A holistic lens on entrepreneurial learning from failure: continuing the legacy of Jason Cope

Abstract

Purpose

The paper follows Jason Cope's (2011) vision of a holistic perspective on the failure-based learning process. By analyzing the research since Cope's first attempt, which is often fragmentary in nature, and providing novel empirical insights, the paper aims to draw a new comprehensive picture of all five phases of entrepreneurial learning and their interplay.

Design/methodology/approach

The study features an interpretative phenomenological analysis of in-depth interviews with 18 failed entrepreneurs. Findings are presented and discussed in line with experiential learning theory and Cope's conceptual framework of five interrelated learning timeframes spanning from the descent into failure until re-emergence.

Findings

The study reveals different patterns of how entrepreneurs experience failure, ranging from abrupt to gradual descent paths, different management and coping behaviors, and varying learning effects depending on the new professional setting (entrepreneurial vs non-entrepreneurial). Analyzing the entrepreneurs' experiences throughout the process shows different paths and connections between individual phases. Findings indicate that the learning timeframes may overlap, appear in different orders, loop, or (partly) stay absent, indicating that the individual learning process is even more dynamic and heterogeneous than hitherto known.

Originality/value

The paper contributes to the field of entrepreneurial learning from failure, advancing Cope's seminal work on the learning process and -contents by providing novel empirical insights and discussing them in the light of recent scientific findings. Since entrepreneurial learning from failure is a complex and dynamic process, using a holistic lens in the analysis contributes to a better understanding of this phenomenon as an integrated whole.

Citation

Lattacher, W., Wdowiak, M.A., Schwarz, E.J. and Audretsch, D.B. (2024), "A holistic lens on entrepreneurial learning from failure: continuing the legacy of Jason Cope", International Journal of Entrepreneurial Behavior & Research, Vol. 30 No. 11, pp. 205-235. https://doi.org/10.1108/IJEBR-06-2023-0635

The real voyage of discovery consists not in seeking new landscapes, but in having new eyes. (Marcel Proust)

1. Introduction

A good decade ago, one of the most influential studies on entrepreneurial learning was published: Cope's (2011) “Entrepreneurial learning from failure: An interpretative phenomenological analysis”. The paper provided pioneering insights into the failure-induced learning process and -contents (Mueller and Shepherd, 2016; Pittaway and Thorpe, 2012) and set the stage for a holistic perspective on entrepreneurial learning from failure. Employing experiential learning theory, Cope contextualized the learning process into a five-stage framework (five “learning timeframes”), ranging from the descent into failure to the application of learning contents in the subsequent activity. The study's empirical findings covered the effects of failure, emotional recovery, reflection on the failure, and the resulting learning contents. While the findings were pioneering, representing the most comprehensive empirical representation of learning from failure, they were exploratory and did not cover the entire five-stage process. Inconsistent with the framework and experiential learning theory, the empirical work omitted the early stages of the learning process (i.e. the descent into and the experience of managing failure) and did not fully cover the last stage, i.e. re-emergence, as it did not analyze learning application.

Subsequent research made essential steps in detailing some of Cope's insights and exploring learning application upon entrepreneurial re-engagement. Most of this work takes a reductionist approach, narrowing the phenomenon under study. By breaking the phenomenon down to a narrower focus, researchers can examine one component in much greater depth (Dafermos, 2014). In particular, subsequent research focused on how learning is shaped by emotion (Byrne and Shepherd, 2015), failure attribution (Riar et al., 2021; Walsh and Cunningham, 2017; Yamakawa and Cardon, 2015; Yamakawa et al., 2015), narcissism (Liu et al., 2019, 2021; Wang et al., 2022), stigmatization (Singh et al., 2015), social relationships (Heinze, 2013), and serial entrepreneurship (Nahata, 2019). Recently, research seems to be increasingly engaging in endeavors with “broader” foci, collecting influence factors of learning from failure (Wei et al., 2019), re-conceptualizing the process from failure to re-emergence with learning as one area of interest (Costa et al., 2023; Ucbasaran et al., 2013), and extending the process to organizational imprinting of learnings in subsequent ventures (Amankwah-Amoah et al., 2018).

Despite the merits of these works, the earlier stages of the learning process (descent into and management experience of failure), which yield pivotal implications for the later reflection (Shepherd et al., 2009), are still largely unexplored. Moreover, regarding learning application, research made progress merely in the case of entrepreneurial re-engagement after failure. Other possible application contexts – especially the most common case of a change into an employed work setting (Burton et al., 2016; Omorede, 2021) – have not been adequately addressed. Consequently, existing knowledge about the learning process still yields significant gaps. These gaps limit understanding how the learning stages (“timeframes”) relate to each other and form the learning process. Cope himself recognized “the complex, interdependent relationships between different timeframes” as “key areas for research” (p. 620). His vision of a holistic understanding of how and what entrepreneurs learn is also today unfulfilled. A recent review by Costa et al. (2023) also emphasizes that “what, why, and how do entrepreneurs learn” are still essential avenues of future exploration.

The present study follows this vision and uses a holistic lens to analyze the process of entrepreneurial learning from failure. A holistic approach has ancient roots and can be traced back to Aristotle's formula: “The whole is more than the sum of the parts”. This approach assumes that the individual parts can only be understood if they are considered in the context of the whole, i.e. how these parts interact and influence each other (Ralston, 2014). For complex phenomena like human behavior, a holistic lens has the advantage of assessing multiple factors that might contribute to explaining it. In addition, looking at factors as a whole makes it easier to keep in view the intrinsic purpose and function of behavior. The following study aims to provide insights into all five learning timeframes from failure and how they relate to each other. Specifically, the study answers the following two research questions:

RQ1.

How do entrepreneurs descend into, manage, perceive, recover, and re-emerge from failure and thereby shape their learning process?

RQ2.

How do the learning timeframes from failure relate to each other?

Like Cope's seminal work, this study adopts an experiential learning theory perspective, conceptualizing learning as a process of experiencing, reflecting, and acting. The perspective is analyzed in the light of contemporary literature, which represents a considerably advanced state of knowledge. The empirical data stems from in-depth interviews with 18 failed entrepreneurs in knowledge-intensive industries, which are analyzed using interpretative phenomenological analysis.

The study contributes to research in the following ways. It uses a holistic lens to look at the phenomenon under investigation. Especially when studying complex phenomena such as learning processes, it is helpful to regularly take a step back and look at the whole picture (with “new eyes”). Doing so makes it possible to obtain a new, sharper overall portrait that reveals new important threads, which still need to be illuminated to gain an even clearer insight and understanding of the phenomenon (Alexandrova et al., 2021). Every attempt to look at the picture holistically is the opportunity and the way to recognize the interplay of (new) individual parts. Holism can thus be seen as a kind of gradual phenomenon (Verschuren, 2001). In this study, the picture of entrepreneurial learning from failure becomes sharper by using a holistic lens. First, the insights gained go beyond the work of Cope and later researchers as they cover the whole process - all five timeframes of learning from failure. In particular, the study examines the timeframes that have been under-researched so far (descent into failure, experience of coping with failure, application of learning in a non-entrepreneurial setting). Second, the holistic lens allows us to recognize how the five learning timeframes relate to each other. Specifically, the study illuminates that the timeframes can overlap, occur in a different order, form a loop, or not occur at all. Further, if the learning process is considered as an integral whole, its intrinsic function is kept in mind, i.e. the application of the learning to the new context. The study looks at the re-emergence phase more broadly than previous research and reveals the application of numerous learning outcomes in new professional contexts that go beyond a new entrepreneurial episode. Finally, the study addresses the individual learning processes as a collective experience and shows important learning effects in the social context (e.g. the experience of solidarity and trust), representing an essential social capital for future endeavors. Overall, the analysis helps to obtain a clearer picture of the learning process from failure as a whole and as a dynamic (and social) phenomenon. The paper contributes to a more comprehensive view of entrepreneurial learning from failure and, thus, a better understanding of this phenomenon that has been called for by Cope himself and contemporary research (Costa et al., 2023; Wang and Chugh, 2014).

The article continues by revisiting Cope's work and relevant advancements by later research, followed by presenting the methodology and empirical findings. The findings are subsequently discussed, deriving implications for theory and practice that foster a comprehensive understanding of entrepreneurial learning from failure.

2. Developments in understanding entrepreneurial learning from failure

Entrepreneurial learning from failure emerged as a promising research field, situated at the confluence of two vivid research streams: Entrepreneurial learning, which started to flourish by facilitating the understanding of entrepreneurship as a dynamic process of becoming (Minniti and Bygrave, 2001) and failure, which research uncovered as an inherently entrepreneurial phenomenon and a context for rich learning (Ucbasaran et al., 2013). Recent reviews (Costa et al., 2023; Lattacher and Wdowiak, 2020; Nogueira, 2019) illustrate that especially over the last decade, entrepreneurial learning from failure attracted high interest by researchers, who were able to provide in-depth insights into several influence factors of the learning process (e.g., emotions, attribution, personal characteristics) and resulting learning contents (e.g. relating management, self-reflexivity). These reviews, at the same time, highlight that Jason Cope had influenced the field like few others, introducing a dynamic learning perspective to entrepreneurship research with a particular focus on learning from critical events (Cope, 2003, 2005; Cope and Watts, 2000) and – with his seminal paper (2011) – providing initial empirical insights into how and what entrepreneurs learn.

In this seminal paper (2011), he follows Mcgrath (1999) and Politis and Gabrielsson (2009), conceptualizing failure as the “termination of a business that has fallen short of its goals” (p. 605). This definition covers bankruptcy but also cases where the business fails to satisfy principal shareholder expectations.

Cope builds on experiential learning theory to account for this learning opportunity and applies a learning lens to failure (Dewey, 1933; Kolb, 1984; Lewin, 1951). According to this theory, the entrepreneur updates a subjective stock of knowledge based on experiences. In the case of failure, the experience that triggers and informs the learning is highly critical (Espinoza-Benavides and Díaz, 2019). Drawing on the work of Fiol and Lyles (1985), Cope postulates that such critical incidents can stimulate higher-level learning, which is associated with a challenging, fundamental form of learning deeply affecting existing mental models. He highlights that this change in mental models does not happen immediately but that both – the failure and the resulting learning – are processes rather than isolated events (Cardon and Mcgrath, 1999). Cope defines five learning timeframes of failure (p. 619) that are essential for a holistic understanding of the phenomenon under study: (1) the descent into failure, (2) the experience of managing failure, (3) the aftermath of failure, (4) the recovery from failure, and (5) the re-emergence from failure, including learning contents and their application in future actions. These learning timeframes are interdependent, so covering all of them in one study contributes to a more accurate picture of the learning process.

Below, the current state of research on each of these timeframes is presented. After a brief summary of Cope’s findings, all relevant additional insights from earlier and later studies are discussed.

2.1 The descent into failure

The first timeframe, “descent into failure”, refers to the venture's demise with its decline patterns and underlying causes. While being aware that the descent into failure is an essential part of the overall failure experience, Cope skims over this stage and only provides short summaries of the analyzed entrepreneurs' professional lives and failures.

The literature before and after Cope provides – if one disregards organizational-level studies on causes of business failure (Carter and Auken, 2006; Garcia Martinez et al., 2019; Ulmer and Nielsen, 1947) and on the process of organizational decline (Habersang et al., 2019; Kücher and Feldbauer-Durstmüller, 2019; Moulton et al., 1996) limited insights into this timeframe.

Shepherd et al. (2009) provide ideas on how entrepreneurs take financially costly actions to slow down their venture's demise for their emotional well-being (“anticipatory grieving”). Pretorius (2008) refers to findings from organizational studies (Weitzel et al., 1991) suggesting a five-stage descent process (from blindness for decline to dissolution). Regarding the triggers and causes for the decline, research on individual-level causes of entrepreneurial failure (Romanelli, 1989; Shepherd, 2003; Shepherd et al., 2000) indicates a wide heterogeneity of individual-level failure causes. More recent studies add to the discussion the importance of the (subjective) attribution of failure causes (Munawaroh et al., 2023; Riar et al., 2021; Shore et al., 2023; Walsh and Cunningham, 2017; Yamakawa and Cardon, 2015).

Despite these conceptual and highly specialized contributions, little is known about how entrepreneurs experience the decline of their company and the related causes and triggers. Consequently, this learning timeframe requires exploration – as also recently emphasized by Jenkins (2022).

2.2 The experience of managing failure

The second timeframe, “experience of managing failure”, covers entrepreneurs' actions upon failing. These actions influence the failure process and form central parts of the experience that entrepreneurs may learn from. Like the descent into failure (timeframe 1), Cope has not covered the experience of managing failure empirically.

Similarly, this learning timeframe represents largely “terra incognita” in the remaining entrepreneurship literature.

Some ideas may be found in the work of Stambaugh and Mitchell (2018), who focused on expertise built during the fight against failure. The authors expect entrepreneurs to make considerable efforts to avoid the adverse effects of failure. This argument aligns with findings on the entrepreneur's tendency to procrastinate business failure (Shepherd et al., 2009; Yamakawa and Cardon, 2017). Regarding the intensity of the learning potential, Stambaugh and Mitchell (2018) argue for an inverted U-relationship: while some degree of intensity is necessary for triggering and informing cognitive processes, an overly intense management situation may lead to a cognitive overload and the decision to give up. Another issue that the authors mention is the duration of a failure-management experience, arguing that a too-short period impedes learning possibilities.

While the studies presented above provide some interesting ideas for the timeframe, the individual experiencing and acting of entrepreneurs upon failing is yet to be empirically explored.

2.3 The aftermath of failure

The third timeframe, “aftermath of failure”, targets the effects that the failure exposes on the entrepreneur. Cope starts his empirical study with this stage, providing an extensive overview of the failure's effects on six impact spheres: financial, emotional, physiological, social, professional, and entrepreneurial. While present in all cases, the financial effects strongly depended on the individual wealth situation and the degree of financial investment, ranging from minor setbacks to severe, life-limiting debts. The emotional impact was the most challenging part for Cope's study participants. They reported “shocks to the system”, extreme emotional commitment and stress, or even feelings of depression. In some cases, the adverse emotional effects also exposed physiological costs, like medically required breaks over several months and taking drugs against high blood pressure. Social effects included a diminishing social environment and the – self-imposed – feeling of guilt, shame, and embarrassment in relation to investors. Regarding future career options, failed entrepreneurs reported entrepreneurial costs in terms of diminishing self-efficacy and risk-taking propensity. In contrast, the professional impact of the failure on job opportunities was limited and, in some cases, even positive.

The aftermath of failure received considerable interest from contemporary research, as the conceptual reviews of Ucbasaran et al. (2013) and Klimas et al. (2021) demonstrate. The latter characterizes failure as a multilevel process inducing different types of effects staggered over time.

Most attention from subsequent research received emotional and social effects. With regard to the former, later studies partially relativize Cope's very negative characterization of the failure's impact on the emotional level. Heinze (2013) fundamentally supports Cope's findings of high emotional costs (mainly related to grief) but also recognizes positive emotional consequences, i.e. a sense of liberation. Corner et al. (2017) go further and provide initial empirical evidence that the intense grief described by Cope, along with subsequent recovery, does not apply in every case. Instead, they discover patterns of resilience, i.e. the ability to maintain relatively stable, healthy levels of psychological and emotional functioning over time after experiencing trauma or severe loss (Bonanno, 2004, 2005; Eicher et al., 2015). Indeed, according to Corner et al. (2017), there exist cases where entrepreneurs do not experience a strong emotional impact at all or report a rapid disappearance of negative emotions without extensive recovery efforts. Similarly, a recent quantitative study by Shore et al. (2023) finds no relationship between the extent of failure experience and the level of negative emotional response. Concerning the latter, Amankwah-Amoah et al. (2018) point out that the causes of grief need not directly result from the loss of the business but may also arise from perceptions of negative consequences for the family environment. Addressing the broader social environment, Singh et al. (2015) argue that stigmatization does not begin with the actual failure but should be seen as a process that already starts with the descent into failure.

2.4 The recovery from failure

The fourth timeframe, “recovery”, targets the rehabilitation from the failure and associated learning processes. Cope illustrates the recovery phase as a gradual process of healing and learning. He builds on and empirically supports Shepherd's (2003) assumption that entrepreneurs should oscillate between focusing on emotional healing (restoration orientation) and problem analysis (loss orientation) in order to recover effectively. From a temporal perspective, he finds that a stage of healing (typically even initiated by completely stepping back for some time) is necessary before meaningful critical reflection can occur.

The patterns and interrelationships of reflection, emotions, and other factors and entrepreneurial learning were also central to the research that emerged after Cope. Wei et al. (2019) summarized several potential individual, corporate, and environmental factors influencing the reflection and learning process. Particularly interesting are novel and partly contradictory findings on emotional recovery patterns. Corner et al. (2017) identified cases of resilient entrepreneurs who reported far milder and shorter periods of negative emotional impact from failure and consequently did not require recovery, as outlined by Cope. Singh et al. (2015) discovered an alternative mechanism of emotional coping. Instead of a gradual reduction of negative emotions through oscillation, the authors found so-called epiphanies, i.e. crucial moments of insight that provide the foundation for new knowledge. Another recent discussion concerns the influence of personality factors – especially narcissism – on reflection. Findings indicate that narcists engage in selective information processing and deny their failures, resulting in limited reflection and learning (Liu et al., 2019, 2021; Navis and Ozbek, 2016).

2.5 The re-emergence from failure

The fifth timeframe, “re-emergence from failure”, entails the application of generated learning contents in future actions. Cope (2011) provides rich empirical evidence on the contents (that might subsequently be applied) using a framework featuring four categories of learning outcomes (Cope, 2005): learning about oneself, learning about the venture (and its demise), learning about networks and relationships, and learning about venture management.

Later studies specialized in specific learning outcomes like changes in risk behavior (Dias and Teixeira, 2017) and the awareness of more social acceptance of failure (Heinze, 2013). Furthermore, Walsh and Cunningham (2017) provided a link between attribution patterns and learning outcomes. They argue that internal failure attribution fosters predominantly learning about oneself, whereas external attribution primarily promotes learning about the venture and its social environment.

While insights into learning outcomes are relevant, they only represent an interim result, as experiential learning theory requires learning outcomes to be applied in future actions to close the learning cycle. Similarly, the timeframe “re-emergence” ultimately analyzes the application of learnings in subsequent action. Integrating the application of learning contents is essential for evaluating the ultimate value of failure for learning and subsequent activities. Cope highlights that there exist different ways of application to be discovered. However, he does not provide empirical insights but urges future research to proceed.

The awareness of the pivotal role of learning application has been visible in the literature both before and after Cope (Amankwah-Amoah et al., 2018; Boso et al., 2019; Shepherd, 2003; Ucbasaran et al., 2013; Walsh and Cunningham, 2017; Widiawan and Igel, 2023). While – as Amankwah-Amoah et al. (2022a) and Burton et al. (2016) illustrate – there exist several pathways of re-emergence, research primarily focuses on entrepreneurial re-engagement as an application mode (i.e. the application of learnings to subsequent entrepreneurial endeavors). In particular, the conditions for re-emergence and the influence of learning on the new venture's performance were of greater interest. Relating the former, research discussed a range of factors that can influence the decision to re-engage in entrepreneurship: positive prior founding experience (Amaral et al., 2011), perceived controllability (Williams et al., 2020), individual demographic factors (Hessels et al., 2011; Stam et al., 2008), traits like extraversion (Widiawan and Igel, 2023), national and institutional contexts (Autio and Acs, 2010; Candon et al., 2023; Guerrero and Espinoza-Benavides, 2021; Hessels et al., 2011; Simmons et al., 2014), overconfidence (Nielsen and Sarasvathy, 2016), entrepreneurial self-efficacy (Hsu et al., 2017), digital technology capability (Wang et al., 2023), and an interplay of emotional and attributional responses (Williams et al., 2020).

Regarding the latter, Heinze (2013) finds exploratory evidence for the successful application of learning contents in subsequent ventures. In a similar vein, Boso et al. (2019) highlight learning as the central facilitator for an advanced performance when opening a new business. Initial ideas on how learning influences behavior can be found in recent works of Lin et al. (2019) and Amankwah-Amoah et al. (2018). Lin et al. (2019) show how learning influences strategy upon re-emergence, finding that learning can guide entrepreneurs to more proactive strategic actions. Amankwah-Amoah et al. (2018) explore how failed African Entrepreneurs learn, re-emerge, and imprint their learnings in the new company’s culture. Despite these works, an extensive collection of applicable learning contents is missing so far.

Moreover, research also demonstrates limitations of the applicability of learning. Eggers and Song (2015) discover that serial entrepreneurs are likely to blame their external environment for failing and consequently change industry upon re-emergence, diminishing the applicability of industry-related learnings. In contrast, the leadership style remains unchanged given the external failure attribution. Amankwah-Amoah et al. (2022b) provide a differentiated picture of the learnings' value for re-emergence, distinguishing varying stages (pre-foundation, formation, post-formation) of re-emergence.

Regarding non-entrepreneurial application, especially the highly relevant application of learning in employed work settings, there exists little research so far. The awareness that the transition from self-employment to employed settings represents a fruitful research context has only recently been raised (Burton et al., 2016; Debrulle, 2016; Jenkins, 2022; Mattes, 2016; Omorede, 2021). Early empirical work predominantly adopts an employer perspective and reflects on how employers perceive previously failed entrepreneurial episodes in the hiring process (Jenkins et al., 2023; Küsshauer and Baum, 2023; Waddingham et al., 2022). In contrast, the value of learning for re-emerging ex-entrepreneurs is barely researched. So far, only Rieger et al. (2023) provide some insights, arguing that previously failed entrepreneurs can have a career advantage over graduates who started directly as employees when changing to employed settings. This pertains to cases of adopting higher-hierarchy positions that require more general human capital combined with a rather short failed entrepreneurial episode.

3. Method

3.1 Research design

This study adopts a qualitative methodology, which is an appropriate strategy for revealing the complex and dynamic nature of underresearched areas and, thus, provides a better and more detailed understanding of the issue (Ghauri and Grønhaug, 2010; Suddaby et al., 2015). This is particularly true for a phenomenon that reflects the variety of participants' subjective “lived experiences” (Miles and Huberman, 1994).

The design of this study was informed by interpretative phenomenological analysis (IPA), which aims to understand lived experiences through the analysis of hermeneutics, that is, the meaning that individuals attach to the experience under interrogation (Griffin and May, 2018). According to IPA, context – the situated nature of the experience, involving, for example, social or psychological factors – contributes significantly to a better understanding of the phenomenon under study and is indispensable to consider in the analysis.

3.2 Research setting and data collection

This study builds on a purposive sample of 18 entrepreneurs who experienced business failure, i.e. the “termination of a business that has fallen short of its goals”. A research collaboration with six of the seven business incubators at Austrian universities allowed reaching entrepreneurs from knowledge-intensive industries who were willing to speak openly about their failure. Table 1 provides an overview of the study participants and relevant personal and venture-related information.

Primary data were obtained through semi-structured interviews. The interviews were conducted between February and April of 2017 and predominantly lasted more than one hour. Interviewees were asked to openly retell their story from the descent into failure until re-emergence. The learning timeframes of Cope acted thereby as guiding interview questions. In order to be well prepared for these sensitive interviews, the designed interview schedule was pre-tested. Interviews were voice-recorded and transcribed verbatim. In addition to the interviews, personal accounts of the interviewers and data from business incubators on the cases studied that were useful for the IPA analysis were considered.

3.3 Data analysis

Data analysis was informed by Smith and Osborn (2008). Accordingly, after the close-reading phase of one data item, a stepwise approach was taken first to describe (emergent themes) and then to interpret (superordinate themes) data within that one case. The former involved noting the researcher's initial thoughts and comments associated with the identified themes in the data item. The latter involved looking for patterns of connection between emergent themes. Thus, the themes identified were further clustered and grouped under higher-level (more abstract or theoretical) themes. It was an iterative process in which the superordinate themes were checked against the primary source material for their closeness to the actual words of the participants and their own sensemaking about the experience under study. After such a detailed, in-depth analysis of one data item was completed, the procedure was repeated with other cases. Again, the developed structure of superordinate and higher-level themes was iteratively updated. As a last step, the data was assigned to the five learning timeframes of Cope. Below, the final structure is presented, featuring one illustrative statement per superordinate theme (see Table 2).

During data analysis, careful attention was paid to the reflexivity of research (Bourdieu and Wacquant, 1992), meaning that interpretations were always checked for subjective biases of the researchers involved. In intense disputes, care was always taken to emphasize the informants' voices and not to shadow them with one's own worldviews and preconceptions about the phenomenon.

4. Findings

4.1 Descent into failure

The descent into failure took a variety of patterns and was subject to various causes and triggers among study participants. The descent paths ranged from the gradual realization that set goals were unachievable and stepwise counteracting measures to abrupt stalls without many intervention opportunities for the entrepreneurs. Gradual descents over a longer time were often linked to insufficient customer bases (Walter, Paul, Edgar). As Edgar explains, “one still hopes that it will become better, i.e. that customers do not only declare interest in but really purchase the product”. In this hope, the entrepreneurs procrastinated necessary exit decisions until becoming increasingly disillusioned. A similar pattern is visible in young companies that require financing to enter and accelerate a stage of growth (Daniel). In contrast to these gradual descents into failure, there also existed cases featuring an abrupt stall pattern, i.e. cases of immediate business deaths. Quinn, for example, experienced the descent into failure as “a lightning death if there is such a thing. […]”. He had invested nearly one million euros in developing a software architecture that subsequently proved unable to handle the required data amounts, with no chance of upgrading the software's capabilities due to technical reasons.

Among the various triggers and causes attributed to the descent into failure were insufficient funding, a lack and the loss of customers, as well as setbacks in product development most frequently reported. In terms of funding issues, entrepreneurs primarily blamed banks and investors for rigid and restrictive policies. Frank claimed that “banks were very inflexible and said they did not want to take this risk […] and that investors were hard to find within this short amount of time.” While Alina felt that receiving funding was problematic from the very beginning of the entrepreneurial journey, others experienced funding problems when attempting to grow. Olivia, for example, recognized that “without a business angel […] there is no real chance to approach the existing market in a way that allows us to become known”. Similarly, Michael ran out of cash when commercializing his product. Frank, who had already been promised follow-up financing after a long development time, experienced that this promise did not hold. He consequently negotiated with other potential investors, yet with clear ideas about the conditions, which finally impeded the closing of a financing agreement. As far as the relationship with customers is concerned, problems in gaining and keeping customers were thematized. In particular, cases in which key customers canceled their orders were especially critical and instantly led to failure. For example, Paul was confronted with the postponement of a substantial order by a public health institution. This meant losing most of his turnover – a turnover featuring an exceptionally high contribution margin. The ensuing financing gap could not be covered fast enough, precipitating the company's bankruptcy. Also, Kevin lost hundreds of thousands in sales in one fell swoop when a well-known beverage chain cut its budget. Even if orders held as planned, some entrepreneurs were confronted with a more and more deteriorating frequency of payments, causing cash flow issues leading to failure. The third range of primary failure triggers is linked to product development. Among the failures are cases where the intended product offering turned out to be technically not feasible. Lucas described: “Well, the product development was started, and before this, the technical feasibility study took place. And well … we developed the entire concept, and then the result of the feasibility study was basically that it was unviable.” Particularly problematic were negative feasibility results in the medical field. Nicolas, pursuing a business idea in medical technology, highlighted the issue of very long development cycles and limited options for lean start-up-like market tests.

4.2 Experience of managing failure

The experience of managing failure was characterized by manifold and effort-taking activities to counteract or delay the company's closure. Indeed, most study participants struggled with the decision to close their businesses. Edgar illustrates the overall sentiment, stating, “Well, now you have been pursuing your idea so far – you simply cannot give up now, you have to go on … there is this inner pressure to keep going”. Instead of giving up, entrepreneurs further increased their efforts and took a range of measures, including the search for additional funding, cost reduction, and the generation of additional revenues.

As with Alina, Frank, Paul, Christian, Michael, and Quinn, a considerable amount of study participants reported on desperate attempts to convince existing and new investors to support their companies financially. Kevin tried to become more efficient by cost reductions, dismissing employees, and focusing on different customer segments. On the revenue side, Lucas and colleagues, whose idea was a peer-to-peer platform for art, started selling their own products to bridge the time gap until his platform became better known by potential suppliers. Paul, realizing that the Austrian market might not be ready for his healthcare offering, tried to approach customers in Germany and Slovenia. Also, Nicolas, who was confronted with his primary distribution partner not fulfilling the contract, searched intensively for new distribution partners.

All the abovementioned measures have in common that they could – at most – delay the failure. When closing the firm, study participants, of course, had to take different, pre-defined measures – ranging from the filing for bankruptcy with following court cases also affecting their private assets (like in the case of Paul) to the mere fulfilling of existing duties and closing the firm (like in the case of Lucas).

In retrospect, the sentiment exists that managing the failure was extraordinarily stressful and that an earlier acceptance of the exit would have been advisable. Apart from actions targeted at rescuing the failing company, some entrepreneurs focused on securing their personal financial survival, e.g. by accepting part-time positions in another firm. Walter, for example, started to work at the company where he is now an attorney to feed himself and his family. Michael took steps to secure his business idea for the future by buying the rights out of insolvency.

4.3 Aftermath of failure

The failure impacted the study participants in multifaceted ways. Significant effects were perceived in terms of their financial, emotional, and social situation. Indeed, most entrepreneurs experienced a financial loss, persistently challenging their personal and professional life. Paul explained: “I am still struggling with the consequences, as one can imagine. I managed to get a settlement out of court with my creditors, and I'm paying them in installments over the next seven years … just like in the case of classical bankruptcy, at a rate of 20%.” Frank, too, was confronted with fundamental financial questions like how to pay his rent and finance his next project.

Apart from financial issues, findings reveal a variety of negative emotions like grief, anger, and exhaustion. Kevin was particularly strongly impacted, experiencing sleepless nights. Less impacted but still sad was Hanna, stating that until today she finds her entrepreneurial project exciting and that, in retrospect, she looks back at an enjoyable and exciting time. Hanna, like about one-third of all study participants, also experienced some positive emotional effects. As she states: “well, at the time of the exit, there was somehow also a big relief”. Although the emotional impact was typically felt strongest directly after the failure, certain study participants showed delayed emotional reactions. Barbara, for example, did not realize how sad she was about losing her company until talking about her failure at an entrepreneurial failure event.

A differentiated picture in terms of effect direction is also recognizable regarding the social environment. Most entrepreneurs experienced positive social consequences after business failure, particularly stemming from their closer personal circle (family, friends). Christian contested that “especially friends reacted totally positive”. The reactions ranged from – “if you need someone, just call us” – to “help offers in terms of financial resources or job search”. In addition, although participants reported an anti-failure culture in society, they did not experience the stigma of failure on the individual level. In team-founded firms, the entrepreneurs expressed both negative (e.g. no contact with each other) and positive (e.g. still working together) effects of social relationships with their team members.

4.4 Recovery from failure

The recovery from failure turned out to be highly complex, with an extremely variable process of emotional healing, reflection, and striving for subsequent action. Study participants employed varying emotional healing strategies focused on down-regulating negative emotions like anger and frustration. Kevin, for example, suffering a psychological crisis, started to make diary-like recordings. Others took a time-out to relieve the stress or tried to suppress the failure. Igor, who took a half year off, found this time for himself to be very rewarding. Paul instead attempted to overcome the failure as fast as possible by suppressing it. He stated that he did not want “to hear anything more about the failure” and just looked into the future. This suppression was not always a free decision. Instead, for Paul and others, pressing economic problems such as paying off the debt or covering the living costs made searching for a new job the number one priority.

Besides emotional healing, the recovery stage was characterized by reflection efforts targeted to make sense of the failure. This reflection varied widely from person to person. Some very deliberately started an intense reflection on what had happened. Christian, for example, reported, “I very much engaged in looking into the back mirror: what were the mistakes that we made”. Others – as illustrated above for the case of Paul – tried to limit reflection efforts. Pertaining to its relationship to emotional healing, some cases supported the notion that recovery is essential before reflection can happen. Edgar, for example, first repressed thinking of what had happened. After about six to twelve months, during which he stabilized emotionally, he felt able to engage in reflection. However, empirical evidence also includes cases where foregoing emotional recovery was unnecessary (e.g. in the case of Gerhard, who – after closing the business immediately continued his university studies) or possible (e.g. in the case of John, who directly changed into a new job).

Resulting from the strongly varying emotional and professional situations, the point in time when reflection took place was highly heterogeneous among study participants. Hanna already started intense reflection during the descent into failure. Also Nicolas stated: “We reflected very early”. Others reflected after emotional recovery, and in one case, reflection was even delayed until the conceptualization of a new venture: Lucas reported to have thought about “which failures not to repeat” only once he decided to form a new business.

As heterogeneous as the time patterns were also the triggers for reflection. They include negative and positive events that draw the attention of the failed entrepreneur (back) to the failure. Quinn felt that especially the difficulties after the failure led him to reflect. Similarly, Hanna reported that intense reflection was triggered once she was confronted with debts by unknown creditors. For Barbara, preparing a talk about her failure at a dedicated event let her think in detail about what had happened. In the case of Edgar and John, the recognition of similar business ideas became triggers for reflection. Edgar recounted two separate occurrences where he became aware of a similar business idea that brought him back to reflecting on his business. Interestingly, Edgar and John reported having felt happy for their successful peers, even if their own endeavors had failed.

4.5 Re-emergence from failure

Learning research suggests that the application of learning contents in subsequent activities (e.g. opening a new business or changing to an employed work setting) is influenced by learning outcomes resulting from preceding timeframes in the learning process.

4.5.1 Learning outcomes

The study participants reported a large variety of learning outcomes related to their failure, which are presented below following the conceptualization of Cope.

Learning about oneself primarily featured changes in the readiness to assume risks and the coping with stress. Regarding readiness to take risks, most participants reflected that they had become more risk-friendly. This tendency might result from two streams of learning. First, entrepreneurs discovered an increased acceptance of risk to be an essential key to success. Edgar learned to have “more courage to take risks – also to deliberately take a risk and consciously ignore it”. Second, experiencing failure familiarizes entrepreneurs with the feared consequences of failure, showing them that life continues. Olivia mentions, “I do not fear the risk of what can happen as I have gone through it already – and this was a pure learning process”. Concerning dealing with stress, participants reported having become hardened. As a consequence, they felt better able to work under stressful conditions.

Learning about the venture and its demise primarily refers to improved awareness of the venture's strengths and weaknesses, including reasons for its failure. Study participants were able to provide a reflected assessment of the strengths and weaknesses of their ventures. Given that their entrepreneurial episodes had failed, the overall evaluation of the venture was rather negatively connotated. As illustrated in the section “descent into failure”, financial issues and an inadequate product-market fit were most frequently addressed as critical factors for success or – in this case – the failure. Nevertheless, entrepreneurs also reported several strengths they perceived about their ventures. Especially the business idea itself was – despite the failure – often seen in a positive light. Frank, who aimed to launch an innovative snack box, concludes, “the idea itself is actually still great as it is very easy to demonstrate”.

Learning about the social environment resulting from the failure refers to various levels, including the founding team, networks, and the wider environment. Regarding the founding team, study participants unanimously agreed that team constellation requires careful planning. While agreeing on the sharing-related advantages teamwork provides, some participants stressed possible limitations due to inflexibility. Concerning the team members' professional backgrounds, interviewees highlighted the importance of expert knowledge. The team should cover all relevant fields of expertise without extensive redundancies. In any case, team constellation should not solely be based on sympathy but should also account for a strong emphasis on members' professional backgrounds. Regarding networks, interviewees generally agreed on their importance. However, some study participants claimed networks to be overrated, given that entrepreneurs might receive positive feedback without real subsequent support. In order to prevent “being stuck” in many less supportive networks, participants learned to become selective and be present in just a tiny range of networks, holding closer, regular contact with respective partners. With regard to the broader business environment, entrepreneurs expressed rather negative associations. This particularly applies to the entrepreneurial ecosystem. Study participants recognized a tension between a start-up hype on the one side and an anti-failure culture on the other side. During and after failure, interviewees felt considerable stigmatization from the public. The negative attitude towards failure led some entrepreneurs to avoid further activity and investment.

Learning about venture management comprises knowledge gains about running and controlling businesses more effectively. This study found project- and strategic management, product development, and user integration as the most prominent areas of learning. With regard to project management, entrepreneurs reported improvements in planning and managing deadlines and milestones, as well as financing the projects. As far as strategic management is concerned, interviewees highlighted the importance of organic and stepwise growth. They now see a small-scale start combined with extensive and continuous market research as essential based on their experiences. Further, the study participants claimed that a conscious choice of strategic partners and orders is indispensable. Concerning product development, the participants recognized product-market fit as a critical factor for success. Some reported a lack of customer focus stemming from neglecting real consumer needs (as opposed to technical details) and comprehensive market approaches. Consequently, they intend to follow a “lean start-up”-like approach in their future endeavors. This should comprise early consultation of target groups using minimum viable versions of one's product.

4.5.2 Application of learning contents in subsequent activities

The re-emergence stories of the study participants provide rich evidence for the application of learning contents in subsequent activities. However, the degree to which learnings could be utilized varied considerably between those who re-emerged as entrepreneurs and those who entered employed work settings.

The five study participants who re-emerged as entrepreneurs unanimously reported that they could and did successfully apply learning outcomes. Kevin states that the opportunity to apply learnings from the failure effectively was a key rationale for his reentry into entrepreneurship: “It would be a pity not to use the experiences and know-how gathered, and why should I gift them to an employer?” The learning outcomes that proved particularly relevant upon re-emergence fall into three groups. The first entails strategic cornerstones of the new foundation, like the team size and the venture's legal frame. Gerhard and Lucas, who previously operated in larger team contexts (six persons), chose to re-emerge only in a team of two persons. In light of personal disappointments in the failed venturing spell, Hanna paid close attention to a watertight syndicate agreement. She thereby took care to consider not only the welfare of the company but also her own. A second aspect strongly influenced by learning outcomes was the behavior during the start-up process, which tended to be “leaner” and more resource-focused than previously. Michael, who did lots of networking in the start-up community in his prior business, significantly reduced these efforts and instead focused on product development and intensive market tests. Gerhard tried to include fast feedback rounds in his product development circle and consistently increased the speed of feedback loops. Lucas introduced the principle that “product-market fit comes first now […] minimum viable products and lean market tests … no programming prior to that”. A third application area of learning contents is daily operations, like communication and customer relations. In this regard, Hanna felt that she considerably improved her way of approaching customers. Kevin, unlike in his former business, did not enter legal battles with customers in case of minor pending payments.

Apart from founding a new company, also employed working settings represent interesting contexts for applying failure-based learnings. In the present study, 12 failed entrepreneurs chose this option and closed dependent working contracts. One further study participant was still searching for a job. The rationale behind an employed working relationship was the wish for a safe and regular income. As Barbara states, “I can't stand financial uncertainty at the moment […] I don't have to have this insecurity anymore for a few years”. Nevertheless, the clear majority of now employed could very well imagine re-emerging as entrepreneurs somewhen in the future – yet under certain conditions that are too linked to the experience of failure, e.g. in case of clearly defined paybacks (Edgar), a waterproof business model (Barbara) and a collaboration with a partner (Igor).

In contrast to re-emerged entrepreneurs, study participants who decided for employed work settings show heterogeneous opinions about the applicability of learnings. This heterogeneity ranges from Walter, who, for his current position as an authorized signatory, states, “I have learned so much and now, of course, fully apply these learnings”, to Alina, who, as a process engineer, can hardly draw on prior experience. In total, eight participants in employed settings have a clearly or rather positive perception of the ability and utility of learning application. Most of the reported learnings to be applied relate to dealing with risk and the bearing of leadership responsibilities. Barbara feels to have improved in terms of risk assessment. John learned to rest his mind despite difficult and risky situations: “Well, do I have to sleep badly because of something like that? Or can I say after coming home, I am going to handle it (the next day)?”. In the case of Christian, the experience of failure resulted in an improved understanding for the burden of carrying risks. He states: “Today it is very easy for me to talk to my boss at eye level when it comes to entrepreneurial risks as I experienced everything on my own and know how this feels to an entrepreneur.” In addition to the aforementioned learning contents, a range of further learnings was reported to be applied. However, these learnings rather resulted from routine experiences during the failed episode. Among them are cost and financial planning, price negotiation, business planning, and leadership skills. Unlike those failed entrepreneurs who re-emerged, the group of employed and unemployed study participants also contains cases where participants explicitly deny the applicability of learning outcomes. In the case of Frank, who is unemployed and searching for a job, and Nicolas, who just started a new profession, the application context has been missing so far. In other cases, the context was different in terms of industry or did not require entrepreneurial skills given the job characteristics (e.g. occupations as IT specialists – Alina and Daniel or medical doctor – Paul).

5. Discussion

The present study's findings spread along all five of Cope's learning timeframes as it seeks to take a holistic lens. Of particular interest are the insights into the hitherto underresearched early stages of descending into and managing the failure and the final process stage of applying learnings in novel entrepreneurial and employed settings. Furthermore, the relationship of the timeframes calls for closer analysis.

Findings showed that the beginning of the learning cycle (i.e. the descent into the failure and related management actions; learning timeframes 1 and 2), can feature a variety of descent patterns – ranging from a stepwise decline to a “shock death”. In the first case, the entrepreneurs gradually realized that the business persistently remained below expectations, eventually increasing the pressure to cease operations. Findings that entrepreneurs tried to balance financial costs and emotional impact support the propositions on emotion-regulating anticipatory grieving by Shepherd et al. (2009). Moreover, insights demonstrate escalation of commitment tendencies (Yamakawa and Cardon, 2017), as study participants lamented the extensive resources they had invested in the failing company. In the second case, entrepreneurs suffered a sudden death of their venture due to an unexpected withdrawal of a key investor or customer or because of a technical breakdown. Interestingly, considering the learning outcomes from the failure, findings imply that both descent patterns (and the nuances in between) enable learning. Thus, despite their mitigating effects on the failure's perceived emotional impact (Shepherd et al., 2009), gradual declines appear to trigger sufficient alertness to enable critical reflection. Similarly, sudden deaths – while requiring stronger efforts of post-failure recovery – represent experiences that entrepreneurs can reflect on.

Regarding the perceived failure causes, the present study finds – in line with former research (Romanelli, 1989; Shepherd, 2003; Shepherd et al., 2000) – a large variety of perceived reasons why the ventures failed. These reasons were predominantly found in the external environment (Eggers and Song, 2015). Many interviewees mentioned insufficient funding, lacks and losses of customers, as well as setbacks in product development, which appeared to be the primary triggers of failure and to provoke both – gradual and radical declines.

Findings regarding the management of failure reflect procrastination tendencies manifested in high levels of personal effort (Shepherd et al., 2009; Yamakawa and Cardon, 2017). The intensive actions to fight the failure and mitigate its impact align with the ideas of Stambaugh and Mitchell (2018). Adding to the existing research, this study provides insights into the concrete management measures applied, i.e. cost reduction, seeking further revenue sources and investors. Moreover, findings reveal that entrepreneurs perceived the descent into failure and the related management situation not only as extraordinarily stressful but retrospectively as something to be avoided.

The present study reveals considerable differences in the applicability of learning contents (timeframe 5) depending on the mode of re-emergence. In the case of re-engagement in entrepreneurship (i.e. opening a new business), previously failed entrepreneurs extensively use the knowledge that resulted from the failure (and its preceding episode). Thus, recently assumed learning-based performance advantages upon re-emergence (Boso et al., 2019) appear plausible. Moreover, findings yield evidence for more proactive strategic actions (Lin et al., 2019), like a clear legal structure and a very close-to-market lean startup approach. In line with the theorizing of Eggers and Song (2015), four out of five entrepreneurs who started new businesses changed the industry. However, as all industries were strongly linked to the previous ones, no strong evidence for a loss of industry-specific knowledge was found. Following Amankwah-Amoah et al. (2022b), who differentiated three stages of re-engagement in entrepreneurship, this study finds applicable learning contents for each stage. During pre-foundation, strategic cornerstones like the chosen team size and legal agreements were strongly influenced by learnings. During formation and development, a “leaner” approach resulted from insights taken from the failure. In the post-formation stage, changes in daily operations like customer relations and communication as well as more selective networking show visible links to learnings.

In the case of a transition into employed work settings – which, in line with Koch et al. (2021) and Marshall (2016) – turned out to be a highly relevant application context, failed entrepreneurs reported strongly varying degrees to which they could and did apply learning contents from failure. Those who profited the most from failure-based learnings were entrepreneurs who took over senior positions in companies, supporting recent findings of Rieger et al. (2023). In these cases, especially risk management and the bearing of leadership responsibilities benefitted from failure-based learning. De-novo employees without senior positions reported learning benefits in terms of project work and customer interaction. However, these aspects mainly relate to lower-level learnings derived from routine activity during the past entrepreneurial episode. Failed entrepreneurs who found no or almost no applicability of learnings are characterized by working in industries far from their former business or contexts without an entrepreneurial nature (e.g. medical doctor). Nonetheless, the experience of failure was an essential source of self-development for all participants. Risk assessment, self-confidence and stress resistance were among the most important skills that are transferable to both re-emergence contexts.

Pertaining to the relationship of the learning timeframes, the study's findings indicate that the process may have a non-linear character. The timeframes can overlap, loop, appear in different orders, or even (partly) lack in the learning process. Cases of overlapping timeframes included a descent into failure where the entrepreneur already experienced and managed extensive negative emotions and reflection efforts. Further, a re-engagement in entrepreneurship was found to be accompanied by intense reflection on the previous failure. Loops became visible when entrepreneurs who had recovered and re-emerged were reminded of their failed business and started thinking and grieving about their failure again. Differing sequences of how the learning process is undergone were mostly related to situations where entrepreneurs had to re-emerge for financial reasons as fast as possible, leaving little space for recovery and reflection at first. Higher-level learning requires going through these important emotional and reflective processes (Cope, 2005). Also, the experience of grief after the loss, if the failure was felt that way, needs to be encountered for emotional integration (meaning finding) and growth (Shepherd, 2003). In the two contexts mentioned, failed entrepreneurs returned later to recover and reflect on the failure, making the research more sensitive to the learning process's temporal component. In the case of a firm closure associated with a relatively limited negative financial impact, the confronted entrepreneur reported very little necessary coping, representing an example of a timeframe that only partly materialized. These examples show that Cope's timeframes represent a good generic concept to describe the failure-based learning process but that there exist cases where individual learning processes differ in terms of the sequence and extent to which the timeframes appear. Reasons include the considerable heterogeneity in person-related factors (e.g. emotional resilience, openness to reflect) and factual necessities (e.g. the requirement to instantly search for a job, limiting reflection possibilities).

6. Conclusion

The present study provides essential contributions to research and interesting implications for practice.

6.1 Research contribution

Following Cope's vision of a holistic understanding of entrepreneurial learning from failure, the present study yields essential contributions beyond his and later research. The choice to take a holistic view on entrepreneurial learning from failure opens up several levels of investigation of this phenomenon: (1) as a whole process with all its stages; (2) as a complex and dynamic process with the interdependencies between the different stages; (3) as an integral whole with its intrinsic purpose; and (4) as a part of a broader social and societal context (a collective level).

Consequently, the study first covers the entire process with all five learning timeframes from failure. In this way, it also explores and provides insights into those timeframes that had been key research gaps in the learning process so far: descent into failure, management experience of failure, and re-emergence from failure via paid employment. Findings on heterogeneous descent patterns, concrete management practices, and varying degrees of applicability depending on the mode of re-emergence (entrepreneurial vs paid employment) enlighten hitherto blind spots of the learning process.

Secondly, the research detects important non-linear interdependencies between different timeframes. Empirical evidence shows that there can be overlaps and loops between timeframes and that timeframes can appear in different orders or be (partially or entirely) absent from the learning process. These dynamics can be traced back to the extent of the failure, personal situation, and personal characteristics. While existing research had hitherto, at best, assumed that learning does not linearly follow the proposed stages (Cope, 2011), the present study is the first to provide empirical evidence on how learning paths may differ.

When examining human behavior - such as the learning process - as an integral whole, it helps to keep in mind the intrinsic function that this process must fulfill. In the context of entrepreneurial learning from failure, it becomes important what the entrepreneur applies to the new context from what he or she has learned from the failure itself, management processes, and associated personal growth. The phase of re-emergence becomes crucial here. Entrepreneurship research so far has been primarily interested in whether the entrepreneur has founded a new company and which learning effects he or she applies in the new company (Amankwah-Amoah et al., 2018; Boso et al., 2019; Heinze, 2013; Lin et al., 2019; Stam et al., 2008). However, a holistic perspective broadens the scope of this analysis. The study showed that many of the entrepreneurs who had not started a new company were able to utilize numerous learning outcomes in a new professional context. The acquired entrepreneurial skills and mindset as well as personal growth experienced, constitute valuable assets for companies employing the previously failed entrepreneurs. Already viewing entrepreneurial failure as one of the critical events from which valuable learning can be gained (Cope, 2003) makes it easier to view the applicability of learning effects through a broad lens and not limit it to a new entrepreneurial episode. It seems important to speak of learning from entrepreneurial failure (as opposed to a narrower version of entrepreneurial learning from failure) so that the transfer of learning effects to diverse professional contexts can be directly recognized and addressed.

Finally, considering individual learning processes a collective experience, important learning effects arise in a social context. The study results show a picture of strong social support from family and friends in the event of failure itself and in recovery. The experience of failure can thus be a collective experience of solidarity and crucial social support to overcome a crisis together; an experience that fosters greater trust and hence generates a collective good (Putnam et al., 1993). Trust is an essential social capital (Crowley and Barlow, 2022) that impacts a sense of belonging and connectedness and is an important collective good for the context of future crises and uncertainties, such as we are increasingly facing in Europe in recent years with the outbreak of the pandemic and the war in Ukraine, as well as of a new entrepreneurial episode.

6.2 Practical implications

First of all, the findings reinforce the encouraging remarks of Cope, Ucbasaran, Shepherd, and others: learning from failure is challenging but possible (Cope, 2011; Dias and Teixeira, 2017; Shepherd, 2003; Singh et al., 2007; Ucbasaran et al., 2013). This insight appears particularly comforting in times of multiple crises that make failure in entrepreneurship even more probable than before. Especially the recent COVID-19 pandemic challenged entrepreneurs as the entire humanity to a degree hitherto hard to imagine (Semerci, 2022), which entrepreneurship research had only started to explore (Afshan et al., 2021; Gupta et al., 2024). Revisiting this study's findings under the impression of the recent pandemic reveals that this difficult context yields both aspects that can hinder and facilitate learning from failure. A crisis can tempt people to unreflectively blame the problematic framework conditions for their failure. Additionally, the range of opportunities for re-emergence and, thus, the application of learning during times of crisis seems to be limited. On the other hand, Heinze (2013) argued in the aftermath of the financial crisis of 2008–2009 that failure had become more socially accepted, reducing stigmatization.

Detached from the current crisis context, however, the findings of this study have a number of other practical implications. Failed entrepreneurs can benefit most from their learnings when they continue with an entrepreneurial career or as employees in a leadership position. Independent from the mode of re-emergence, an industry close to the previous field of operation should be targeted. Another practical implication for entrepreneurs relates to the findings that critical reflection is central to learning and recovery but can happen at different points in time. This might comfort entrepreneurs who find themselves unable to reflect close to the failure of their business for emotional or professional reasons.

Policymakers may, too, take valuable lessons from this study's findings. First, insights strongly call for initiatives that support entrepreneurial re-engagement after failure (Schermuly et al., 2021). Specifically, results demonstrate not only that entrepreneurs do learn from failed endeavors but also show that opening a new business is a highly effective way to apply those learnings. Societal stigmatization of failed entrepreneurs hampers reflection and, thus, learning and the probability of successfully starting a new business. Policymakers could strive to improve the environmental conditions for entrepreneurs, including the fight against societal stigmatization and a better financial ecosystem for re-starters (Nielsen and Sarasvathy, 2016; Singh et al., 2015). When it comes to concrete actions for entrepreneurs during failure, providing psychological support in the grief and reflection phases could also be an important aid. Psychological assistance provides valuable emotional support to rise after the loss and can also act as a catalyst for the reflection process, which is essential for the learning process. Finally, policymakers should support initiatives targeting knowledge spillovers from failed entrepreneurs to latent and emergent entrepreneurs. Such initiatives may include internet forums, events, and other means representing platforms for failed entrepreneurs to talk about their learnings openly. In this way, the learning outcomes of those who change into employed positions also spill over to a larger audience.

6.3 Limitations and future research

Learning from failure is a highly challenging research area. Recruiting study participants to openly speak about failing, appropriately capturing the learning contents, and making sense of the complex influence factors on learning impose high demands on the researchers. While striving to meet these challenges, the present study yields some limitations, some of which are unavoidable and some yield potential for future researchers.

First, one may argue that ex-post interviews of failed entrepreneurs are subject to a retrospective bias (Frankish et al., 2012; Podsakoff and Organ, 1986; Walsh and Cunningham, 2016). While this argument cannot be entirely rejected, the chosen approach of ex-post phenomenological interviews is not only state-of-the-art in the field but was joined by additional measures to reduce the danger of a retrospective bias further. Failure as an ex-post facto phenomenon was not (Sheppard and Chowdhury, 2005) and still is not examined in real-time. To some extent, this may be due to the challenges that a longitudinal study on failure would pose: accompanying entrepreneurs in the expectation that some might fail and maintaining their cooperation even during the tough time of failing is a difficult endeavor. However, weighty arguments apart from feasibility exist that back the decision for ex-post interviews. Ex-post interviews can prevent biases that continuing interviews in real time may impose on reflection behavior, as interviews might trigger reflections that otherwise do not happen.

Moreover, the phenomenological approach of the study and narrative interview style let entrepreneurs “live through” their experience again and thereby facilitate a good recall even if some time has passed since the failure (Singh et al., 2015). In addition, failure's highly critical nature makes recalling experiences per se easier (Cope and Watts, 2000). What must be considered – and was considered – is adequate timing of the interviews. There should be some time passed since the failure to allow for recovery and reflection to emerge, and overly extensive or strongly differing timespans among study participants should be avoided. In this regard, Cope's study faced a drawback as some cases ranged back more than 20 years at the time of data collection. In contrast, in the present study, retrospective bias is limited as entrepreneurs experienced their exits on average three years before the interview, with seven years being the longest timespan.

Second, the extent and contents of learning derived from the study participants' perceptions. As Cassar and Craig (2009) stress – there may be a bias between what entrepreneurs believe they experience and what they actually experience. Moreover, in attempting to find something good about failure, entrepreneurs might talk themselves into believing in their learning progress (Bacon, 2016; Lattacher and Wdowiak, 2020; Nielsen and Sarasvathy, 2016). Despite these potential shortcomings, relying on individual perceptions still appears to be the best option available. Nevertheless, future research that moves beyond subjective perceptions could apply more objective measures, e.g. document analysis. Even more insightful would be to utilize and compare objectively designed pre-post tests of learning contents in case these tests are conducted, for instance, in the context of accelerator programs.

Third, this study was exploratory, providing interesting novel insights into the learning process from failure. However, further research is needed on the relationships between the timeframes in a longer perspective. For instance, entrepreneurs may not be able to experience the different stages of the learning process during failure but return to them at a later stage to complete the learning circle. Applying different theoretical frameworks (e.g. Kolb’s learning circle, Affective Events Theory according to He et al.) could provide interesting perspectives in this regard. The application of learning requires further research in at least two directions – on the one hand, a more nuanced understanding of the relevance of learning contents for different stages in re-emergence and, on the other hand, a particular focus on transitioning to paid employment. Lastly, a critical evaluation of the actual value of failure for entrepreneurial learning appears interesting as this study revealed that some reported learnings are of lower-level character and stem from gradual learning during the entrepreneurial episode.

Table 1

Study participants

#	Name	Age	Sex	Highest education	Number of companies founded so far	Sector	Year of foundation	Year of exit	Individual vs team foundation (number. of members)	Current occupation
1	Alina	54	f	University degree	2	Information technology	2010	2014	Team (2)	Employed
2	Barbara	46	f	University degree	1	Consulting and information technology	2008	2012	Individual	Employed
3	Christian	38	m	High school (technical focus)	1	Medical technology	2009	2015	Team (2)	Employed
4	Daniel	50	m	University degree	2	Plant engineering	2009	2012	Individual	Employed
5	Edgar	38	m	University degree	1	Software services	2012	2016	Team (2)	Employed
6	Frank	43	m	High school (technical focus)	2	Product development	2015	2017	Individual	No occupation
7	Gerhard	30	m	University degree	2	Information technology, e-commerce	2008	2010	Team (6)	Self-employed
8	Hanna	33	f	University degree (Ph.D.)	3	Publishing	2013	2017	Team (3)	Self-employed
9	Igor	40	m	University degree	1	Transport	2004	2013	Team (4)	Employed
10	John	30	m	University degree	1	Information technology, e-commerce	2008	2010	Team (6)	Employed
11	Kevin	45	m	University degree	6	Marketing and communication	2002	2012	Team (6)	Self-employed
12	Lucas	30	m	University degree	3	Information technology, e-commerce	2008	2010	Team (6)	Self-employed
13	Michael	43	m	University degree	3	Medical technology	2006	2015	Team (2)	Self-employed
14	Nicolas	39	m	High school (technical focus)	2	Development and distribution of consumer goods	2010	2016	Team (4)	Undisclosed
15	Olivia	32	f	University degree	1	Information technology, e-commerce	2008	2010	Team (6)	Employed
16	Paul	49	m	University degree	2	Information technology, consulting in health care	2009	2017	Team (2)	Employed
17	Quinn	46	m	High school and advanced training	1	Software development	2010	2013	Team (2)	Employed
18	Roland	41	m	University degree	1	Information technology, software development	2005	2016	Team (2)	Employed

Source(s): Authors' own work

Table 2

Data analysis

Example quotes	Participant	Superordinate theme	Higher-level theme	Timeframe
Phew. So I don't think it's a point in time where you say “now I failed”, but it's one step, one step after the other and I think it's being procrastinated	Edgar	Gradual descent	Decline patterns	Descent into failure
At the end of 2014, an Austrian health insurance company postponed a very, very large order for us. This was almost 30 million euros in revenue for us with a very, very large contribution margin and it actually plunged us into a very large financing gap in the short term, which we were unable to close	Christian	Abrupt stall
So you say: yes, now it's failing because of the money. And then you just try to find another way. Find some other solution, find other partners, but then it couldn't be done	Alina	Funding issues	Triggers and causes
Large customers simply cut their budgets. So (company name), it felt like, had 300,000 euros in fee revenue a year, which was gone overnight	Kevin	Relationship with customers
Because the project is technically very complex and food databases don't work	Alina	Product development
In other words, the team had already noticed that two camps were forming due to different opinions and target groups, and the fact that everyone had the same number of voting rights - which was a bit of a disaster - meant that everyone was blocking each other	Lucas	Funding team problems
And then I was really angry	Frank	grief and rage	Feelings	Experience of managing failure
“Ah, now you've pursued your idea so far, now you can't give up, now you have to keep doing it!” and so on. So this external pressure to keep going	Edgar	hardship of closing
I think it was more this overload and double and triple burden that tore me down	Paul	stress
We actually realized immediately with this postponement of orders that we were facing a financing problem. We then actually started immediately. We contacted potential investors	Christian	counteracting: investor search	Firm-related measures
We reacted in an entrepreneurial way, which meant reducing costs, laying off employees, and focusing on other customer segments	Kevin	counteracting: cost reduction
So we approached a wide variety of markets and did not receive a prompt commitment to acquire the product or the license for the product in any market, let's put it that way	Paul	counteracting: market refocusation
I then started working at (company name) - the current company where I work as an authorized signatory - in parallel in 2011, simply to financially secure myself and my family	Roland	securing financial survival	Person-centered measures
We then bought out the assets with another investor, so to speak, and founded (company name)	Michael	securing opportunities
And in the end, we were left with the liabilities to the banks. Yes	Barbara	individual bankruptcy	Financial effects	Aftermath of failure
Of course, you were at the bottom financially and sometimes you didn't know how you were going to survive the next day	Paul	difficulties of meeting daily expenses
What's missing now more than before are the financial resources, which, of course, when you've been working on something for so many years on low pay, it really drains your savings	Alina	difficulties to re-start with a new project
I'm sad about it because I still find the project very exciting today and because it was also a very fun and interesting time for the most part	Hanna	grief	Emotional effects
And then I think I was quite annoyed about it for a while	Hanna	anger
My personal, psychological energy level was actually close to zero or negative	Edgar	exhaustion
So at the time of the exit, it was somehow a great relief	Hanna	relief
Our friends, in particular, reacted very positively. The reactions actually ranged from “if you need someone to be there for you, get in touch” to “can we help you financially or with finding a job or whatever”	Christian	support from friends	Social effects
The problem was more with one of the partners who was involved in the project. And there were actually problems with him	Igor	team conflicts
One person in that group is still one of my best friends today, or still is. And I'm glad that this no longer had an effect and I'm glad that we have a good relationship today	Olivia	team support
I also kept diary-like notes back then	Kevin	Measures: down-regulating emotions	Emotional healing	Recovery from failure
I took half a year off, which was very pleasant	Igor	Measures: taking time-out
Phew. Mmm. Suppression, I would say	Edgar	Measures: suppress failure
I don't think I wanted anything more to do with it in the first instance, because of course there was so much to do with the new company	Hanna	No healing possible
And then I actually looked at it very intensively in the rear-view mirror: what were the mistakes we actually made? It was actually clear to me that we didn't need to look for the mistakes somewhere else	Christian	Deliberate reflection	Reflection
No, I just want to meet my payments and that's it. What should I think about that? That's in the past, nobody cares	Paul	Limiting reflection
Hopefully no more weird bills from 2014 will pop up now	Hanna	negative follow-up-events	Triggers for Reflection
Well, reflection happens relatively frequent. It was at the Fuck-up Night (= event where failures are presented) …	Hanna	reporting events
… there was the innovation prize for two students who basically did the same thing and exhibited at Cebit and they won 50,000 euros, where you say “Mhm, great thing, quite nice, I could have had it too … ”	Edgar	recognition of similar business ideas
I think I've dealt with a lot of my frustration and anger about the behavior of not giving up these shares or changing any of my behavior, I think I've already dealt with it very strongly within the company	Hanna	early reflection	Timing of reflection
B: No, no. At first I tried to close it, but yes. I: And then a bit later after … B: Exactly, exactly, exactly. You just want to get it over with as quickly as possible, yes. You want to (unintelligible; 22:03) look back	Kevin	delayed reflection
That you also learn to deal with risk, so the topic of risk wasn't taught or shown either	Kevin	risk bearing ability	Learning about oneself	Re-emergence from failure
Stressful situations, that's a bit of a problem with the word, that you say: ok, you don't need to say stress in that sense, because then it becomes one. Yes, probably yes, but I can't say how to deal with it differently	Michael	stress bearing ability
Yes, so it was a whole series. For one thing, I got to know myself better. I learned what I can do well	Frank	improved self-awareness
If you take a look behind the scenes at why the company worked and grew the way it did, then you also understand the industry a bit	John	improved awareness of strengths and weaknesses	Learning about the venture and its demise
There was simply too little customer development	Frank	improved awareness about failure reasons
The division of roles, who does what, was also a lesson learned from the situation	Hanna	team: careful planning necessary	Learning about the social environment
However, it was very important to have someone who didn't do it all alone. Simply because sometimes you want to discuss things and don't want to make decisions alone, so it's easier if you have someone with you	Alina	team: awareness of advantages
You often see that there are teams with far too many people in them	Edgar	team: limitations due to inflexibility
Different composition, especially in terms of skills, is extremely important, i.e. the team of associates	Kevin	team: heterogeneous background important
I think what I've also learned is that you shouldn't start a company with someone or hire someone just because you like them a lot	Hanna	team: profession over friendship
That they are extremely important. So nothing works without networks and certainly not if you're self-employed	Igor	networks: important success factor
You shouldn't overestimate them	Michael	networks: overrated
There are some real phonies who pose as consultants and maybe even want money for it or boast about a network that then turns out to be very poor	Alina	networks: selective approach necessary
I think that the current start-up hype that prevails and where people are saying “Hey, we need a lot more entrepreneurs and we'll promote everything back and front”	Lucas	environment: start-up hype
It's unpleasant, it's still not respected in our society if you fail with something	Alina	environment: stigmatization of failure
During my time at the company, I learned how to set up efficient reporting and project management through my research work. I have certainly, how should I put it, matured a lot	Quinn	project-/strategic management	Learning about venture management
You always have to try out how something works and realize as quickly as possible, okay, it's not working, let's try something new	Gerhard	product development
I think that's an important point that I can recommend to every founder. So get the product out there as quickly as possible and get feedback, that's something you also take away from it	Gerhard	user integration
I have a syndicate agreement with my new partner, now based on my experience, parallel to the GmbH agreement, where we take shares out of the company	Hanna	strategic cornerstones	Applicable learning contents in entrepreneurship
You always have to try out how something works and realize as quickly as possible, okay, it's not working, let's try something new	Gerhard	leaner start-up process
Well, I think I'm better at external communication	Hanna	daily operations
When it comes to entrepreneurial risk, because I've simply felt it all first-hand and because I know, because I can also understand how an owner sometimes feels	Christian	understanding of entrepreneurial risk
But I would need the financial security I mentioned - I would need a waterproof business model	Barbara	waterproof business model	Requirements for re-transition to entrepreneurship
Okay, if at least the shareholder shares (unintelligible; 24:56) work is put in, that it would at least be generously compensated financially or something like that	Hanna	paybacks
On the one hand, I probably wouldn't start a business on my own, so if I did, then only with a partner	Igor	collaboration setting
Yes, as I said, I've learned so much and taken so much with me and of course I'm now applying all of it	Roland	mixed-high	Applicability of learning
No, what I learned from the last company is actually difficult to use	Alina	mixed-low
So you don't take so many risks anymore, you're more risk-conscious	Edgar	dealing with risk	Applicable learning contents in employee settings
Well, now I'm a bit older - so what I'm already noticing is that I'm slowly gaining the confidence to lead	Barbara	bearing of leadership responsibilities

Source(s): Authors' own work