Sparked by an ever-increasing amount of data, organizations have begun to analyze the data of their workforce in hopes of improving their business outcomes (Cascio, Boudreau, & Fink, 2019; Levenson, 2005). This practice is called people analytics and refers to “the analysis of employee and workforce data to reveal insights and provide recommendations to improve business outcomes” (Ferrar & Green, 2021). People analytics can support any employee-related decision (Ellmer & Reichel, 2021; Huselid & Minbaeva, 2019), help the Human Resources Management (HRM) function become more strategic (Angrave, Charlwood, Kirkpatrick, Lawrence, & Stuart, 2016), and allow an organization to prepare for the future (Guenole, Ferrar, & Feinzig, 2017). Practically, people analytics can, for example, identify internal and external talents, create succession pipelines, predict which talents may be tempted to leave the organization and provide recommendations on how they may be retained most efficiently (Minbaeva & Vardi, 2018; Rosenbaum, 2019; Yuan, Kroon, & Kramer, 2021). Due to these proposed benefits, organizations invest heavily in people analytics (Ledet, McNulty, Morales, & Shandell, 2021). Nevertheless, most organizations struggle to use it effectively (Ledet et al., 2020; Orgvue, 2019; Inc. Sierra-Cedar, 2019). Therefore, this dissertation aims to answer the following research question: How people analytics can be used to gain insights into and provide recommendations to enhance business outcomes? To answer this question, this dissertation addressed three challenges from the people analytics literature after a general introduction of the topic and challenges in chapter 1. The challenges, their importance and the results of the different chapters are briefly discussed in the following. How can an effective people analytics function be created? (challenge 1) This dissertation investigates what a people analytics function requires to be effective. This is important, as there is a rather limited understanding of how people analytics can be implemented effectively within the people analytics literature (Fernandez & Gallardo-Gallardo, 2020; Qamar & Samad, 2021). To address this issue, I conducted a narrative literature review (chapter 2) and follow-up qualitative research (chapter 3). For the literature review, the people analytics literature and broader more advanced, business intelligence domain that people analytics is part of (Davenport & Harris, 2017; Holsapple et al., 2014) were investigated. Based upon this, a number of crucial elements for an effective people analytics function were identified. However, a number of gaps within the literature were also found. Specifically, the relationships between the different elements and the processes a people analytics function requires to transform its inputs into outputs remained unclear. To address these gaps in our knowledge, qualitative follow-up research was conducted (chapter 3). To this end, 36 in-depth interviews with members of nine people analytics functions and their stakeholders were conducted. Based on the findings, eight processes were identified to transform the inputs into outputs. Some of these are related to the projects of a people analytics function (i.e. project selection, management, execution and the compliant and ethical behavior of people analytics experts) and others to their stakeholders (i.e. the attitude of stakeholders, collaborations, partnerships and the transparency of people analytics function to their stakeholders). Furthermore, the “People Analytics Effectiveness Model” together with seven propositions to guide future research were developed. These propositions illustrated on one hand the relative importance of the different elements a people analytics function requires. For example, having data was found to be more crucial than a specific organizational culture. On the other hand, the propositions showed the relationships between the different elements: Delivering high-quality people analytics products, for instance, increased the reputation of the people analytics function. Furthermore, as the reputation increased, people analytics functions were typically provided with more inputs and better contextual factors, such as access to new datasets and increased support from senior management. How can people analytics be used to enhance employee well-being and performance? (challenge 2) This dissertation demonstrates how people analytics can be used to enhance employee well-being and performance through two use cases. This is relevant, as organizations increasingly consider how the interest of the manager and employees may be achieved in conjunction (Battilana, Obloj, Pache, & Sengul, 2020; Paauwe, 2004). However, there are few empirical studies on people analytics that demonstrate it can provide insights and recommendations that support employee well-being or performance (Margherita, 2021). In chapter 4, I therefore demonstrate how people analytics can be used to evaluate whether the decision of a company to adopt the agile way of working is beneficial to employee well-being and performance. The agile way of working is an increasingly popular way of working among teams, that is characterized by self-management, face-to-face communication, reflexivity, a quick product turnaround and customer interaction (Beck et al., 2001). To do this, I developed a survey focused on the agile way of working and tested among 97 teams from an organization whether the agile way of working leads to beneficial outcomes. Based upon the results, it appeared that this was indeed the case: The agile way of working was found to be related to increased levels of team engagement and performance regardless of teams’ functional domains. Moreover, it was found that these effects are partially mediated by psychological safety climate. Following this research, the company central to this research now has data-driven insights that support the decision to implement the agile way of working across a variety of functional domains. In chapter 5, I show how people analytics can be used to provide insights about employee well-being and performance and inform job design practices. Specifically, I tested in line with the HRM literature (e.g. Ayala et al., 2017; Benitez et al., 2019; Tordera et al., 2020) whether complex trade-off patterns may occur between employee well-being and performance. Based upon data of 5,729 employees working in a large financial organization, I find support for the notion that five well-being and performance profiles exist: 1. Low well-being/low performance, 2. low well-being/medium performance, 3. high well-being/medium performance, 4. high well-being/high performance, and 5. high well-being/top performance. Furthermore, it appeared that specific job demands and resources are related to these well-being and performance profiles. Specifically, employees with more learning and development opportunities, more social support from colleagues, more autonomy, and less role-conflict were related to the high well-being profiles. Additionally, employees with more role clarity, more performance feedback, more autonomy, and less work pressure were related to the high- and top-performance profiles. Finally, communication and social support from the manager were found to be relatively weak antecedents of the different profiles. How can people analytics departments benefit from a collaboration with academia? (challenge 3) The final challenge this dissertation addresses is how people analytics departments may benefit from a collaboration with academia. This is an important topic, as a competency gap among people analytics practitioners has been identified as being one of the main obstacles for organizations to use people analytics effectively (Fernandez & Gallardo-Gallardo, 2020; McCartney et al., 2020). Specifically, Human Resource (HR) professionals usually fall short of statistical skills and statistically strong individuals usually lack business acumen and HR knowledge (Andersen, 2016; McCartney et al., 2020; Rasmussen & Ulrich, 2015). As a potential solution, a collaboration with academia has been suggested (Simón & Ferreiro, 2018; Van der Togt & Rasmussen, 2017). Specifically, the so-called “boundary spawners”, in which for example PhD candidates bridge the gap between academia and a people analytics department is frequently mentioned within the people analytics literature (Minbaeva, 2018; Van der Togt & Rasmussen, 2017). To illustrate how this may work in practice, chapter 6 of this dissertation discusses the benefits, challenges and potential ways to navigate through these challenges based upon my own experience of working in a joined PhD trajectory for 4.5 years. In total, six benefits and five challenges were identified in this chapter. Among the benefits, the opportunity to conduct relevant research for both parties and the time and opportunity to identify and address real and pressing business needs are for instance discussed. With regards to the challenges, topics such as the different potential interest for both parties and limitations regarding the data are described. Discussion After addressing the challenges, the discussion follows in chapter 7. This chapter holds a summary of the main findings of this dissertation, their theoretical and practical contributions, strengths and limitations and points of reflection. The primary contribution of this dissertation is to explore how people analytics can be used to gain insights into and provide recommendations to enhance business outcomes. To this end, the discussion chapter described what a people analytics function requires to be effective, investigated two potential use cases and showed how a collaboration with academics may be beneficial and challenging. Furthermore, four points of reflection are discussed within this chapter. First, I describe how people analytics can contribute to and benefit from the employee experience. The employee experience is one of the actual trends within the field of HRM and emphasizes that organizations need to consider the wants, needs and expectations of their employees from the moment of their recruitment all the way to the moment they leave the organization. Furthermore, as each employee is different, employee experience experts emphasize the need to offer a differentiating employee experience depending on the wants, needs and expectations of specific employees (Dye et al., 2020; Whitter, 2019). In this section, five concrete ways in which people analytics can support employee experience experts through data-driven insight are discussed. Furthermore, the reverse value of the employee experience for people analytics is also discussed. Specifically, whereas a substantial amount of HR professionals are confused or skeptical about the use of people analytics (Guenole & Feinzig, 2018), the far majority is enthousiastic about improving the employee experience (Dye et al., 2020). By offering insights and recommendations on a topic HR professionals are enthousiastic about, it is suggested people analytics can improve the number of data-driven decisions taken within the HRM function, and through this, enhance employee well-being and performance. Second, I explore how data science and HRM research can become more intertwined. On one hand, it is discussed how HRM scholars can utilize the data sources and analysis techniques used by data scientists to make new contributions to the HRM literature. Specifically, the analysis of non-survey data, such as unstructured text and (HRM) system data, is highlighted as a method to unveil relevant insights into the sentiment, behavior, and perceptions of employees (e.g., Gloor et al., 2017; Yang et al., 2021). On the other hand, it is suggested that data scientists may benefit more from using survey data, theories, and analysis and interpretation techniques common among HRM scholars. This could help them to avoid oversimplifying reality (e.g., human beings are more complex and unpredictable than the numbers captured in the HR information system or their model output may suggest) and avoid misinterpretations, miscalculations and errors as a result (Giermindl et al., 2021). Third, I discuss the topic of ethics within people analytics. Despite of the benefits of people analytics that this dissertation highlighted, people analytics has also been used by organizations for unethical matters, such as intrusively tracking employees (Ajunwa et al., 2017; Tursunbayeva et al., 2021), (unintentionally) discrimination (Dastin, 2018) or even firing employees (e.g., Business Internet Tech, 2021). Therefore, the ethical aspect of people analytics are highlighted in this section. On one side of the spectrum, I discuss that it is always necessary to operate within the boundaries of the law but not always sufficient, and explore the negative consequences of behaving unethically for the people analytics function itself. On the other end of the spectrum, I also discuss three examples in which I believe it is ethically just to push for the use of people analytics. Specifically, I advocate that data-driven insights can bring more equality and fairness to the workplace, increase the employability of employees and enhance employee well-being. Therefore, it is concluded in this section people analytics is not necessarly good or evil and that it should be reviewed on a case-by-case basis whether it is ethical to use people analytics. Fourth, I discuss the governance of people analytics. Although in this dissertation various governance aspects are discussed (e.g., data governance, governance of the people analytics function), I suggest people analytics scholars and practitioners should also pay attention to the question of who owns people analytics. This is important, as software providers are increasingly facilitating HR experts and line managers to run their own (semi) automated advanced analytics models. However, as these professionals typically lack the skills, there is a high risk of misinterpretation of the results, finding incorrect findings due to pure chance (e.g., as a result of the error margin for all statistical models) and statistical artifacts such as reverse causal relationships and spurious effects. I therefore recommend caution in enabling professionals who lack the capabilities to run advanced analytical models in fear of wasting valuable organizational resources on the wrong actions, and to focus on building their analytical capability first. Finally, I conclude this dissertation by emphasizing that the age of people analytics is just beginning. Continued attention from academics and practitioners will therefore be needed to ensure that the right bridges are built between different worlds to be effective at people analytics: These are the worlds of HRM and technology; the worlds of academia and practice; the worlds of data science practitioners and HR practitioners; the worlds of subjectivity and objectivity; and the worlds of employee well-being and performance.