The practice of talent management: a framework and typology

In the midst of intensified global competition, weakening bonds between employers and employees, and the ongoing shift toward knowledge-based economies, talent management (TM) has become a key business activity and a critically important decision area (Sparrow and Makram, 2015; Collings and Mellahi, 2009). In essence, the dilemma is that talent has become an attractive source of competitive advantage and at the same time, many organizations are encountering talent shortages and finding it increasingly difficult to attract, manage, and retain talented people (Schuler et al., 2011).

The strong interest in talent issues amongst practitioners has been followed by growing scholarly interest and an increasing number of academic publications on TM (Thunnissen et al., 2013a). However, scholars consistently point to two main shortcomings in the literature. The first is the conceptual ambiguity of “TM.” While substantial advances have lately been made toward establishing mature theoretical frameworks for the concept of “talent” (Dries, 2013; Gallardo-Gallardo et al., 2013; Meyers et al., 2013; Thunnissen and Van Arensbergen, 2015; Nijs et al., 2014; Tansley, 2011), the framework for TM is considerably less developed (Al Ariss et al., 2014; Collings and Mellahi, 2009; Thunnissen et al., 2013b). Specifically, Dries (2013) states that many scholars fail to define “which specific practices fall under the TM label” (p. 274). In many ways, it is still unclear what TM actually is.

The second shortcoming is the paucity of empirical research (Collings and Mellahi, 2009; Dries, 2013; Gallardo-Gallardo et al., 2013; Meyers et al., 2013). Reviews point to an “urgent need” for empirical studies to move the field beyond “assumptions and hypotheses” (Thunnissen et al., 2013b, p. 334). In particular, there is a lack of rigorous research that pays close attention to actual organizational practices and develops an in-depth understanding of how organizations define talent and carry out TM activities (Meyers and van Woerkom, 2014; Meyers et al., 2013). An important avenue for such research is the comparative investigation of “typologies that are dominant in organizational reality” (Thunnissen et al., 2013b, p. 334). Such typologies inform us of the various ways in which TM is manifested in contemporary organizations and constitute an important foundation for further research on the antecedents and outcomes of different TM approaches.

This paper answers to the call for robust empirical research on TM and poses the following research question:

In order to answer our question, we develop a comprehensive framework for comparative analysis of TM in organizations. The aim is to develop a framework that accounts for systematic variation in the empirical phenomenon of TM. Next, we present an in-depth study of the TM practices of 30 organizations. The main findings are a typology consisting of four distinct TM types as well as a description of the same.

The paper aims to advance TM research in two ways. Addressing the first shortcoming in the literature, it aims to contribute to a deeper understanding of the conceptual boundaries of TM by integrating theoretical insights in extant literature into a holistic framework that delineates the main dimensions of a talent definition and the main practices of TM. Drawing on this base to address the second shortcoming, it aims to contribute empirically by investigating the different ways in which TM is carried out in organizational settings. Together, the framework and the typology constitute an important step for further theory development in TM (Doty and Glick, 1994; McKelvey, 1975; Meyer et al., 1993). Specifically, they provide a solid basis for investigating the conditions under which organizations adopt different types of TM and can therefore contribute to the development of a greater contextual understanding of TM (Thunnissen et al., 2013b; Gallardo-Gallardo et al., 2015).

A recent, comprehensive review concludes that three main areas are addressed in the academic literature on TM: the definition of talent, outcomes and effects of TM, and TM practices and activities (Thunnissen et al., 2013a). Outcomes are crucial but are beyond the scope of this paper. Here, we focus on the definition of talent, TM practices, and the implications talent definitions may have for TM practices.

Although the literature has long made a rough distinction between inclusive and exclusive approaches to TM (e.g. Lewis and Heckman, 2006; Collings and Mellahi, 2009; Stahl et al., 2012), scholars have lately begun to explore in more depth the relationship between talent definitions and TM practices. Little work has, however, been done to draw up a systematic TM typology. The configurations of talent definitions and TM practices that are drawn up in the research works of Gallardo-Gallardo et al. (2013), Meyers et al. (2013) and Meyers and van Woerkom (2014) are an important first step toward conceptually defining TM types. A common shortcoming of these contributions is, however, that the proposed types are unsystematically and cursorily described. Previous typologies also put disproportionate emphasis on one aspect of the talent definition, such as the innate-acquired dimension, thereby possibly concealing the diversity of different views on talent in practice (Thunnissen and Van Arensbergen, 2015). Moreover, there is a paucity of empirical research that compares configurations of talent definitions and TM practices and identifies distinct types that exist in organizations. The studies conducted to date (e.g. Iles et al., 2010a; Valverde et al., 2013; Cooke et al., 2014; Thunnissen and Van Arensbergen, 2015) constitute valuable empirical contributions, but the field suffers from the lack of a comprehensive, systematic, and detailed framework, which makes it difficult to explore similarities and differences between organizations. In the following, we develop such a framework for comparative analysis based on a synthesis of the literature.

Our framework for identifying TM types begins with the talent definition. A number of dimensions of talent have previously been discussed in the literature. In the current framework we include five well-established dimensions derived from theoretical perspectives on talent that exist in six streams of literature: HRM, I/O psychology, educational psychology, vocational psychology, positive psychology, and social psychology (Gallardo-Gallardo et al., 2013; Meyers et al., 2013; Dries, 2013). Each dimension should be seen as a continuum rather than mutually exclusive poles. Nonetheless, the position an organization takes in each dimension has implications for their TM practices (Dries, 2013; Thunnissen and Van Arensbergen, 2015). The first dimension is whether talent is seen as subject or object. Subject views define talent as individuals, whereas object views define it as characteristics of individuals, such as abilities and knowledge. The second is whether talent is inclusive and thus refers to all employees of an organization, or exclusive and refers to an elite subset of employees. The third dimension is innate-acquired: is talent a natural ability or something that can be taught and learned? The fourth is whether talent is based on input or output. Input-based views define talent as the employee’s motivation, interest, ambition, values, and career orientation, whereas output-based views define it as the employee’s ability, manifested in performance and achievements. Lastly, talent may be seen as a quality that remains unchanged when transferred between contexts, or as something that emerges and is recognized only in specific contexts.

The literature on TM discusses a number of different TM practices. Integrating practices that often have been discussed separately in the literature, the proposed framework includes five key practices: recruitment, talent identification, talent development, career management and succession planning, and retention management. Main dimensions along which each practice varies are discussed below.

In the TM context, recruitment is defined as the activities to identify and hire talented external job applicants (Meyers et al., 2013). The first dimension is how high the reliance on recruitment is. When high, organizations “buy” talented individuals from the market for positions at all levels; when low, organizations rely on a “make” strategy and recruit primarily at entry level (Cappelli, 2008; Collings and Mellahi, 2009; Stahl et al., 2012). Second, organizations may recruit to fast-track programs tailored to talents, or to regular entry-level positions (Dries and Pepermans, 2008; Stahl et al., 2007; Dries and De Gieter, 2014). Third, organizations may employ formal selection procedures or make selection decisions primarily based on personal judgment (Silzer and Church, 2009; Mäkelä et al., 2010; Meyers et al., 2013; Nijs et al., 2014). When formal selection procedures are employed, there are clearly stated talent definitions and assessment policies. In addition, validated assessment tools with high methodological standards are used, and assessors strive for objectivity. When decisions are based on personal judgment, talent assessment is holistic and based largely on assessors’ personal understanding.

The second practice is talent identification, i.e. activities to find talent amongst current employees (Meyers et al., 2013). The first dimension is whether regular talent reviews are carried out to create talent pools (Boudreau and Ramstad, 2005; Stahl et al., 2007; Mäkelä et al., 2010). Reviews may aim either to identify those that rank at the top (Becker et al., 2009) or to identify each employee’s strengths (Downs and Swailes, 2013). When talent reviews are not carried out, organizations may rely on succession planning or informal identification processes (Lewis and Heckman, 2006). The second dimension concerns the degree of reliance on formal procedures (see recruitment). Third, identification may be based on input or output (Meyers et al., 2013; Mäkelä et al., 2010; Silzer and Church, 2009). Output-based assessment is concerned with what employees already have accomplished, while input-based assessment is concerned with what the individual brings into the next job, e.g. motivation and cultural fit.

The third practice is talent development, i.e. activities to nurture talent. First, activities may be mainly program-based and take place through formal learning models, or they may be experience-based and take place through job activities such as challenging assignments and rotations (Yost and Chang, 2009). Second, development activities may be exclusive or inclusive (Stahl et al., 2007; Collings and Mellahi, 2009; Yost and Chang, 2009). Exclusive programs, where admission is through nomination, are high-profile programs tailored to an elite group. Inclusive programs provide learning opportunities to the whole workforce. The third dimension concerns whether talent pool activities are organized (Boudreau and Ramstad, 2005; Stahl et al., 2007; Collings and Mellahi, 2009). Some organizations see their talents as a group that should develop together, and consequently organize activities to that purpose, e.g. network events and joint projects.

The fourth TM practice is career management and succession planning, i.e. activities aimed at ensuring that organizations optimally use their talented employees by facilitating their flow into the right jobs. The first dimension is career management focus (Dries and Pepermans, 2008; De Vos and Dries, 2013). When organizations focus on traditional organizational careers, they prioritize the organization’s needs and implement formalized career practices focused on upward mobility. Careers tend to be bounded and paternalistic. When organizations focus on self-managed careers, they foreground the needs and responsibilities of the employees. Careers tend to be boundary-less and self-directed. The second dimension concerns how clear career paths are (Dries and Pepermans, 2008; Meyers et al., 2013; Thunnissen et al., 2013a). When clearly defined, alternative career paths are drawn up by the organization. When loosely defined, career paths may be drawn up, but it is emphasized that careers may be advanced in different ways. When ad hoc, they are formed depending on the supply of talents and jobs. The third dimension is whether mainly vertical or horizontal promotion is focused, i.e. whether upward mobility or skills development within the same role or at another job on the same level is emphasized (Dries and Pepermans, 2008; Meyers et al., 2013). The fourth dimension has to do with how employees’ flow into the right jobs is facilitated (Boudreau and Ramstad, 2005; Stahl et al., 2007; Mäkelä et al., 2010). Organizations may regularly review their talent pools to match talents with current job openings, or they may utilize succession planning or an open labor market.

The final TM practice is retention management, i.e. activities to prevent employee turnover. The first dimension concerns the degree of emphasis on creating a strong employer brand and refers to how strong the organization’s efforts are to differentiate itself from competitors by formulating an employee value proposition (EVP) that appeals to talents (Stahl et al., 2012; Meyers et al., 2013; Thunnissen et al., 2013b). The second dimension is the degree to which organizations take specific measures to enhance loyalty among talents (Stahl et al., 2007).

We conducted an in-depth, comparative study of the TM practices of 30 organizations based in Sweden. In this section, we describe the data collection and the analytical process that led to the development of our main findings: a typology consisting of four distinct TM types and descriptions of the types. Our overall research design was abductive (Alvesson and Sköldberg, 2009; Mantere, 2008). Abductive approaches involve an iterative process of movement back and forth between the studied phenomenon, the researcher’s pre-understanding, existing literature, and the empirical data. The aim is to reach what is called “the best explanation,” i.e. an understanding of the phenomenon that best takes these different components into account.

We started out with a general research question on what organizations “see” and “do” as TM. Turning to the academic literature on TM, we noted a lack of clarity regarding what TM actually consists of. However, five practices (recruitment, talent identification, talent development, career management and succession planning, and retention management) were recurrently mentioned in different places throughout the texts, and we decided to focus our data collection on these.

The first step in selecting organizations was specifying the population (Eisenhardt, 1989). The population included organizations that were explicitly engaged in TM and had implemented numerous activities specifically aimed at managing talent. In order to encourage variation, the population included organizations in different industries. Because formal TM tends not to be implemented in small organizations with limited resources (Schuler et al., 2011; Festing et al., 2013), the population included organizations that were large, although not necessarily in absolute numbers but relative to other organizations in their industry. In the second step, specific organizations were chosen from this population. This selection was not random, but based on the principle that the chosen organizations should be likely to render what was being studied “transparently observable” (Eisenhardt, 1989, p. 537). Specifically, we chose organizations that were known to be most mature in their TM work and that we thought would “see” and “do” TM differently. In this manner, we selected 30 organizations to include in the study (see Table I).

Data were collected through in-depth interviews. We considered interviews to be appropriate since we aimed not only to record which practices the organizations engaged in, but also to gain a detailed understanding of how these practices were shaped and enacted (Edmondson and McManus, 2007; Brinkmann and Kvale, 2015; Monks et al., 2013). The interviews were semi-structured. We followed an interview guide that covered the five practices identified in our initial reading of the literature. On a practical level, this means that we asked participants to define talent and describe what specific activities they carried out as part of TM. All five practices were discussed in each interview, but more or less extensively depending on how much emphasis was put on the practice in the particular organization. Through follow-up questions, we encouraged the interviewees to expand on their descriptions and probed into the details of the practices. Our questions therefore passed through a funnel-like structure. For example, we began by asking the general question “what activities do you include in TM,” continued with slightly more specific questions on the practices such as “how do you identify talent,” which were then followed up with specific questions such as “are there explicit criteria that guide your decisions,” “do you use any tools,” or “how do you assess the information that is collected.”

The study comprised interviews with 56 organizational representatives. The persons within the HR function responsible for TM were interviewed. In some cases, they held the title of vice president HR, but more often the title was the head of TM or similar. They were chosen because they were responsible for the design, implementation, and follow-up of TM in their organizations. We argue that their accounts can be seen as representative of their organizations by virtue of them being key decision-makers regarding their organizations’ TM policies and practices. When possible, which was in approximately one third of the organizations, interviews were also held with other senior managers, such as heads of business areas or functions, who were operationally involved in TM practices.

The interviews, which lasted between 1 and 1.5 hours, were carried out by two of the authors. They were audio-recorded and fully transcribed. The analysis was based on the original transcripts, but excerpts have since been translated from Swedish to English for the purpose of dissemination.

Already in our first reading of the transcripts, it became clear that there were large variations in how the organizations described their view on talent, what specific activities they included in TM and how much emphasis they put on each practice. However, there seemed to be different patterns in this variation that could be indicative of the occurrence of different types of TM in the empirical data. Searching for ways to make sense of this, we turned again to the literature and looked in detail at publications that conceptually define TM types (e.g. Gallardo-Gallardo et al., 2013; Meyers et al., 2013). Finding that the proposed types were sketchily described and that they thus could not be used to understand the patterns we sensed in the data, we began to put together a framework that could be used to identify TM types by searching through academic publications on TM for discussions on the practices. We found that each practice could be described in terms of a number of dimensions along which the practice varied. Putting these together, we created the framework that is presented in this paper’s literature review.

In the first step of the analysis, we conducted qualitative content analysis to analyze the interview transcripts (Wilkinson, 2004; Saldaña, 2012). The transcripts were analyzed one by one in order to investigate how TM is carried out in each organization. We employed structural coding, which “applies a content-based or conceptual phrase representing a topic of inquiry to a segment of data that relates to a specific research question used to frame the interview” (Saldaña, 2012, p. 84). Using the TM framework as a scheme, we coded sections of the transcripts where the different talent dimensions, practices, and practice dimensions are mentioned. These interview excerpts were placed in data tables, one for each organization. Together, the 30 tables comprised a systematic representation of the studied organizations’ configurations of talent definition and TM practices. When there was more than one interviewee in an organization, the transcripts were coded into the same table. Inconsistencies between interviewees were few, but when they occurred we contacted the interviewees to make sure we had a correct understanding of the practice in question.

In the second step, we looked at the coding with a more holistic view and explored patterns across organizations to see if we could identify TM types. We began by closely studying the configurations generated in the first step, looking for commonalities and differences. Once we began to see the contours of types in the material, we moved iteratively between the data tables and the preliminary types, gradually clarifying the common characteristics of each type and the differences between types. In this manner, we derived from the data a typology, where each of the four types constitutes a distinct pattern in how a subset of the studied organizations defined talent and carried out TM practices. This analysis also revealed that each type has a principal practice, i.e. one specific practice was pointed at by interviewees in each type as especially significant in characterizing their TM work. We then went back to the full transcripts to ensure that the types we had identified were faithful descriptions and that the typology included all of the organizations. Finally, we created a summarizing data table with short but illustrative interview excerpts for each type (see Table AI) and constructed descriptions of each type.

The credibility of the results (Silverman, 2011) was enhanced through a procedure involving independent work followed by discussion to resolve differences. In the first step of the analysis, all three authors initially coded the same data separately. We then compared our independent coding, discussed the instances where it diverged, and clarified definitions of the coding scheme. In the second step, each author independently studied all 30 configurations to draw up preliminary types which we then discussed. Similarly, each author independently categorized each organization as a specific type. At this stage, there was a high level of agreement and the final typology emerged.

The findings offer detailed insights into how TM is carried out in practice. Four distinct types of TM are described below. Table AI (Appendix) contains interview excerpts illustrating each type (the number after each quote refers to the interviewees’ organization as listed in Table I). The typology is summarized in Table II.

Organizations with humanistic TM hold an inclusive view in which literally all employees are seen as talents. Top performers have one type of talent, but other employees have other types of talent, and TM encompasses all. There is a firm belief that talent can be developed throughout a person’s lifetime and that the organization can become an important arena for that development by offering opportunities for learning. Ability is a part of talent, but more important are the individual’s interests and desires. These organizations pay more attention to how the employee wants to develop and what she/he would enjoy working with than how well she has done in the past. Talent is seen as context-dependent, to the extent that it is argued that a person who is underperforming in one part of the organization may well be recognized as a talent in another.

There is little reliance on recruitment of talents from outside. Buying external “hotshots” (16) would send the message that current employees are not good enough. Similarly, they do not offer fast-track programs, as that would mean singling some employees out as better than the rest and giving them “an easy ride” (15). Selection relies on clearly defined criteria and formal tools are used to assess these criteria. It is emphasized that this guarantees fair, equal treatment of candidates.

In order to identify talent, talent reviews are performed, most often annually. The purpose is to bring to light employees’ strengths and find the right place for them in the organization. These reviews encompass employees that have indicated that they want to develop beyond their current role. The performance of each of these employees is discussed, but the focus lies on a wide range of input-related factors such as personal interests. Plans are made both for those that come out well in the discussions and those that do not. Assessments are informal and based on a holistic understanding of talent rather than explicitly defined criteria. The stated reason for this is that they want to focus their attention on the actual people involved rather than the technicalities and “mathematics” (11) of assessment. It is accepted, even embraced, that assessment is a subjective process. Formal assessment tools may be used, but the results are not taken at face value since they are seen as simplistic.

Talent development is the principal practice of the humanistic type of TM. There is a strong belief that every employee will be valuable if developed in the appropriate way. Development opportunities are thus offered to all, not just a chosen few. Equal emphasis is placed on experience-based and program-based development. The former takes place mainly through job rotations, where people move between different jobs and departments. Program-based development takes place through courses and trainings, often run by internal academies. Admission is through application rather than nomination. Some programs address specific job-related skills and develop people within their current role, while others address general leadership skills and develop people for vertical advancement. Talent pool activities are not organized, since there is a wish not to single out specific groups as more talented than others.

Since the organizations want to encourage their employees to develop according to their interests, career paths are only loosely defined. Consequentially, careers can be horizontal as well as vertical, and both are equally valued. The main idea is to emphasize that it is not only becoming a manager that makes a career. Despite the lack of formalized career practices, responsibility for career does not rest only on the individual. Once an employee has communicated how she wants to develop, managers are expected to support them. Talent pool reviews for matching purposes are not undertaken. Instead, there is succession planning for a few top positions. Those identified as successors are not guaranteed promotion when such positions become available, but they are encouraged to apply.

These organizations differentiate themselves from their competitors by emphasizing that they “take care” of their employees. In particular, the EVP highlights development opportunities for all employees. Instead of offering the most competitive salaries they profile themselves as employee-friendly organizations that support work-life balance through reasonable working hours and flexible work practices. They do not employ specific activities to breed loyalty among talents, but rather emphasize the strong culture of inclusiveness.

Organizations with competitive TM hold an exclusive view on talent. In any group of employees, there will always be a small number of “stars” (23) whose excellent performance and high potential set them apart. Each employee has an inborn capacity to reach a certain organizational level. Talent processes are therefore aimed at finding and putting all efforts on the few that can make it to the top. These organizations see talent as being about ability rather than input-related factors such as motivation and interest, and the focus on past performance is very strong. Since they believe that a talent will be a talent no matter where she works, there is a strong sense of competing against other employers for talents.

Consequentially, these organizations focus more on buying than making talent. They hire to all levels and are continuously on the lookout for the best in order to sharpen the organization’s competitive edge. In particular, they put a great deal of effort into hiring the best young talent. High-status fast-track programs such as graduate and trainee programs are seen as a key to attracting this group, who are talked about as future leaders of their organizations and expected to advance quickly in the ranks. Selection to these and other positions relies heavily on formal assessment tools, including a wide range of tests and assessment center methodology, to ensure that the objectively best candidate is selected.

Talent identification is the principal practice of the competitive type. Substantial time and effort is put into “the enormous machinery” (12) of organization-wide reviews aiming to identify the talented few. All employees are placed in a grid with performance and potential on the axes. While performance relates to employees’ current role and potential is defined as readiness for promotion and likeliness to succeed, both are discussed on the basis of the output that the employee has already achieved. Forced ranking is rare, but there is usually an expected distribution. Those that are identified as talents are placed in different talent pools and given individual development plans. The identification procedure is described as rigorous and laborious. The main means to attain objective assessment is using a large number of formal assessment tools. Calibration is also considered a key to ensuring that employees are assessed in an objective manner.

Talent development is mainly program-based. Talents are nominated to exclusive programs designed to have a “turbo effect” (21) on their careers, primarily by providing a strategic outlook on the organization and developing leadership skills. The symbolic value is significant; programs are seen as a means to make talents feel recognized and appreciated. Similarly, a number of different talent pool activities, such as shadow management teams, are organized to stimulate talents, make them visible to management, and provide opportunities for talents from different parts of the organization to network.

In organizations with competitive TM, there are clearly defined career paths for leaders, specialists, and project leaders. Talents follow the leadership track and are expected to become managers at successively higher levels. Careers are traditional, particularly in the sense that the organization takes the main responsibility for providing talents with the right career opportunities. Talent pool reviews are organized, often monthly, to discuss talents ready for promotion and match them with current job openings. This is seen as a key to ensuring that TM supports the organization’s strategic goals.

Since the view on talent is that it is transferable, these organizations assume that they are looking for the same talents as their competitors. In this war for talent, maintaining a strong employer brand is seen as a major concern. The EVP emphasizes clear career structures, exclusive talent programs and personalized development. Especially the latter two are also seen as essential for building loyalty among talents. Talent programs give talents the feeling that they are valued, while personalized development plans assure that they will be able to advance quickly. Talents will turn elsewhere if they don’t feel that the organization has plans for them.

Organizations with elitist TM see talents as the crème de la crème. The exceedingly selective recruitment process ensures that all employees are the best, but talents outperform even these peers. Talents are described as “aces” (8) that advance more quickly in the career ladder. The talent definition is clearly connected to the partner model these organizations often have; talents are employees that have the potential to become partners. Talent is seen as an expression of qualities that are inborn; however, it is understood that talent must be formed by the organization to become “fully fledged” (7). These organizations see talent as being largely about ability, defined in terms of delivering in the professional role. However, input-related factors are also important in defining talent, not least motivation and values. Talent is seen as transferable, and these organizations struggle to keep their talents onboard but are also proud to see that their talents are highly attractive on the external labor market.

The principal practice in the elitist TM type is recruitment. A large number of people, including partners, put a great deal of time and effort into attracting and selecting the best of the best. These organizations recruit almost exclusively at entry level, and there is a strong reliance on making or, rather, refining talents – transforming them “from pupae into butterflies” (7). Only students with the highest grades from the best universities are considered. Formal assessment tools are often used but interviews are considered most important. Partners’ personal judgment is crucial for assessing whether candidates are well-suited in terms of motivation level and cultural fit. These organizations do not have fast tracks; instead, all new recruits enter to the same position, and all must advance in the same career structure without skipping any steps.

Regular talent reviews are carried out in which employees are assessed not by their managers but by people “a few notches above on the totem pole” (19). The aim is to explicitly compare and rank employees in order to identify the very best. Performance is measured in terms of clearly defined goals, whereas potential is defined as the capacity to grow the business. Excellent performance is required; however, it is “not enough” (10). Since all employees are expected to be high performers, excellent performance is considered a threshold. To be identified as a talent, a person also needs to conduct herself in a certain way (9). The review process is structured and objective in the sense that it is based on a complex system of transparent and quantifiable criteria. However, these organizations emphasize that final talent decisions must be made based on the intuitive personal judgment of the assessors.

Talent development is mainly experience-based. Talents may be assigned partners that watch over their careers, or they may be expected to find partners who will take them under their wing. Either way, having a strong relationship to a senior person is a key to being given development opportunities. Senior people can provide opportunities that employees cannot apply for or access on their own, such as “tagging along” (10) to meetings and trips. Although there are some exclusive programs, these are not considered important for talent development, nor are formal talent pool activities organized.

There is only one career path in the organization. It usually includes four to five steps, each with a specific title. Employees are expected to take each step in the ladder within a specified time. Careers are paternalistic in the sense that they “lie in the hands of the partner” (9), who has the power to make or break a career. Reviews of the talent pool for placement purposes are not seen as necessary, since responsibility is placed on senior people to make sure that “their” talents are optimally used.

Organizations with elitist TM emphasize maintaining a strong employer brand, especially among students. They participate in career fairs and organize activities targeted at establishing relationships with the most talented students. The EVP includes clear career structures and competitive compensation, but most important is the opportunity to work with the best and with “the most exciting business on the market” (9). These organizations employ several different strategies to prevent talent turnover, most importantly the possibility of advancement to partnership and annual performance-based bonuses.

Organizations with entrepreneurial TM hold a rather inclusive view on talent. While not claiming that all employees are talents, they do say that all employees have the potential to become talents if they find “the right environment” (14). Talent is entirely about input in terms of the individual’s ambition to seek out challenges and take on new responsibilities. These organizations are very result-oriented but they see performance as an outcome of an attitude of “appetite and fire” (29), and it is this attitude rather than performance itself that defines talent. As people take on new challenges, gain experience, and learn from that experience, they become increasingly talented. In these organizations, there is a very strong emphasis on cultural fit, which means that a person that is a talent in one organization may not be seen as a talent in another organization.

As a result of the heavy emphasis on culture, these organizations see successful talent provision as the result of home-growing. External recruitment is kept at a minimum, and managers that grow talents that go on to new internal assignments are acknowledged and rewarded. The selection procedure looks not for the right formal competencies but for the right attitude. Skills can be acquired through training, but not drive and ambition. Formal assessment tools cannot capture whether a candidate has the right attitude or not; instead, it is something assessors develop an eye for over time. A great deal of emphasis is placed on interviews and discussions with candidates. These organizations do not offer fast tracks, largely because they believe that employees need to gain experience and prove themselves to become talents.

In organizations with entrepreneurial TM, talent identification is based on self-identification rather than talent reviews. This is because conducting talent reviews to pick out some people would contradict the culture they wish to foster, which emphasizes individual drive and ambition. When talents step forward and ask for challenges, they are told to apply for their next assignment. Talent identification practices are thus kept simple and informal, and based on personal judgment rather than “academic” (14) assessment systems.

Talent is developed from within by offering all employees a wide range of missions and challenges, e.g. project leader assignments, overseas assignments, and P&L responsibility. There is a little program-based development. These organizations argue that programs, especially exclusive ones to which the participants are handpicked, foster a sense of entitlement and discourage talents from taking personal responsibility for their careers. For the same reasons, these organizations do not organize specific talent pool activities.

Although there are few formalized practices for career management and succession planning, this is seen as the principal practice in the entrepreneurial TM type. Core to this practice is instead the open internal labor market. One key aspect is that there are plenty of challenging openings, not just a few which are the gateways to advancement. Another is that anyone can apply for these openings. The idea of the self-managed career is very strong; again, the emphasis lies on individual responsibility. Career paths are so loosely defined that they may be considered ad hoc. They are formed depending on what challenging assignments are available at a certain point in time and what the talents want.

These organizations differentiate themselves by emphasizing that they give all employees opportunities to prove themselves. The lack of predictability, formalized career practices and talent programs is not seen as a disadvantage; instead, the open labor market is highlighted. An important part of the EVP is the idea of challenge: in these organizations talents will be challenged and are expected to challenge others. Giving talents own areas of responsibility is seen as the most effective method for breeding loyalty. They say that talents crave autonomy and mastery, and put a great deal of effort into organizing work accordingly.

The research presented here contributes to the existing knowledge on TM in several ways. First, it contributes to a deeper understanding of the conceptual boundaries of TM. While the notion of talent has become increasingly well conceptualized in the literature, it is still unclear what TM actually involves (Collings and Mellahi, 2009; Gallardo-Gallardo et al., 2013; Meyers et al., 2013; Thunnissen et al., 2013a). The framework presented in this paper addresses this need for precision by systematically laying out which specific practices fall under the TM label. Importantly, it also specifies the dimensions along which these main practices may vary. This framework has several strengths. First, it integrates theoretical insights dispersed in various research streams. Second, it may be used as a conceptual basis for empirical studies identifying TM types across organizations. Third, the framework is sufficiently detailed to go beyond observing whether a practice is found in an organization and instead probe into how each practice is shaped. As Meyers and van Woerkom (2014) argue, this is an important avenue for gaining a deeper understanding of TM, since any particular practice may be carried out in different ways. In sum, the framework represents a valuable theoretical contribution that may conspire to make research on TM less fragmented, and knowledge more cumulative.

Second, the study is to our knowledge among the first to systematically map in detail how TM is carried out in organizations. The typology constitutes an important empirical contribution as it gives an in-depth understanding of four types of TM in contemporary organizations. This kind of investigation is necessary for the field to move beyond assumptions and hypothesis about what TM might be, as well as prescriptive notions of what it should be (Collings and Mellahi, 2009; Dries, 2013; Iles et al., 2010b; Thunnissen et al., 2013a), toward a well-grounded understanding of how TM is manifested in organizations.

Third, the research deepens our understanding of how talent definitions carry implications for TM practices. While the literature has long made a rough distinction between inclusive and exclusive approaches to TM (Lewis and Heckman, 2006; Stahl et al., 2012; Cappelli and Keller, 2014), the findings presented in this paper nuance our understanding of what it means to take an exclusive or inclusive approach. It shows that there is not one exclusive and one inclusive approach, but rather that the competitive and elitist TM types represent two distinct exclusive approaches and that the humanistic and entrepreneurial types represent two distinct inclusive approaches. This is an important contribution in particular with regards to inclusive approaches, since they are increasingly recognized as legitimate TM approaches but have been paid rather scant attention to date (Meyers and van Woerkom, 2014; Swailes et al., 2014).

The differentiation of four different TM types further points at a second talent definition dimension with far-reaching consequences for how TM practices are carried out. This concerns whether talent is defined primarily in terms of input or output. With regards to talent definition, what differentiates the competitive from the elitist type, and the humanistic from the entrepreneurial type, is the degree to which they see talent as being about the employee’s motivation and interest or the employee’s past performance. This difference seems to be related to differences in how talent recruitment and identification are carried out. The findings suggest that the more the organization focuses on output, the more likely it is to use formalized methods of assessment for recruitment and identification of talents.

Interestingly, despite recent arguments in the literature for moving from an individualistic view on talent towards an emphasis on specific skills and competences (Boudreau and Ramstad, 2005; Collings and Mellahi, 2009; Jones et al., 2012), none of the organizations in our study adopted a view on talent as object. Organizations in the current study use talent as a synonym for people and focus TM practices on talented people (subject) rather than their skills and knowledge (object). By extension, this means that the studied organizations do not grasp the possibility of identifying the pivotal positions where specific sets of skills and knowledge may have significant impact on organizational performance. Thus the strategic outlook on TM, in which the identification of strategic jobs is a central tenet (Huselid and Becker, 2011; Huselid et al., 2005), does not yet seem to have had a strong impact on practice.

Fourth, although our study was not designed to investigate the conditions under which an organization is likely to adopt a certain TM type, the findings invite a discussion regarding if and how contextual factors such as industry, size, and culture play a part in this decision. This is particularly pertinent, since scholars have increasingly emphasized the need to take a contextualized approach to TM (Thunnissen et al., 2013b; Gallardo-Gallardo et al., 2015; Collings and Mellahi, 2009; Stahl et al., 2012). Starting with industry and size, we find conflicting indications. On the one hand, all four public management organizations have adopted humanistic TM. Also, elitist TM is found in all of the commercial and professional services organizations. Competitive TM appears to be more common in larger organizations. On the other hand, among the four banks and investment banks we find humanistic, competitive as well as elitist TM. There are larger and smaller organizations adopting each type of TM. Based on our detailed reading of the empirical data, we would conclude that what seems to influence the decision to adopt a certain type of TM the most is not industry or size, but organizational culture.

Regardless of what type of TM they adopted, the organizations spoke lengthily about the importance of fitting their TM with their organizational culture and its underlying values. For example, organizations with humanistic TM frequently corroborated their view on talent by stating that an exclusive, innate view on talent and TM practices that only encompass a few would be contradictory to their culture of promoting egalitarianism and teamwork (Iles et al., 2010a). Similarly, organizations with competitive TM argued that an inclusive, input-based talent definition and TM practices that addressed all employees would run counter to their “performance-driven culture” (Stahl et al., 2007). Those with elitist TM argued that their view on talent as innate but formable fit exceptionally well with their culture of emphasizing both high performance and self-conduct (Rivera, 2012). Finally, organizations with entrepreneurial TM associated their view on talent and the practice of offering all employees a wide range of assignments with a culture that puts drive, passion, and energy first.

Although no definitive conclusions can be drawn about the antecedents and outcomes of different TM approaches on the basis of the study presented here, we nevertheless argue that the findings constitute an important step toward theory development in TM. The value of typologies lies in that they pin down the distinct characteristics of complex organizational phenomena and are expected to exhaustively describe the holistic patterns in which these characteristics occur (Doty and Glick, 1994; McKelvey, 1975; Meyer et al., 1993). Rich descriptions of these patterns may then be used to develop and test theoretical propositions about the exogenous and endogenous forces that generate specific configurations and about the outcomes of different configurations. This leads to inquiry focusing on “the patterning of organizational elements” rather than separate analyses of the relationships between “narrowly drawn” sets of variables (Meyer et al., 1993, p. 1,181). Specifically, the typology presented here may be used to develop theory on how contextual factors affect an organization’s TM. It is necessary to be specific about the type of TM that is being adopted in order to be able to investigate the influence of factors such as industry, size and organizational culture. Similarly, studies of the outcomes of TM, such as those Collings and Mellahi (2009) have called for, must be specific about what type of TM it is that the observed outcomes are outcomes of.

Fifth and finally, this paper can provide important insights to the practitioners who are considering implementing TM or planning to change the focus of their organizations’ TM. Outside the academic literature, TM is often described in terms of “best practice”, which may lead practitioners to implement a type of TM that is unsuitable for their organization. We believe that the findings, in particular the descriptions of the TM types, point to important decisions to be made when designing TM and provide guidance in this decision-making.

Even as the concept of TM has become ubiquitous in contemporary business settings, it has remained ambiguous. The purpose of the present paper was to develop a deeper understanding of the conceptual and empirical boundaries of TM so that scholars and practitioners may enhance their knowledge of what TM actually is and how it is carried out. To this end, we developed a framework for comparative analysis and applied it on 30 organizations to empirically derive a typology consisting four distinct TM types.

The study is, however, subject to a number of limitations that provide opportunities for future research. First, as the typology is based on insights from a rather limited number of organizations, which although mainly multinational are grounded in a specific national context, it should be validated on a larger sample of organizations. Future studies might explore questions such as whether the identified types are found in other contexts, how common they are, and whether there are other yet unidentified types. Additionally, future research might investigate if and how contextual factors such as size, industry, and culture affect which TM type an organization adopts. Second, while the limited sample size of the current study enabled us to generate an empirically rich understanding of the identified TM types, it limited our ability to draw systematic conclusions about the realized outcomes of different types of TM. Future research, based on e.g. survey methodology and archival data, might explore how employees perceive different TM types, how they affect individual outcomes such as employee satisfaction and engagement, as well as how different TM types affect organizational outcomes such as retention rate and organizational performance.

This paper is based on research funded by The Swedish Foundation for Humanities and Social Sciences under Grant No. P12-0986:1.

Table AI