This study focuses on the development of a talent management framework for the private sector. There seems to be a lack of a uniform theoretical talent management framework in the private sector, which can be used as a guideline to drive talent management.

Collings and Mellahi (2009) asserted that since a group of McKinsey consultants coined the phrase ‘war for talent’ in 1997, the topic of talent management has received a remarkable degree of practitioner and academic interest. This relatively recent emphasis on talent management represents a paradigm shift from more traditional human resource to strategic talent management, which according to Silzer and Dowell (2009) is driven by corporate strategy, incorporated with other processes, managed as a core business practice and deep-seated as a talent mindset. The Chartered Institute of Personnel and Development (2006) further views excellent and best practice level of talent management as a strategy designed to deliver corporate and human resource management strategies. Often there is a formal talent management initiative linked to the human resources management function and flowing vertically from the corporate strategy-making process.

According to Powell and Lubitsh (2007), talent management has moved rapidly up the corporate agenda in recent years, and this is evident in the amount of research papers published over the last decade (e.g. Ashton & Morton, 2005; Bersin, 2006; Chikumbi, 2011; Church, Rotolo, Ginther & Levine, 2015; Collings, 2014; Festing & Shafer, 2014; Nijs, Gallardo-Gallardo, Dries & Sels, 2014; Meyer, 2005; Odierno, 2015; Prinsloo, 2012). Egerova (2014) is of the opinion that the increasing attention to talent is affected by factors including globalisation, knowledge-based competition, changing the world of work as well as new forms of organisations and demographic changes.

Despite the growing popularity of talent management and over a decade of debate and hype, the concept of talent management remains unclear (Collings & Mellahi, 2009; Festing & Schafer, 2014; Nijs et al. 2014). Furthermore, the current state of talent management literature is exacerbated by the fact that in addition to ambiguities around the definition of the concept, there is also alarmingly limited theoretical development in the area as highlighted in the research conducted by Arris, Cascio and Paauwe (2013). Consequently, even though business and consulting firms in the private sector have been driving talent management as a strategic priority, there seems to be an overall lack of theoretical frameworks in the academic field.

Overall, there also seems to be a lack of linkage between talent management practices and the broader human resource systems (Ntonga, 2007), as well as an absence of clear succession planning, skills audits and organisational culture driving talent management (Chikumbi, 2011).

This study attempted to bring to light critical factors which must be taken into account to ensure that organisations can achieve successful implementation of an integrated, holistic and comprehensive talent management framework. The framework will strive to guide talent management practices within companies in the private sector, and also improve and advance the maturity of talent management application.

The primary aim of the study was to develop a talent management framework for the private sector. The secondary aim of this study was to conduct a modernist qualitative research study in order to explore and gain in-depth knowledge on talent management and what it entails.

More specifically, the study aimed to:

The study will contribute theoretically and practically to the operation of businesses in the private sector. From a theoretical perspective, the concept of talent management is illuminated, and knowledge in the field of talent management is extended through the development of a talent management framework. The practical contribution includes access to a well-integrated talent management framework, the purpose of which is to facilitate the implementation and improvement of talent management.

Furthermore, human resource practitioners and talent management specialists will benefit from the developed talent management framework as it will aid and guide the best practice of talent management. The implementation of the best practice of talent management will work only if the planning, implementation, tools, practices, tactics and dimensions of talent management are correctly applied as advised by the research insights.

Lastly, the current status of qualitative research, analytic induction and constructivist grounded theory will be expanded in the areas of talent management and human resource studies in South Africa.

A conceptual overview of talent management is provided below, as well as a discussion on key aspects linked to talent management, followed by a description of the research design and results.

Meyers and Van Woerkom (2013, p. 23) refer to talent management as ‘the systematic utilisation of human resource management (HRM) activities to attract, identify, develop, and retain individuals who are considered to be “talented”’. Meyer (2005) highlights that the attraction, growth and retention of talent are key factors for modern organisations that are knowledge driven. To be known as an employer of choice is considered a key organisational goal with direct benefits.

Chikumbi (2011) in a quantitative study investigating talent management and staff retention at the Bank of Zambia reveals that successful management of talent and employee retention leads to the organisation attaining a competitive edge. The study highlights that improved talent management leads to increased productivity, motivated staff, innovation and high employee contribution towards the organisation.

Odierno (2015) states that talent management offers real and tangible benefits of reducing recruitment costs, effective knowledge transfer, realisation of business strategy, delivery of cutting-edge services and products and the creation of a competitive advantage, in spite of the many challenges that organisations face in order to be sustainable. Ashton and Morton (2005) argue that good talent management is of strategic importance and can differentiate an organisation when it becomes a core competence, and its talent significantly improves strategic execution and operational excellence.

Prinsloo (2012), in her study of talent management and the psychological contract, provides evidence that management can use talent management practices to strengthen the psychological contract, leading to lower turnover, improved motivation productivity and loyalty. Bersin (2006) asserts that entry into the new era of talent management was because of more challenging people-related issues requiring tighter integration between human resource silos and the business. As an organisation strives to meet business goals, it must ensure that it has continuous and integrated processes for recruiting, training, managing, supporting and remunerating these people, and thus the need for talent management.

The next section will highlight the key aspects that could be linked to talent management (as identified out of the literature).

The qualitative research approach was selected because of the nature of the research process being flexible and emergent. Leedy and Ormrod (2010) indicate that qualitative research seeks a better understanding of complex situations and is sometimes exploratory in nature. A modernist qualitative research method was utilised in order to explore and gain in-depth knowledge on talent management and what it entails. With the study, the researcher hopes to further examine effective and leading best practices and challenges in the implementation process of talent management. Analytic induction and constructivist grounded theory was applied to analyse the data.

A multiple site case study design was adopted as the research was conducted in two different companies, mainly for the purposes of comparison and verification and to observe similarities and characteristics in the data sets collected (Moore, Lapan & Quartaroli, 2012). According to Bromley (1990), case study research is a systematic inquiry into an event or a set of related events which aims to describe and explain the phenomenon of interest. Leedy and Ormrod (2010) explain that in other instances researchers study two or more cases, often cases that differ in certain key ways in order to make comparisons, build theory or propose generalisations. Such studies are called multiple or collective case studies.

The field setting can be described as those companies who are identified as ‘Best Employers’ and are certified as such because they meet the requirements of being competitive in the areas of human resource management. They are serious about distinguishing themselves amongst industry peers in the critical areas of pay and benefits, training and development, career opportunities, working conditions and company culture. According to the Corporate Research Foundation (2011), widely known as the Corporate Research Foundation Institute, both the companies are listed in the top 10 ranking as best employers in the private sector.

Quandt, McDonald, Bell and Arcury (1999) indicated that gaining entry into the research field is quite an intensive process which requires perseverance, persistence and a lot of planning and communication skills. Yin (2011) stated that doing research implies interacting with the real world, situations and the people in them. In order to enter and exit research settings, some formality is required, particularly in obtaining the necessary permissions to do the study. In gaining entry to the field, key informants who provided authorisation and permission to conduct the study in the two identified companies provided a list of potential participants for the study. The potential participants were approached through electronic mail to communicate the purpose of the study, and thereafter the interviews were scheduled. The researcher adopted a formal interviewer role, attempting to elicit responses from the interviewees. The researcher used notes, as participants did not feel comfortable with the use of a voice-recording device. The researcher also assumed the role of an observer of the participants, the work environment and of the general company employees.

The purposive non-random or non-probability sampling method was selected for the study. In addition, convenience sampling was used with regard to the unsolicited documents issued to support the interview data. In this type of sampling, the researcher determines the most typical characteristics of the participants that could be included in the sample. Inclusion criteria are created based on the judgement of the researcher and are used to deliberately include specific participants in the study (Botma, Greeff, Mulaudzi & Wright, 2010). Characteristics essential for membership in the sample were as follows: the participant must be employed in an organisation that clearly practices talent management or must have been exposed to talent management practices as a manager, human resource practitioner or a beneficiary of the talent management programme.

The participants identified for the study were human resource practitioners, management members and beneficiaries of the talent management programmes. The selected organisations are widely and publicly recognised for having effective talent management programmes.

From Company A, nine participants took part in the first round of interviews, of which five were white females, two Indian females, one African female and one African male, all aged between 31 and 55 years, with one exception aged below 30. Four participants from Company B took part in this first round of interviews, of which all were white, two males and two females aged between 36 and 45. Interviews with participants lasted for 60–90 min.

The second round of interviews (to test the framework and collect additional data) were conducted with one person from Company A, who issued the researcher with seven documents, while from Company B four participants were interviewed. Four of the participants had a degree and had completed postgraduate studies. Four of the participants were already at middle and senior management levels in their careers, while one was at junior management and executive level.

To achieve the objectives and aims of the research, the researcher opted to combine a few methods of data collection. Botma et al. (2010) stated that qualitative researchers typically gather multiple forms of data and thus make use of a combination of research methods. The data collection methods refer to interviewing, observations, document collection and field notes. During the first round of the one-on-one interviews, the researcher followed a semi-structured approach of interviewing (at the organisations’ premises) in order to ensure that the open-ended questions were aligned with the research questions. This allowed flexibility and additional questions to arise during the interview process. The unstructured, in-depth interview was used during the second round of interviews to verify the initial proposed framework and the initial data analyses and coding of themes. Furthermore, the researcher conducted informal and unstructured observations of the participants, the settings, the environments and how people interacted in the company environment. The unsolicited documents in the form of internal company publications (agendas and minutes of meetings, internal office memos, financial records and annual reports) were issued to the researcher to support the data.

Leedy and Ormrod (2010) suggested that in conducting a productive interview, the researcher needs to record all responses verbatim, whether through the use of handwritten notes, shorthand and tape recordings or via direct capturing onto a computer, especially if it is an unstructured, in-depth interview. Interview data (from conversations, responses and information given by the research participants during the semi-structured interviews) were recorded in a written format. All the handwritten data were transferred and converted into an electronic format through a word processor. Furthermore, precautions were taken to safeguard data through backup on a compact disk, flash drive or a storage device that can be kept in a safe place.

In the current study, Johnson’s (2004) analytical induction procedure was applied, as implemented in Bondas (2006), as well as Charmaz’s (2000) constructivist grounded theory for operationalisation of data analysis. Findings of Marshall and Rossman (1995) were incorporated to cover the data testing, validation and verification phase. Analytical induction according to Smelser and Baltes (2001) is a research logic used to collect data, develop analysis and organise the presentation of research findings. Its formal objective is causal explanation, a specification of the individually necessary and jointly sufficient conditions for emergence of some part of social life. A slightly modified approach of analytical induction was applied as an overall and overarching data analysis method. Charmaz (2006) defines grounded theory as an inductive, iterative and comparative method geared towards theory construction. Grounded theory served the purpose of operationalisation of the data analysis, more specifically for coding purposes. The analysis process entailed a movement between deductive (theory and literature) and inductive analysis approaches (emergent insights from the data), resulting in the development of the talent management framework. Thomas (2006) views deductive data analysis as set out to test and illuminate consistency with prior assumptions, knowledge and theory.

The initial coding process achieved the clustering of similar data sets, and also the naming of these. Once the initial coding was completed, the researcher then progressed to the focused coding where the dense data were further de-layered into simpler and more specific sub-themes. These sub-themes are substantiated and supported by excerpts from the interview scripts. The focused coding stage resulted in the reduction of the data even further to identify and develop relevant sub-themes.

Theoretical coding consequently led to identification of the interrelatedness and the relationships between the concepts, themes and sub-themes, leading to the development of the talent management framework.

Furthermore, the sense-making of the data which resulted in the initial and proposed talent management framework was supported and influenced by a combination of prior knowledge (deductive) as well as emergent insights (inductive) from the data. Thomas (2006) refers to inductive analysis as a detailed process of reading the data to derive concepts, themes, models and frameworks. Thomas (2006) views deductive data analysis as set out to test and illuminate consistency with prior assumptions, knowledge and theory.

Following the analytic induction process, the researcher visited Companies A and B to verify, test, triangulate and validate the initially developed framework through unstructured follow-up interviews with the five participants. The initially proposed talent management framework was electronically mailed to the participants in order for them to advise on whether the framework could be improved, and to indicate the aspects needing improvement. The sixth participant opted not to have an interview but confirmed via electronic mail that the framework and the analysed data constituted a true reflection of what was discussed in the initial round of the interviews conducted. Out of the six participants, two took part in the initial interviews and confirmed and validated the data in comparison to the initially proposed framework. While the other four were not involved in the initial interviews, they could only comment mostly on the comprehensiveness and relevance and could also advise on what concepts should be incorporated into the initially proposed framework to improve it further.

Furthermore, the two companies offered the researcher eight documents in the form of internal publications. The offering of these documents was completely unsolicited. Therefore, the researcher applied initial, focused and theoretical coding principles to analyse the unstructured interviews and the documents.

In conducting the study, the researcher had to ensure the quality of the study in every step of the research, while also adhering to ethical conduct and standards.

For the trustworthiness of the study, the researcher had to ensure that the insights were that of the research participants and not her own. The researcher ensured the correct meaning and wording of what participants shared with her, by writing down and repeating it in the presence of the interviewee. Credibility triangulation was applied through collecting data from the two cases or companies as a basis for comparison. Furthermore, the researcher conducted the second round of interviews at the same companies with a mix of old and new participants. In addition, unsolicited documents were also used as a source of triangulation, verification and validation of the data, as well as for testing the proposed talent management framework.

Table 1 represents excerpts from interviews with all participants on how they defined talent. There seemed to be strong cohesion and consensus about the definition of talent. No conflicting views about the profiling of talent, even between Companies A and B, were noted. The data were further clustered into sub-themes to highlight the important concepts which led to the overall definition of talent.

Table 2 gives a brief definition of talent management as per the participants’ data input. The most important view arising from the data is the perception of talent management in the context of the overall employee life cycle (from hiring to exiting).

Table 3 demonstrates the participants’ views on what a talent management programme should entail in the best practice environment. The various sub-themes are segmented and supported by ideas and insights from the participants. The content of talent management and sub-themes are critical in the ultimate formulation of talent management dimensions or elements.

In Table 4, the participants and sources indicated how information technology and systems can assist and support the talent management processes and implementation. Having systems and processes will not necessarily make the practice of talent management successful but will enable the effectiveness of the process. These systems are mainly for reporting and data management purposes.

In this theme (see Table 5), the participants emphasised the importance of the overall culture, values and behaviour at work, which set the mood or temperature of the workplace environment. The point that is driven is that it is important to create a healthy and humane work environment where people can look forward to coming to work.

In Table 6, the participants and sources highlighted the importance of role clarity in the implementation of talent management. Clear accountabilities and deliverables need to be set, as to who does what, and who is responsible for what with regard to talent management.

Participants with exposure to talent management practices have reflected on the potential and the real challenges that come with the implementation of talent management (see Table 7). Issues that could potentially derail talent management implementation are flagged, thus creating awareness to help prevent future pitfalls. Much as the companies strive for people management best practices, certain areas still need improvement.

In this section (see Table 8), talent management is not viewed in isolation from the business operations. The importance and necessity of talent management is shared by the participant as a business imperative necessary for continuity and success of the business. The value add of talent management is highlighted below in the sub-themes that arose.

The participants shared their views on how to further enhance and improve current talent management practices by applying sophisticated measures relating to marketing and communication skills (see Table 9). In other words, talent management needs to be packaged attractively to appeal to the relevant audience.

The focus of this section (see Table 10) is to assess and determine talent management needs at various career life stages. In other words, what talent management interventions are needed for talent at various career life stages or cycles? This is extremely important as a blanket approach and one-glove-fits-all approach to talent management delivery must be avoided.

The proposed talent management framework is discussed in Figure 1.

The framework consists of the following:

Role players are the key stakeholders in the business who are responsible for the development, planning, implementation and improvement of talent management projects in the organisation. It is therefore critical to set key performance indicators for all the role players which pertain to talent management activities, and set reward systems which encourage strong performance.

The limitations of the study are mainly because of the purposive non-probability sampling method used. Consequently, the sampling method does not allow for generalisation of the findings to the general population (Thomas, 2006).

Furthermore, it would have been preferred to include a much bigger and broader sample (participants) and more cases (companies in the private sector), but this was not possible. The study is not rendered invalid and insufficient because of the elements stated above. With the few participants and two cases consulted in the study, the researcher has managed to collect comprehensive and detailed data, referred to as ‘thick’ descriptions. During the interviews and the data analysis process, the researcher realised and confirmed data saturation and sufficiency, as concepts, themes and patterns were repeated by the participants.

Anecdotally, it is known that talent management is a valuable best practice to have in an organisation. It would be of substantial value to have a set, defined, specific and consistent formulae to calculate the real economic value of investing in talent management activities quantitatively. Often, the investment in certain assets or finances spent in business activities is justified by possible financial and economic benefits that might be realised, and these financial returns are easy to track. Therefore, the practice of talent management will benefit from the ability to report on the actual economic value realised. This capability will improve buy in and a willingness to invest in high-quality talent management initiatives and resources.

In conclusion, the dimensions of talent management were identified and developed. The delivered talent management framework will guide and support the elevation of talent management practices in organisations in the private sector. The framework is comprehensive enough to cover all key and core dimensions of talent management practice. Ultimately, the framework is developed in such a way that it can lead to best practice.