This study is conducted by a systematic review, following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) standards to ensure a transparent, rigorous and replicable review process. The PRISMA approach comprises three stages: doing a keyword database search (identification); applying academic filtration, manual screening and coding (screening); and employing snowball sampling to choose the final sample of articles for study (included) (Sarker & Bartok 2024; Venesz, Dőry & Raišienė 2022). Figure 1 displays a flowchart illustrating the process of searching and selecting.

A comprehensive search was conducted across three academic databases: Scopus, EBSCOhost and ScienceDirect. These databases were selected because they index high-quality, peer-reviewed journals that are accredited by the Department of Higher Education and Training (DHET) and widely recognised by international bodies such as the United Nations Educational, Scientific, and Cultural Organisation (UNESCO) and the Organisation for Economic Co-operation and Development (OECD) and aligned with the Sustainable Development Goals (SDGs). ScienceDirect covers studies on the social sciences and humanities. EBSCOhost shelters high-quality articles licensed from reputable publishers recognised by library professionals, and Scopus is a comprehensive, multidisciplinary and trusted academic database. Thorough the application of (Table 2), the search was limited to works published between January 2019 and March 2025, to capture recent developments in university-led employability and entrepreneurship support activities. The EndNote software was used to organise and track eligible studies for systematic analysis.

‘Graduates’ employability’ and ‘entrepreneurship’ strategies were gathered from reputable sources and published research. The initial search yielded 101 publications; the researcher established a protocol to select relevant studies, reducing the count to 97 journal articles after filtration. After the duplicates were removed, 83 articles were reviewed for relevance to graduates’ employability and entrepreneurship support in universities. Consequently, 30 articles were excluded, leading to a final selection of 53 relevant studies for analysis, demonstrating consensus among the examined studies.

Finally, 53 full-text publications were retrieved and assessed for suitability based on three criteria: they must be empirical, focus on graduates’ employability or entrepreneurship, and examine these aspects within academic settings. Publications not meeting at least one criterion were disqualified. Ultimately, 53 publications met the criteria for analysis (see Figure 1).

The coding process was conducted by using NVivo (version 15) to enhance rigour and transparency. After finalising our codebook through initial manual piloting, all 53 full-text articles were imported into NVivo as source documents. Codes defined and refined during the codebook development were created as ‘nodes’ in NVivo, and each document was systematically coded against these nodes. NVivo’s query tools (e.g. text search and coding comparison) were then used for verifying coding consistency, uncovering co-occurrence patterns and ensuring that no relevant passages were overlooked. An export of the NVivo coding matrix provided a comprehensive audit trail of every coding decision, including source references and coding frequencies.

Bias was minimised through a multicoder, blinded approach. Documents were coded in a randomised order, and coders were ‘blind’ to author names and publication outlets. Intercoder reliability was assessed via NVivo’s coding comparison feature, yielding a Cohen’s κ of 0.82; discrepancies were discussed during consensus meetings and resolved by a third reviewer (co-author). All codebook revisions, adjudication notes and NVivo query outputs were logged to maintain an audit trail. Finally, a 20% subsample of NVivo-coded documents was rechecked by all authors against the original texts to confirm fidelity, and no systematic deviations were found, thereby underpinning the validity and reliability of our synthesis.

This systematic review uses the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) approach to identify pertinent studies (Figure 1). The search criteria and databases were expanded to encompass the worldwide aspect of the following the research protocol depicted in Table 1. The search was confined to three databases recognised for their quality and commitment to science, to ensure the rigour and quality of the papers included in the review. The quality assurance of the selected papers was upheld through a comprehensive examination, yet this resulted in the selection of only 53 papers from six continents.

Ethical clearance to conduct this study was obtained from the Durban University of Technology Institutional Research Ethics Committee on 01 July 2025. The ethical clearance number is IREC 057/25.

The findings highlight the key domains utilised by universities in developed and developing countries. This helps to consider the various perspectives explaining the impacts of university support on graduates’ employability and the context of entrepreneurship. Given the diversity of support, it can be concluded that the mechanisms for applying these supports are pretty fragmented. No single form of support dominates the research for popularity in academic papers. Table 3 illustrates the types of support that were cited as reasons for the research on graduates’ employability and entrepreneurship support.

In the sample, the most frequently cited form of support for explaining how universities impact graduates’ employability and entrepreneurship was the curriculum (n = 21). This curriculum approach accounts for 39% of the samples, and the 21 universities in our sample have implemented it only in developing countries. Curriculum transformation involves embedding employability support and integrating entrepreneurship education into the curriculum (Koseda et al. 2024; Oyinlola et al. 2024; Santos-Jaén et al. 2022). The second-most cited forms of support are extracurricular activities (n = 10) and university–industry collaboration (n = 9), and the universities are from developed countries and utilise extracurricular activities to enhance entrepreneurship competencies and employability skills (Bodolica et al. 2021; Surendran et al. 2023). Six studies from developing countries represented support structures, while four of the articles from these identified policy and institutional support as a means to promote employability skills and an entrepreneurial mindset (Al-Abri et al. 2024; Langseth, Jacobsen & Haugsbakken 2022; Ncanywa 2024). Lastly, two studies from developed countries emphasised community engagement (Medina-Bueno et al. 2025), while 3.3% of studies from developing countries addressed the resource provision necessary to support employability and entrepreneurship (Borah et al. 2021).

Most research utilised quantitative methods, with 38 out of 53 studies employing exclusively quantitative designs, primarily through questionnaire surveys (n = 38). In contrast, a smaller group used secondary data (n = 13) for their analysis (Figure 2). The advantages of these methods include the ability to collect data from a large number of individuals, resulting in robust outcomes that can make broad assertions about graduates’ job prospects and business success. Standard tools allow for easy comparisons across different situations and times, and statistical tests can help identify connections among factors (like how curricular support affects entrepreneurial intentions). However, they have drawbacks, as they heavily depend on self-reported answers, which can lead to biased responses (such as wanting to appear favourable), and they typically capture only a single point in time, making it challenging to determine cause and effect. Moreover, survey items may oversimplify complex phenomena, overlooking individual students’ nuanced motivations and experiences.

In contrast, 17 studies employed qualitative methods, primarily using interviews (n = 22) and, to a lesser extent, focus groups (n = 7). The advantages of qualitative methods are that they can deeply explore and understand how students and stakeholders view employability and entrepreneurship support, as well as highlighting specific challenges, like those faced by female graduates in particular cultural settings. Rich textual data allow for the emergence of unanticipated themes and the generation of theory grounded in participants’ lived experiences. However, smaller, often purposive samples limit these approaches by hindering broad generalisability. They also require significant interpretive skills from researchers and are vulnerable to subjectivity and potential confirmation biases during coding and analysis.

Mixed methods designs, employed by 17 papers, combine quantitative and qualitative data, offering benefits such as using different methods concurrently and enhancing result validation. For example, a survey might measure the effectiveness of standard mentorship programmes. At the same time, follow-up interviews elucidate how those programmes function in real-life scenarios and why they succeed or fail in various circumstances. Yet mixed methods studies encounter practical challenges: they require additional resources, time and expertise to design and integrate diverse data streams effectively. Balancing the rigour of each component and weaving them into coherent conclusions can be complex, and publication venues may favour one method over the other, leading to uneven reporting of methods and results.

The reviewed studies reveal a striking diversity of support mechanisms, ranging from curriculum integration and extracurricular activities to industry partnerships and policy initiatives. However, they also indicate a lack of coherence and coordination across these approaches. No single form of support dominates: curricula account for roughly 39% of interventions, while extracurricular initiatives, university–industry collaborations, support structures, policy efforts, community engagement and resource provisions each occupy much smaller niches. This fragmentation suggests that universities often implement discrete, siloed programmes rather than a coordinated strategy, undermining potential synergies (e.g. embedding entrepreneurship courses alongside WIL and mentorship) and risking patchy student experiences.

Such fragmentation has significant methodological and evaluative implications. Because interventions are unevenly distributed across contexts, with curriculum-driven support in developing countries versus extracurricular and collaboration models in wealthier settings, it becomes difficult to compare outcomes or build cumulative evidence on ‘what works’ for graduate employability and entrepreneurship. Moreover, fragmented mechanisms may foster piecemeal data collection and reporting, complicating meta-analysis and hindering policymakers and practitioners from identifying best practices that are both scalable and adaptable across diverse institutional environments.

From a strategic standpoint, the observed fragmentation points to missed opportunities for holistic support. When career services, academic curricula, incubator centres and industry partnerships operate in isolation, universities risk duplicating their efforts, misallocating scarce resources and failing to engage key stakeholders collaboratively. This siloed approach is especially problematic for female graduates, who already face systemic barriers, as it limits tailored, wrap-around support that could bridge gaps among academic learning, professional networks and entrepreneurial ecosystems. Addressing these shortcomings requires moving beyond isolated interventions towards an integrated model, such as the holistic support framework proposed in this study, which aligns curricular, extracurricular, structural and stakeholder-driven initiatives into a coherent institutional strategy.

Such fragmentation has significant methodological and evaluative implications. Because support programmes vary significantly between different areas, developing countries focus on curriculum-based help. In contrast, richer countries use extracurricular activities and collaboration, making it hard to compare results or gather solid evidence on what improves graduate employability and entrepreneurship. Moreover, these fragmented arrangements can result in uneven documentation practices, complicating meta-analysis and hindering the ability of policymakers and practitioners to identify best practices that are both scalable and adaptable across diverse institutional environments.

From a strategic standpoint, the observed fragmentation points to missed opportunities for holistic support. When career services, academic curricula, incubation centres and industry partnerships operate in isolation, Arthur and Koomson (2024) argue that universities risk duplicating their efforts, misallocating scarce resources and failing to engage key stakeholders collaboratively. This siloed approach is especially problematic for female graduates, who already face systemic barriers (Al-Saadi et al. 2024). According to Aldhaheri (2023), it limits tailored, wrap-around support that could bridge gaps among academic learning, professional networks and entrepreneurial ecosystems. Addressing these shortcomings requires moving beyond isolated interventions towards an integrated model, such as the holistic support framework proposed in this study, that aligns curricular, extracurricular, structural and stakeholder-driven initiatives into a coherent institutional strategy.

This study’s findings indicate that higher education institutions across various global contexts have adopted various best practices to enhance graduate employability and foster entrepreneurial capabilities. However, these practices often appear fragmented in the literature. Moqokama and Van der Westhuizen (2024) argued that many institutions concentrate exclusively on either employability skills development or entrepreneurial education, rather than adopting an integrated, systemic approach. In response, this section proposes a holistic framework displayed in (Figure 10) that brings together these elements in a cohesive model to guide universities in designing effective, contextually responsive interventions.

The proposed framework draws from empirical evidence and established practices observed in both developed and developing countries, reflecting contextual differences across Europe, Asia and Africa. European institutions are noted for their structured and institutionalised employability systems (Guerrero, Heaton & Urbano 2020; Succi & Canovi 2019), while Asian universities demonstrate a strong emphasis on digital innovation and government-backed entrepreneurial initiatives (Khan et al. 2024; Song & Lu 2024). In contrast, African institutions often focus on grassroots entrepreneurship and internship expansion (Shenkoya, Hwang & Sung 2023).

This framework addresses the gap raised by UNESCO (2020) and Srivastava et al. (2025) when arguing that there is inequality in students’ development strategies between developed and developing countries. This model intertwines the best practices from these economies into one structure that guides universities when developing their graduates. The main idea of this model is to create a comprehensive system that involves different departments and incorporates input from various stakeholders. This system aligns with the research of Shen et al. (2025) and Santos-Jaén et al. (2022), as it is designed to prepare graduates with the skills, experiences and connections they need to succeed in a rapidly evolving, knowledge-driven economy. The model integrates institutional strategies, academic programming and stakeholder collaboration to provide all-encompassing support for graduate development.

The framework consists of two interconnected structural components:

The proposed framework directly addresses the identified gaps and challenges by integrating various support mechanisms into a cohesive model. This integration is crucial for overcoming the fragmentation observed in current practices. The framework aligns curricular, extracurricular, structural and stakeholder-driven initiatives, ensuring that all aspects of graduate support are interconnected and mutually reinforcing.

The framework emphasises embedding employability and entrepreneurship education in the curriculum. This approach connects with Chowdhury (2020), who found that students’ participation in extracurricular activities is lower and that they prioritise their involvement in the educational curriculum. The framework addresses the gap in which some students might miss out on essential skills as a result of limited access to extracurricular programmes by infusing these components into the core academic experience. Additionally, the inclusion of extracurricular activities such as incubators, accelerators and student competitions provides practical, hands-on experiences that complement theoretical knowledge. This dual approach helps bridge the gap between academic learning and real-world application, enhancing students’ readiness for both employment and entrepreneurial ventures.

Moreover, the framework advocates for active collaboration with industry partners, government bodies, alumni and internal university units. This multi-stakeholder approach ensures that the support provided is relevant to current market demands and leverages external expertise and resources. The framework addresses the challenge of aligning academic programmes with industry needs and across different university departments and external entities by fostering these partnerships. This holistic structure mitigates the risk of isolated interventions and ensures that all support mechanisms are aligned and coordinated. The framework’s structural components, such as interdepartmental support systems and stakeholder integration mechanisms, promote coordination and continuous work synergistically. Furthermore, the framework is designed to be adaptable to various institutional and cultural contexts. By drawing on best practices from developed and developing countries, the framework provides a flexible model that can be tailored to specific regional needs and constraints. This adaptability tackles the challenge of implementing a uniform approach in diverse educational environments.

Implementing the proposed holistic support framework involves several practical steps and considerations. Successful implementation requires a strong commitment from university leadership. Institutions must prioritise employability and entrepreneurship as core components of their strategic plans and allocate sufficient resources to support these initiatives. Pantaruk et al. (2025) argue that establishing effective interdepartmental collaboration is crucial. Husain et al. (2024) point out that universities need to create platforms for regular communication and coordination among different stakeholders, such as career services, academic faculties and entrepreneurship centres. This collaboration ensures that all support mechanisms are aligned and mutually reinforcing, and building and maintaining partnerships with industry, government and alumni is essential (Herbert et al. 2020). Universities should establish formal agreements and regular engagement activities to ensure that these partnerships are productive and sustainable. This policy may involve creating advisory boards, hosting industry events and facilitating internships and mentorship programmes. Integrating employability and entrepreneurship education into the curriculum requires careful planning and design. Dal-Soto et al. (2021) contend that universities should establish clear metrics for success and regularly collect data on student outcomes. This data can inform continuous improvement efforts and demonstrate the impact of the support mechanisms.

However, implementing the framework may encounter several challenges. Resource constraints are a significant concern, particularly for universities in developing countries. Creative solutions such as leveraging technology and forming strategic partnerships can help mitigate these constraints. Introducing new initiatives may encounter resistance from faculty, students or other stakeholders who are accustomed to traditional approaches. Effective communication and change management strategies are essential to address this resistance and build buy-in for the new framework. While the framework is designed to be adaptable, scaling it across different departments and campuses can be complex. Universities need to pilot the framework in specific areas, gather feedback and make necessary adjustments before full-scale implementation. Ensuring the long-term sustainability of the framework requires ongoing commitment and investment. Universities must continuously seek funding opportunities, engage stakeholders and adapt to changing market demands to maintain the relevance and effectiveness of the support mechanisms. By addressing these practical considerations and potential challenges, universities can effectively implement the holistic support framework, enhancing graduate employability and entrepreneurship in alignment with global best practices and local needs.