Henseler et al., 2009).
(4) Discriminant validity -Items factor loading on their latent variable higher than the cross-loadings on the other constructs (Ravand and Baghaei, 2016).
In our case some items had a factor loading lower than 0.6 (namely SS1, SS3, SS4, ESE4, NET2, PR3 and Y3) and, given the exploratory nature of our study, we decided to delete them and re-run the test (Hulland, 1999;Hair et al., 2016). As shown in Tables 3 and 4, this second run passed the four tests.

Hypotheses testing
To assess the quality of the structural model, we looked (Hair et al., 2016) at the structural path coefficients defined with a bootstrap with 5,000 resamples and confirmed the related predicting power of the constructs using R 2 . The related data are reported in Table 5 and Figure 2. We found support for several of our hypotheses (namely H1b, H1c, H2a, H2b, H3a, H3e, H3f, H4a and H4b). We found support for H1b and H1c showing that EA drives entrepreneurs to use ECF both to create a bigger network of relationships (factor loading 5 0.187*) and to exploit these new connections in promoting their business (factor loading 5 0.261*).
We found support for the negative relationship between ESE and both the motivation to leverage investors' feedback to improve products or strategies (factor loading 5 À0.229**) and to exploit the new relationship to widen the network the enterprise is embedded into (factor loading 5 À0.253**).
Looking at the direct effects of entrepreneurs' motivations on campaign characteristics, we found support for the hypotheses linking PMK to COC (factor loading 5 À0.176*) and PR to CCC (factor loading 5 0.307*), while NET had a significantly negative impact on CCC (factor loading 5 À0.348**), though we expected to find a positive one.
Finally, looking at the impact of campaigns' characteristics on ECF performance, we found a negative impact of COC on ECF performance (factor loading 5 À0.210**) and a positive one for CCC (factor loading 5 0.484***), which confirms the related literature.
At the same time, we have found our model only had a limited predicting power for the different campaign characteristics (the R 2 of all these variables was lower than 0.1), but it showed a moderate predicting power for the campaign performance (R 2 of PERF is 0.32).

Discussion
EDM processes in the ECF field are still largely unexplored, and the perspective of entrepreneurs is still not fully understood (Mochkabadi and Volkmann, 2018). In this paper, MD 59,5 we looked into EDM to improve the current knowledge on this topic, and we have tried to shed some light on the importance of entrepreneurial characteristics, motivations and behaviours in this field. This paper contributes to the current literature on EDM and ECF.
Recently, there has been increasing interest in understanding ECF as both its volume and the number of entrepreneurs who use it are growing exponentially (Massolution, 2016;Technavio, 2018;Politecnico di Milano, 2019). Existing studies have enhanced our understanding of ECF performance and investors (Ahlers et al., 2015;Lukkarinen et al., 2016;Vismara, 2016Vismara, , 2018Vulkan et al., 2016;Block et al., 2018b;Polzin et al., 2018;Feola et al., 2019), yet the research on entrepreneurs is at its infancy; further research has been suggested to develop a solid understanding about EDM in this specific context and how entrepreneurs use this new system.

Theoretical implications
Our study focused on entrepreneurs and, therefore, we looked at entrepreneurial characteristics underlying EDM. We distinguished two well-known types of entrepreneurial characteristics, EA and ESE. Then we focused on entrepreneurial motivations in terms of PMK, PR and NET. Finally, we looked at two classes of campaign characteristics (COC and CCC).
Our findings highlight that both entrepreneurial characteristics have a significant impact on motivations in adopting ECF. EA positively influences both NET and PR, confirming, at  Vismara, 2016;Di Pietro et al., 2018). This agrees with some scholars (Gaglio and Katz, 2001;Tang et al., 2012) who hold that EA allows entrepreneurs to identify new opportunities (Gaglio and Winter, 2017a;Roundy et al., 2018;Patel, 2019), which leads them to leverage the crowd to improve their networking capabilities (Ebbers, 2014;Adomako et al., 2018). This is done not only to get access to more networks but also to engage the crowd in promoting their business, further helping them gain access to new relationships. Our model supports the idea that EA allows entrepreneurs to be more sensitive and vigilant to the environment and drives them to interact with other players to discover new opportunities (Ardichvili et al., 2003). Our data support the idea that ESE, however, reduces the perceived need for external inputs, not only for their products or the markets they are targeting (as expected from several studies) (Zhao et al., 2005;Drnov sek et al., 2010) but also for their networking capabilities. This highlights the idea that the more entrepreneurs are focused on their capabilities the less they will look for talent from the outside.
Moreover, our study suggests that both the entrepreneurial characteristics we have considered are needed to explain the three classes of entrepreneurial motivations, even if the low R 2 highlights the need to study these topics more.
Our results show that an entrepreneur's motivation can have a meaningful impact on the characteristics the entrepreneur chooses for their ECF campaign. As expected, when the entrepreneur starts a campaign to get new information or feedback on their products and/or the target market, they will lower their COC (Vismara, 2016;Di Pietro et al., 2018;Estrin et al., 2018). At the same time, we found no evidence of these motivations' effects on the CCC, as was expected from the existing literature (Bretschneider and Leimeister, 2017;Block et al., 2018b).
Our analysis does not support the idea that entrepreneurs wanting to expand their network will lower some COCs (we were expecting a positive effect of these motivations on their CCC) (Brown et al., 2019). According to some authors (Belleflamme et al., 2014;Vismara, 2016), the crowd is mostly composed of amateurs with limited experience who are only useful for increasing word of mouth (Cholakova and Clarysse, 2015;Vismara, 2018). These findings agree with some scholars (Ahlers et al., 2015;Lukkarinen et al., 2016;Vismara, 2016) who hold that when entrepreneurs prefer to attract more experienced investors, they will try to leverage the funding target and equity offered as signs of their commitment.
This study's framework highlights, in support of previous findings by several authors (Mollick, 2014, Block et al., 2108b, that entrepreneurs looking to exploit ECF as a way to promote their business or their business idea will increase the CCC of their campaign (such as adding more photos or updating the campaign more often) (Wald et al., 2019).  Finally, we found strong support for our last two sub-hypotheses. As suggested by prior studies (Ahlers et al., 2015;Lukkarinen et al., 2016;Vismara, 2016Vismara, , 2018Vulkan et al., 2016), campaign characteristics influence ECF performance. Specifically, CCC has a positive impact on ECF performance, while COC has a negative impact.

Managerial and practical implications
The findings of this research, highlighting some suggestions to improve EDM processes in ECF, may also provide useful implications for entrepreneurs and policymakers. This research warns entrepreneurs to be aware of their decision-making processes and to evaluate and monitor them over time. When entrepreneurs improve their understanding of EDMa complex process in which their characteristics, motivations and behaviours are linked to each otherthey will make better decisions, avoiding biases and increasing their effectiveness.
Our results might help them to more clearly define their choices in ECF campaigns and to align their campaign characteristics with the desired results they need to grow their business (Wald et al., 2019). Entrepreneurs, in fact, need to not only manage crowd inputs to get additional resources but also create added value for their companies. The results might help entrepreneurs to enhance their choices and actions to support their business development.
Entrepreneurs adopting ECF should, above all, consider the kind of backers they want to attract. When they want to attract experienced investors, they should set their campaign offer at an adequate level to attract external stakeholders and relevant industry players (Di Pietro et al., 2018). This is an intriguing insight since in the current scenario it is relevant to understanding how entrepreneurs deal with other stakeholders (Caputo et al., 2018). Moreover, our results highlight that entrepreneurs may adopt ECF as an OI platform to get meaningful feedback and as a tool to promote their business; these two motivations are significantly related to ECF performance, showing that these two objectives may reinforce each other.
However, entrepreneurs should pay attention to their self-efficacy and to the idea that ECF may provide them only with financial resources as, according to our data, this is a selfdefeating approach; it increases the focus on campaign characteristics linked to the offer, while reducing those linked to communication, effectively reducing the possibility to reach success. Also, investors, platform managers and authorities would find it useful to understand the EDM process in ECF as presented in this research. This research highlights that policymakers and platform managers should try to engage more sophisticated investors since entrepreneurs still do not consider ECF platforms as a viable source of valuable relationships (Di Pietro et al., 2018). Adding these new investors may prove useful in making ECF more attractive to entrepreneurs without having to lose the knowledge value of the crowd. This is particularly relevant as even if we had found a positive and statistically significant effect of CCC on ECF performance (factor loading 5 0.48), the NET and PR effects on CCC are opposite, potentially negating each other's beneficial effects.
Our data highlight that both the motivation to get feedback from the crowd and the one to leverage the crowd as ambassadors for the company increase ECF performance (Mollick, 2014;Vismara, 2016;Block et al., 2018b;Polzin et al., 2018); accordingly, platform managers should try to help entrepreneurs use them to more actively engage the crowd. One option could be to suggest entrepreneurs provide a minimum number of updates for their campaigns or to provide them with a forum to interact proactively with their potential backers.
EDM may be considered from the perspective of national ECF programs. Often, such programs tend to emphasise the intuitive side of decision-making, while the focus on entrepreneurs remains underestimated. For these reasons, it might be useful for academics and institutions to start looking at entrepreneurs as strategic decision-makers who need to develop and be more aware of their EDM process as well as the role of their characteristics, motivation and behaviours. Hence, authorities and policymakers should offer specific programs in which both EDM and strategic thinking are promoted for entrepreneurial success in new contexts like ECF. Authorities (e.g. Consob in Italy) are interested in the entrepreneurial dynamics of ECF and trying to encourage entrepreneurs to not only use ECF to get financial resources.

Conclusions, limitations and avenues for future research
This study is among the first focused on EDM in the ECF context. We have explored some entrepreneurial decisions that precede the launch of a campaign, and we have found that at the base of EDM there are important entrepreneurial characteristics influencing their motivations and behaviour.
Entrepreneurs that decide to use ECF present different levels of EA and ESE, leading them to search for different opportunities deriving from crowd exploitation (PMK, NET and PR). These motivations in turn influence entrepreneurial behaviours (CCC and COC) and, as a consequence, ECF performance.
This study leveraged well-known constructs in the literature and developed a theoretical framework based on EDM theory to understand decisions by entrepreneurs in a new context.
The added value of this research lies in the analysis of the entrepreneurial aspects underlying entrepreneurial decisions to use ECF. Our results highlight that these aspects are central in decision-making and play a key role in the ECF context. The study highlights that the choice of campaign characteristics is only the last step of a much broader process in which both the characteristics and motivations of the entrepreneur have significant importance. Moreover, our research points to the need to study the pre-campaign phases as a way to further contribute to the EDM literature.
Our research is explorative and limited in several ways. Given the novelty of both the context and the topic, there is a scarcity of other contributions based on entrepreneurial perspectives, and, in particular, there is a lack of studies which specifically explore EDM in ECF. This is the first, and most important, limitation of our research since we cannot make Exploring entrepreneurial characteristics in ECF comparisons with previous studies in this field to find support for our results or not. We acknowledge that this was an explorative study, so we have investigated EDM in the ECF context using well-known constructsi.e. EA and ESEwithout considering other entrepreneurial motivations, traits and characteristics that could be explored to improve our understanding of EDM in ECF. The second limitation is that we focused on three types of entrepreneurial aspects, but other parameters could be examined for the same purpose. This offers several possible avenues of future research. In this vein, other intriguing aspects of EDM to investigate within the ECF context could be related to emotions, cognitive parameters, entrepreneurial capabilities and how the relationship with the surrounding environment affects entrepreneurs' judgments (Sadler-Smith, 2004;Cardon et al., 2012;Foss et al., 2019). Another limitation of the research lies in the sample size, 26.36% of the actual Italian ECF population, but as ECF is growing in numbers and rapidly evolving, thanks to new technologies (e.g. ICOs, STOs and IEOs) (Politecnico di Milano, 2019), our data will probably need to be updated soon even if the number of entrepreneurs fully interviewed represents today a satisfactory sample, given the novelty and the dimension of ECF (different from other models, such as reward crowdfunding). In the future it will be possible to enlarge the survey sample by replicating the interviews through different sources (e.g. phone-based interviews) and enrich the data, thanks to new campaigns that have been launched since this study's inception by other entrepreneurs.
Moreover, as this is a novel topic, scholars still have to develop a full set of validated scales for these constructs that could increase these studies' predictive power; new research attempts should use more engaging methodologies in order to capture more personal characteristics, such as emotions, cognitive parameters and environments. I see links between seemingly unrelated pieces of information AC2 I am good at "connecting dots" AC3 I often see connections between previously unconnected domains of information EJ1 I have a gut feeling for potential opportunities EJ2 I can distinguish between profitable opportunities and not-so-profitable opportunities EJ3 I have a knack for telling high-value opportunities apart from low-value opportunities EJ4 When facing multiple opportunities, I am able to select the good ones ESE1 I'm confident I'm successful at identifying new business opportunities ESE2 I'm confident I'm successful at creating new products ESE3 I'm confident I'm successful at thinking creatively ESE4 I'm confident I'm successful at commercialising an idea or new development PMK1 In ECF I look for learning new product/service features to consider PMK2 In ECF I look for feedback on the early-version of the product/service PMK3 In ECF I look for getting needed foreign country information PMK4 In ECF I look for getting market trends insights PMK5 In ECF I look for information about potential competitors and partners NET1 In ECF I look for connections with key industry players NET2 In ECF I look for contacts with investors to obtain additional financing NET3 In ECF I look for distribution contacts NET4 In ECF I look for connections to favour recruitment of key staff PR1 In ECF I look for leveraging crowd's contacts to help promote my business PR2 In ECF I look for backers acting as marketing channels PR3 In ECF I look for backers helping me in spreading the word through social media MD 59,5