COVID-19 crisis impact on the stability between parties in crowdfunding and crowdsourcing.

This research reviews challenges in building sustainable relationships between the parties involved in the crowdfunding and crowdsourcing projects, which are running in extreme situations, such as the COVID-19 pandemic. This study aims to solve problems that generate the crowdsourcing concerns and to find better alternatives to increase trust for crowdfunding among donors, as this impacts their strategic sustainability in the conditions of turbulence and COVID-induced financial crisis. It was found that factors influence donor decisions in different ways, yet the common tendency for donor activity is non-monotonicity. Future development in the field of sustainable relationships should focus on creating a donor classification system.

Introduction

Crowdfunding become a popular early-stage funding tool in various campaigns, although it has not yet withdrawn from the category of charity tools completely. Many charity campaigns continue to obtain funding through crowdfunding [1]. To date, the majority crowdfunding studies have focused mainly on those factors contributing to a successful campaign and thoroughly examined the experience of the corresponding actors [2]. The results of discussions in various fields suggest the potential for crowdfunding platforms to develop efficiency [3], for capital from the crowd can increase the funds of both business companies and social movements. The recent COVID-19 outbreak showed a high demand for innovations that can be used in the fight against health hazards, specifically the pandemics. The crowdfunding platforms can provide these projects with financial support they may need within a fairly short timeframe. In crowdfunding, donors motivated by getting something in return jointly finance startup projects and make contributions to achieve a funding goal. However, when there is an emergency, such as COVID-19, donors are interested not in financial gain, but in finding quick and reliable solutions.

Practice shows that failure of many crowdfunding campaigns is due to setting of inadequate tasks or a complex return mechanism [4]. Recent studies have shown that crowd’s motives for participating in various forms of crowdsourcing initiatives may vary [5]. Studies on popular European crowdfunding platforms show that crowdfunding not only feeds technology startups but also serves a new source of finance for the existing companies and firms [1]. Furthermore, crowdfunding platforms are of interest to scientists and researchers who seek to implement their research projects focusing on COVID-19-related topics. These projects attract donors with different kinds of motivation and decision criteria [6]. Thus, investigating motivation itself and motivation for donation is crucial for the promotion and development of crowdfunding platforms.

Studies concerning the impact of investing options on funding decisions during the pandemic are nowhere to be found. The options in point may relate to bonus/reward criteria established by project initiators, and donors can choose any of those options themselves. For example, bigger commitments bring more profit or assets [7]. Therefore, the number of options may be a significant factor influencing donor decision-making. Studying the impact of various motivations on fundraising provides an opportunity to improve project design and thus increase the likelihood of reaching success in the campaign. At this point, motivation research is a promising and relevant direction in the field of crowdfunding technologies.

This study aimed to look at the matter from another angle, a little more on donors and a bit less on project initiators, specifically on their relationship with the acceptors of their donations. The study reviewed sources exploring the interaction between the crowdfunding project initiators and donors, with the aim to find a pattern in donor behavior for relationship building and to understand the flow of donors in donor acquisition and retention. This interaction is the key contribution to a successful campaign [8]. The study attempted to identify factors that would make the donor more attached to the project and continually make donations while retaining interest in the project even after the pandemic has come to an end so that others could handle the consequences of the pandemic and situations similar to it. The purpose of this work was to establish factors influencing donors’ intention to invest in a crowdfunding project during the COVID-19 pandemic. For this, quantitative and qualitative reviews of already existing crowdfunding projects were carried out. The quantitative review is performed through meta-analysis, which focused on the overall effectiveness of the project and its ability to meets donor expectations. A systematic literature review was conducted to identify drives of success and failure for crowdfunding and crowdsourcing projects. It is hypothesized that likely differences in views regarding the role of communication, commitment, and trust in initiator-donor relationship can affect the quality and sustainability of this relationship and the campaign success rates.

Literature Review

Globalization and informatization processes that take place while social groups blend together and acquire resistance to conventional incentives bring into the world a need for tracking social trends and identifying patterns on which new social relations are built [9]. This is especially the case when there is a financial crisis, such as one caused by the COVID-19 outbreak, and many projects have lost their financial support and were suspended or closed. The threat of a new outbreak and the following financial collapse reqires project initiators to create medium- and long-term plans of socio-economic development to restore after the crisis [10].

The crowdsensing technology allows individuals having sensor-equipped mobile devices (e.g., smartphones, tablets, smart watches, etc.) to collectively share data, with the purpose of measuring, mapping, analyzing, estimating or inferring (predicting) any processes of common interest [11]. Such systems gather relative information via direct reports and by passive methods, which refer to the automatic sending of information about the needs of people on the Internet. For example, marketers and advertisers can easily know your current interests from your search engine queries and then inconspicuously offer the product or service of interest through advertising. This approach cam significantly increases the effectiveness of advertising but sometimes it may have a negative effect on the consumer, as in the case of aggressive/intrusive scenario, which results in distrust and avoidance by the target audience.

Problems with crowdsourcing and crowdfunding occur due to privacy threats from personal information or tasks. Risks associated with data processing refer to privacy of sensed data (PSD), privacy of computing inputs (PCI), and privacy of computing results (PCR). The sensed information may include confidential or sensitive data such as geolocation details obtained from GPS devices. Even specific environmental data may reveal your position. This information, if disclosed, may cause breakdown in trust and therefore, must be treated as confidential. Privacy threats from tasks embrace task privacy of end users (TPEU) and the task privacy of participants (TPP). Participants in crowdsourcing are condemned to disclose certain information to reach the set goal. Thus, a collision of information processing technologies and social demands takes place, posing a threat to sustainability of projects with share financing.

Yang et al. [12] explored a number of critical security and privacy issues that impede the application of the mobile crowdsourcing network (MCN). When participating in crowdsourcing, mobile users consume their own resources (e.g., battery, mobile data, and memory) and may suffer potential security and privacy threats, which, in turn, affect the quality of relationship between actors. The output results of crowdsourced computing tasks may be sensitive or private. End users do not want the service provider to know the content of the results or obtain some sensitive information from it [13, 14].

It becomes difficult to protect personal information when a group of people involved in the project becomes dynamic because when it happens, the initiator is no longer capable of determining the exact size of that group [15]. People in such an ecosystem respond to crowdsourcing tasks depending on their personal interests, location, or Internet accessibility. Due to the high mobility of the people, the user network topology is in constant change, making it difficult to solve security and privacy issues.

In their work, Yang et al. [12] also examined the service provider. In their definition, the service provider is a “crowdsourcing platform that provides crowdsourcing services to both end users and public crowds” and end users in turn are “customers who purchase or rent crowdsourcing services at certain costs.” Hence, they send service requests to the service provider and receive results from it.

Another problem with the protection of personal information arises with the Internet of Things and intelligent technologies for crowdsourcing [16], such as in smart city projects [17]. Developers need to make sure that project initiators can collect and process big data for the purposes of data analytics and forecasting not only fast but also safely [18]. Yet, organizations may still face data privacy challenges and lose the people’s trust in cases of information misuse [16]. It often happens with crowdsensing and crowdsourcing platforms [19]. This, in turn, is one of the many obstacles to the early integration of smart city services [20, 21]. In such cases, city dwellers are inactive information providers and show no interest in crowdsourcing campaigns [22]. To alleviate this problem, it was proposed to give residents certain tools to track and control how their person information is used. However, it is not easy to enable this function and most crowdsourcing platforms do not provide such an opportunity. The emerging mismatch between social demand and supply leads to destabilization of the strategic sustainability of such projects. Therefore, there must be information security controls to ensure privacy, integrity, and limited access to the information that is submitted to the crowdsourcing system [23].

Zhu and Zhou [24] examined the problems of crowdfunding, the blockchain technology in particular, through the example of China. The major directions that were tackled include the benefits of the blockchain solution; use cases (e.g., registration of stocks and shares of a firm financed by crowdfunding; transaction and transfer of crowdfunding equities; peer to peer transactions between investors and entrepreneurs); and potential of the technology (i.e. facilitates the circulation of crowdfunding equities; enables crowdfunders to be involved in corporate governance through a specially developed voting system).

Among the major challenges associated with crowdsourcing are, among other things, shareholder registration and capital management. The lack of information security is associated with certain risks that discourage investors them from investing in a particular project [25]. In crowdfunding, investors are the only source of money, which may be at risk of misuse due to the lack of legal regulation and the number of investors involved [26]. In addition to that, investors may not have the opportunity to influence corporate governance, which leads to conflicts of interest [27]. There are also regulation issues related to investment quotas, illegal fundraising, detection of fraudulent fund raising, and anti-money laundering [24].

Mobile app crowdsourcing (MCS) is also fraught with security, privacy and trust problems [28]. The main problem with mobile crowdsourcing is the openness and mobility of data, allowing various attackers easy access to user information (such as GPS data or cellular network identifiers) [29]. The openness of MCS enables all mobile users to participate in MCS activities, but the information they offer is not necessarily accurate, a problem that requires the introduction of criteria to differentiate reliable and unreliable information [30].

The major problems associated with crowdsourcing, crowdsensing and crowdfunding platforms revolve around the protection of personal information. Insecure data transmission and ineffective protection of data privacy lead to the loss of investor confidence and, consequently, to a decrease in fundraising (Fig. 1).

Fig. 1

Graph showing cost comparison of products with and without advanced protection

Due to the COVID-19 pandemic, most countries imposed restrictions on movement and working hours, causing businesses to lose money. In these conditions, industries that provide information products and services have gained popularity [31], resulting in the rise of online projects aimed at resolving the consequences of the pandemic crisis. The most acute issues in this context are the lack of jobs and social distancing. This is why moving to crowdfunding platforms seems to be the most appropriate solution. Therefore, this study aims to investigate the influence of economic, social and environmental factors on the success of crowdfunding projects during the COVID-19 pandemic that hit the whole world over the past year. The study focuses on interrelated donor attractions and trust issues to estimate the impact of each factor on one’s decision to participate in the crowdfunding project.

Research Methodology

This paper is a systematic review that uses meta-analysis of empirical data from multiple studies to assess the effect size of incentives in crowdfunding. Because there is very little literature on the problem of crowdfunding during COVID-19, and most sources focus on finding medicines and morbidity statistics, this study relied on materials from DonorsChoose.org, a crowdfunding platform for education projects [32].

Algorithm of Selection of Individual Studies

Because meta-analysis bases on statistical processing of results from single studies, this study converts statistics to a general metric form with Pearson correlation. Here, the magnitude of the impact is estimated for each work examined. Studies in meta-analysis were selected under the following criteria:

Direct or indirect empirical data;

Details on the sample size are present;

Coefficients of correlation between the amount of donation and the donor count are computed and can be converted to other correlation coefficients;

Studies and data sampling are dated no earlier than 2012;

In order to avoid cross-references, only original sources (peer-reviewed journals) are used.

In order to maintain data independence in meta-analysis, one finding per sample was taken. A total of 50 empirical studies were found, of which only 10 met the selection criteria.

A number of scientists focused on two types of behavioral events, donation recurrence (whether one donor will make donations at some time intervals in the future) and donor retention (whether a donor will remain on the crowdfunding platform until a future time) [33]. These terms describe the major aspects of relationship building among donors [34]. Studies normally use large-scale behavioral data collected from crowdfunding. This information is important because the problem of donor attrition (many donors donate only once or very few times within a rather short lifecycle and then leave) is a crucial issue for sustainability of crowdfunding platforms and traditional charitable organizations. From this perspective, analyzing factors and predicting donor behaviors is an urgent task.

A mixed method study conducted in 2017 examines factors that influence backers’ trust in crowdfunding when re-wards are delayed [35]. Based on in-depth interviews with crowdfunding participants, it was found that a rich set of factors influenced the trust of backers, including backers’ role identity and domain knowledge, backers’ research on a creator’s background, communication between participants during delays, and duration of delays. The study is rich in terms of evidences, reports on a regression analysis of 4089 delayed projects, and analyses such indicators as the funding goal, number of backers, percentage of funds raised, number of reward levels, and previous crowdfunding experience.

In the study of Vismara [36] one may find a sample of 271 projects listed on the UK platforms, Crowdcube and Seedrs, between 2011 and 2014. Another paper explores donation in the original sense of the term, e.g., blood donation [37]. We decided to include this work into our research for understanding the basics of donor’s emotional response to donation.

User interactions were studied in Gerber et al. [38] through the example of more than 50 crowdfunding projects. In their next work [3], authors focused on human–computer interaction issues and the scale of study was as large as in 2012. Authors Giudici et al. [39] provide rich statistics using a dataset of 618 proponents that launched 457 crowdfunding projects on 13 Italian reward-based platforms, analyze rewards and valuable empirical experience.

Path Diagram of Direct and Indirect Estimates

Methodology in this study allowed synthesizing direct and indirect estimates of the relative factor effects for the same case under study. The two zero-approximation theories were compared in a direct research. An indirect estimate was obtained from research comparing these two theories by the general comparative method. To investigate indirect effects, a path analysis approach was used in which the extent to which the casual variable changes was measured. Change of the casual variable affects the dependent variable both directly and indirectly through the correlation coefficient (r). This analysis is presented in the form of a path diagram (Fig. 2).

Fig. 2

A Path Diagram: factors 1, 2, and 3 directly impact the effect and interact with each other to indirectly impact the effect

The path diagram illustrates the connection (arrows) of path coefficients (P) and correlation coefficients (r) between various factors. Here, P represents a partial regression coefficient that measures the magnitude of the direct effect of one variable or another, holding other variables constant. The diagram displays the links between three factors, which can have effects that are same but of different magnitude. Factors that match in terms of effect constitute the major factor. The path coefficients P between one factor and the major factor as well as correlation coefficients r between factor pairs may be expressed for the correlation coefficients (r) as follows:

Path coefficients in Eq. (1) can be calculated using simultaneous equations. Path and correlation coefficients can be used to calculate direct and indirect effects (Table 1).

Table 1

Equations for direct, indirect and total effects

	Direct effect	Indirect effect	Total effect
Factor 1	P_1P_4	r_1,2P_2P_4+r_1,3P_3P_4	r_1
Factor 2	P_2P_4	r_1,2P_1P_4+r_2,3P_3P_4	r_2
Factor 3	P_3P_4	r_2,3P_2P_4+r_1,3P_1P_4	r_3

A single factor influencing donor retention can have varying effects investigated by different randomized studies, i.e., an evidence network. Direct and indirect estimates from many different comparators can help compare the hypotheses about the influence of one or another factor. Using this meta-analytic methodology, the study synthesized different pieces of information and evaluated a number of internally consistent hypotheses about the relative influence of donor retention factors.

Results

Based on the empirical data examined, we can argue that donor trust is a key factor to loyalty. The organization’s ability to reach the set goals on time and a transparent public report are the major inducements to attract and maintain a wide audience of crowdfunding platform users and a foundation of sustainable relationships between parties involved in a crowdfunding project. To complete the picture of a good service provider, the impediments involve an incorrect fundraising goal, failure to meet deadlines, and report delays. Donor’s motivation is greatly undermined by the failure to meet the crowdfunding goal as well as by sudden changes of thereof. Therefore, it is extremely important not to allow destabilization between the processes within the project and at the level of obligation fulfillment, as this will result in an information asymmetry. Donors who donated a project, which reached successes afterwards, are more likely to return. There are several reasons for such a behavior: donors trust the platform more, as their donations contribute to the cause, i.e. the public offering makes sense. These phenomena represent a response to the ongoing process of global social transformation and a social request for participation in share-value projects [40]. Aside from that, donors feel themselves involved and receive greater recognition if donations are made to successful projects. It is the sense of impact that comes to the forefront in further cases and serves as emotional filler within a stable relationship between the parties. Results of the empirical data analysis distinguish three factors that impact both the participation in crowdfunding and the amount of donor contribution: economic, social or ecological interests. Table 2 presents a summary of estimates of the direct and indirect effects inflicted by these factors (all data relate to a single work).

Table 2

Estimates of direct and indirect effects of different factors

Factor	Direct effect	Indirect effect	Total effect
Social	0.15	0.01	0.16
Economical	0.10	− 0.2	− 0.1
Ecological	0.08	− 0.12	− 0.04

Data in the Table above show a positive correlation of the social factor. Comparing different works (Fig. 3), however, an unambiguous conclusion was drawn. In various empirical studies, the social factor exhibits a different impact on decision-making. Besides, the histogram below illustrates that indirect effects add more to the total effect than the direct effects. Results concerning other factors are similar.

Fig. 3

Histogram of direct, indirect and total effects of the social factor: a summary of results from multiple works

Despite the advantages of meta-analysis, the researchers split over the validity of indirect comparisons. One of the main points of criticism is the nature of the meta-analytic evidence. Despite the fact that donors were randomized within the frame of each specific project scenario, they were not randomized to all studies included in the review. Moreover, the crowdfunding campaigns lacked balance between the number of donors, project goals and project duration. Therefore, indirect comparisons are non-randomized comparisons and provide observational, rather than randomized, evidence. For this reason, indirect comparisons of donor attraction/retention factors may be more susceptible to bias due to the mixing of sources and the presence of options to choose from (e.g., the choice of a comparator depends on the relative effect of a hypothetical factor).

Various statistics show an inverse correlation (after a certain threshold is reached) between the sense of involvement and the crowdfunding goal; the larger the sum of money needed to make the project possible, the lower the number of donations. Figure 4a illustrates this trend with a graph created on the back of data that were collected during many crowdfunding campaigns. As we can see, the function is not smooth and broad peak; perhaps this means the presence of multiple peaks. This result can be easily explained by heterogeneity of input statistics, which come from different countries, and apply to different cultural and economic contexts of crowdfunding. Figure 4b shows trends given in random papers in order to demonstrate heterogeneity and ambiguity of data under study. The presence of several peaks indicates the connection between the factor that is involved in decision-making and the project cost. Anyway, a common pattern aligning different countries is that after reaching a certain fundraising maximum, the activity of donors monotonically decreases, with the next threshold becoming higher compared to the first goal. This confirms that there is a need to use tools that are most adaptive, correspond to social demands, and provide the most vivid and stable feedback while managing crowdfunding projects.

Fig. 4

a The number of donations in relation to the crowdfunding goal, adapted from papers under consideration; b Comparative trends for three single studies

Donors make donations basing on a comfortable amount, which remains almost unchanged [32]. Thus, the need arises to retain the donor not by the incentive to higher donations, but by maintaining the amount on a steady level. In light of current global trends, donor retention is becoming a fundamental priority.

In 2019, the average amount of donations amounted to $ 303.70, which is 26.5% more than in 2018 ($ 223.08) [32]. About 0.55% of all donors donate over $ 5000 (68.42%). Each of them on average donates $ 75.606 during the whole life. Such donors also have the highest retention rates of 51.69%. At the same time, 84.8% of all donors donate less than $ 100 (8.83% of all donations) with an average donation of $ 42.31. The retention rate of such donors corresponds to 39.54%.

In 2018, the retention rate of novice donors (donated for the first time in 2018) equaled 19.53%; about 29.32% for renewed donors (did not donate in 2018 but donated earlier); and 48.10% for repeat donors (those who donated in 2018 and earlier). This confirms that in 2019 the overall trust for crowdfunding among donors decreased.

Taking into consideration the information above, trust is believed to be a determinant factor for donations. First of all, because the donation is exclusively voluntary and bases on the donor’s desire to contribute to the project implementation through the introduction of financial resources.

This desire can be legally influenced through the use of the following tools:

Creation of a typical donor portrait to prevent difficulties due to cultural differences. For example, donors of Western culture prefer when something is presented with “cool” and “wow!” effect. However, the emphasis on the same emotional aspect for representatives of the East can lead to undesirable consequences [41];

Strategies for building trust. The donor must be sure that after receiving the money, the project will be managed effectively. It is necessary to demonstrate other achievements and social accounts of the project founders, providing information about the team background;

Future donors rewarding. All the donors should receive a reward. Donors or company names can be indicated in a special place of the released product. For example, in a computer game or on the pages of comics, one of the characters may bear the name or even the appearance of the donor. Such a bonus often plays a decisive role in choosing a project;

Development of a funding threshold and a way to receive payments above this threshold. Often after reaching 100% of the amount declared by the project, the fees cannot stop for a long time. Therefore, it is necessary to give a clear signal that you also have ideas for using these additional funds. Otherwise, this will negatively affect the reputation of the team. However, the disbursement of funds should be implemented for purposes close to the original business idea;

Permanent communication during the campaign and the disbursement of funds. It is better to inform donors about the progress of funds expenditures as well as any arising delay and discrepancy. Often, many former donors donate money to projects if they require more than planned through personal communication channels [42]. However, in this case, information on the progress of the implementation of a business idea should be presented to donors on a stable basis, but not only in the moments of inconsistency. It also should be noted that donors are mostly using mobile channels to receive information and make donations [43]. Therefore, it is advisable to include additional features for accessing project information via mobile phones;

Reports on donations receipt effectively influence donor trust. Even if the use of donations is not controlled, the information should not stop flowing into the network.

Thus, relations with donors remain a full-fledged element of crowdfunding activity and require efficient management to maintain a high level of donor trust in the project.

Similar results were obtained in other studies. A study on the knowledge-based and stakeholder management perspectives focused on mathematical modeling of development [44]. It was found that the opportunity to reach intensive development of energy-efficient and renewable energy sources encourages firms to participate in the crowdfunding campaign and support innovative strategies for the more rapid development of new technologies. These findings were obtained through a partial least squares statistical approach and relate to Spanish and Portuguese firms from the renewable energy sector.

Discussion

A meta-analysis of research projects that compare crowdfunding with traditional funding mechanisms [45] show that attention of the crowd is given to projects that convey new simple concepts, have smaller budgets and are likely to be implemented within a shorter timeframe, rather than previous merits and experience of the project creator. According to statistical analysis, students, junior researchers, and female scientists are more likely to succeed than senior scientists. Furthermore, donation-based crowdfunding, along with the operator of the Internet platform, involves people who raise funds on a gratuitous basis [46]. The study showed that the crowd may apply decision criteria than are distinct from those applied in traditional funding agencies. The direct and indirect factors affecting decision-making, however, were not investigated. Therefore, the present approach can help many scientists and researchers find the right path towards success in promoting their project.

Projects that have been already launched attract more donors and funding, and many companies should consider this before starting the fundraising campaign [47]. Hence, the pre-financing planning can significantly increase the chance of raising money. A number of works have emphasized the role of social networks in a successful crowdfunding campaign [48, 49], which help introduce the product to potential donors of different age, gender, and with different interests.

Other important aspects of the crowdfunding platform are the period of funds withdrawal and financing methods [50]. There are fixed and flexible methods of financing. With a fixed method, the financing initiator receives the money if the project has met the goal set. Within the flexible method, various options are possible: the platform may set contributions if the project does not meet the campaign goal; or additional financing of the project can be performed. Consequently, the costs of servicing money raised through crowdfunding, depend on fees charged by the platform, and the degree to which the fundraising goal has been achieved [51]. When successfully implemented crowdfunding transactions positively affect the image of their initiator, the failed ones can damage the reputation of the project team [52]. Besides, there are risks of losing the authorship of the idea embedded in the project. If a business idea that is not protected by copyright, other entrepreneurs can take it with full openness and make it widely available on the Internet with all the ensuing consequences.

Conclusion

Based on the available information, we can make a summary of basic intuitive and rational methods to improve sustainability of relationships between those who are involved in a crowdfunding project. A crowdfunding platform needs to be honest with its users, as trust between parties is an extremely valuable resource. Moreover, it must pay attention to its goal-setting policy and publish reports in time. As is was found, donors donate more often if the project is close to completion or is launched by people, who live or plan to implement their project idea relatively close to them (e.g., in the same country). The influence of donor’s personal data disclosure is also significant. We assume that by disclosing one’s own personal information, donators fulfill their own crave for recognition.

Standardization of reporting and transparency ratings of organizations can increase public trust in crowdfunding processes. Its absence remains the main obstacle to a successful dialogue between donors and donation recipients. As long as there are no unified standards that can improve cooperation within a crowdfunding platform, distrust of the crowdfunding will persist, even when there is a COVID danger or similar threats. However, the initiative to develop convenient and transparent tools and procedures for reporting and communication lies with market participants since the difference in national regulation will not allow creating a universal legal mechanism. Further developments in the field of sustainable relations between crowdfunding project members after the pandemic has ended should focus on the development of a system for clustering donors. The classification of donors will help not only evaluate their reliability and stability but also provide guarantees to future investors. For this, a variety of approaches will be used such as the analysis of social network accounts and other social institutions.