Challenges and Opportunities for Implementing a Data Management Programme in Higher Education

1.Introduction

Universities often find themselves accumulating data without taking effective action, which can lead to passivity. In response to issues, isolated solutions may exacerbate challenges instead of creating sustainable resolutions. While data alone does not inherently enhance decision-making, it is the insights derived from interpreting that data that foster informed actions. Transforming data into actionable insights is crucial, typically by understanding the interplay between various information sources rather than relying solely on a single dataset. No-code tools like Coda and Notion can significantly aid in this process by enabling rapid automation and workflow management. For example, with Coda, teams can create dynamic document-based applications that incorporate real-time data integration, facilitating instant updates and efficient collaboration. Similarly, Notion allows users to create customisable dashboards that gather information from multiple sources, promoting a holistic view of data that enhances decision-making. However, a key challenge remains: the lack of clear requirements knowledge can hinder the effective use of these tools. As requirements evolve, no-code solutions provide the flexibility to easily test new features and modify existing ones without requiring extensive technical expertise. This iterative approach not only speeds up the development process but also encourages continuous refinement as user needs change. Investing in no-code platforms must be paired with thoughtful policies and sustainable support structures. These tools can bridge data silos, reduce redundancy, and promote a more cohesive understanding of information, making data more accessible, usable, and reliable. This integration minimises data collection efforts and cuts down on transactional costs related to data management, ultimately enhancing the organisation's capacity to transform insights into action.

2.The Importance of Data Management in Higher Education

The importance of managing and planning data using professional approaches to data management in higher education institutions is recognisable. These institutions accept that they should use data to make decisions and enhance services and resources for staff and students. These institutions have made significant advances in various complex and diverse data issues—in terms of generation and storage capacity—yet they are often criticised for not exploiting such data more effectively to support higher quality in pedagogy, research, outreach, and administration. (Elisa Raffaghelli, 2020)

Universities have a long history of using data. This includes statutory data about their program offerings, information about research outputs and outcomes, and traces of interaction between prospective and enrolled students and staff with the higher education institution's integrated services. The use of data to aid in research and teaching has been developed and acted upon by a number of the most significant universities in the world, being intrinsic parts of what makes them more innovative, efficient, and capable of producing high-quality publications and graduates. However, while a great deal is made about the importance of understanding the data deluge and grasping its potential, a number of reports and discussions in higher education have noted that, on the whole, the use of data has not impacted most sectors to the extent that it could or should have done so far. It is within such contexts that the principles of management and planning of data using professional data management approaches become very important. While data management is recognised by institutions as effectively achieved, the principles of data management in use may not be formalised. Factors that improve the quality of data management include mechanisms and incentives that encourage the establishment of data structures, standardisation structures, responsibilities, transparent governance, and related advances in culture, training, and application of technology.

3.Key Components of a Data Management Programme

Meaningful programs are built on a firm foundation; the same could be said of data as well. Without a solid underpinning infrastructure, the data will be of little use to those seeking to gain some value from it. In the same way that a successful exercise regimen to improve health requires a balanced approach of strength, endurance, and flexibility, developing an approach that balances access, security, and privacy supports the goals of the institution. These are typically implemented through a variety of technical, administrative, and regularly supported procedures and business rules. Failure to implement one of these key areas might reduce the effectiveness of the implementation and the program as well.

There are a number of building blocks that serve as key components of an effective infrastructure program. These are not all-inclusive, nor are they presented in any order. The building blocks are intentionally described at a high level and are used to set expectations for success within the data management community. It is worthwhile to observe that these components are not all-encompassing; for a data management program to be effective, leadership needs to have an open, engaged view of the potential of data as a strategic asset, an ability to create and articulate that potential, and a commitment to a long-term plan to develop the potential return. These components are necessary but not sufficient.

3.1.Data Governance

Data governance revolves around the overall management of data, emphasising business and policy-driven activities. Formal data governance programs in a higher education institution provide value and accountability across the range of stakeholders with data-related interests and accountabilities, providing policy, procedural, and technical guidance to ensure that institutional data is managed effectively. Competing with organisational data governance will be the governance that is driven by the political and legal directives of the statutory bodies to which a higher education institution is accountable. Institutional data governance will need to be in alignment with these requirements. The increasing value of institutional data in funding, benchmarking, and reputation necessitates increased attention to data quality, accuracy, reliability, and security.

High-level data governance usually concerns itself with the bigger picture of data, establishing infrastructure to embed data practices within decision-making processes. An effective higher education organisation usually has a governance committee examining relevant key performance indicators and performance criteria. In this context, key questions concerning data can be formulated, such as: Where is data owned and officially mastered? Who is responsible for collecting and authorising the data submitted to the regulatory bodies? Are there regulations in place that accurately govern data? Do data tools need to be developed to facilitate effective data-related decision-making processes? Similarly, self-regulation is employed by higher education institutions through peer pressure, auditing, and benchmarking exercises, including ranking league tables that rely heavily on the quality of the submission and verification of the data provided. Higher penalties are often associated with unreliable data submission, questioning the balance between time investment and direct correlation to funding, which policy drivers tend to impose.

3.2.Data Quality Management

Data quality management is a critical sub-process in data management that ensures that the data stored in the data repository, database, or data warehouse is of high quality and suitable for the purpose for which it is intended to be used. Ignoring data quality management can lead directly to the low productivity of the project making use of the data and indirectly enforce bad management practices and decision-making in the organisation using it. Data quality is of fundamental importance, but assessing it and improving it can be difficult to do in practice. The main concern of motivating people to apply data quality improvement techniques to databases is explained. They suggest that while it is often possible to easily estimate the costs of poor data quality, it is much more difficult to estimate the benefits to the organisation of implementing data quality management processes. There are many different facets to data quality; these include accuracy, compliance with data requirements, timeliness of data, and consistency and understandability of the data.

In a survey of data quality, it is expounded upon that there are five essential facets of data quality, although the view can be summarised as the result being suitable, whatever the context. Five dimensions of data quality are highlighted, namely intrinsic, contextual, representational, accessibility, and conflict. Other surveys have also found that there are a number of different dimensions of data quality. A range of data quality management techniques exist for monitoring data quality based on rules and data profiling, for assessing data quality using user surveys, manual data inspection, and relational and statistical analysis, and for the improvement of data quality. These range from data auditing based on business rules and support for data quality rules in the data definition and associated documentation, where automated data profiling applications search the databases to understand the contents and further identify situations that need improvement.

3.3.Data Security and Privacy

Data Security ensures that data cannot be accessed by unauthorized users while Data Privacy decides who can access the data, data owner/custodian. The UK and EU universities gather and store a lot of personal data, which includes students’ academic records and some financial details. They keep records of students’ names, national insurance numbers, passport, credit card, and photographs. Personal information can be identified to an individual or identifiable. Attributes that can be linked to anyone will be personal data.

In the UK the Data Protection Act 2018 and UK General Data Protection Regulation set out the rules on data protection and privacy. The regulations are designed so that the collection and processing of personal data is lawful, fair and for specific purpose. In the European Union also, the General Data Protection Regulation (GDPR) has been laid down that provides guidelines of proper use of personal data. Both the UK GDPR and the EU GDPR seek to protect personal data and the rights of individuals.

Academics must remember that data breaches and invasions of data privacy have real consequences not just for them. When someone accesses, uses, or shares your personal information without your consent, it can cause legal and financial issues. This may include fines and loss of reputation. Under the GDPR rule, for example, companies could be fined a maximum of €20 million or 4% of their total annual worldwide turnover (whichever is higher).

Educational institutions need to have strong data protection mechanisms to comply. This is about maintaining documentation of the types of personal data collected and purpose of data, Keeping records up to date and accuracy, and having appropriate security measures in place to avoid unauthorized access, and loss or destruction of data. Educational institutions must prepare and implement proper data retention policies, ensuring that data gets securely deleted when not in use. Also, institutions must conduct regular training sessions for staff and students to promote awareness of data protection principles.

When universities follow rules and protect data well, they can keep the personal data of their community safe as well as follow the law. This builds trust with the students, staff and others.

3.4.Data Integration

Most of the available data have limitations, at least as a result of their being embedded in a specific subject or space. Researchers generally work with a combination of surveys, direct measurements, and other data to find generalisable results. Because of these limitations with research as a basis, it follows that a data management program should be based on a number of block-level data considerations or imperatives. As a first step towards building a data management program, we suggest considering the form of interchangeable blocks of data that might feature in the program. Attributes and areas to be concerned with in a data management program include access control, standards, data quality, documentation of data, descriptions of data, attitudes towards the sharing of data, long-term planning, and integrated staff training and funding service delivery.

Access control would ensure that data are safely and securely available in the research environment. Standards and best practices in data management and curation would centre on the thrust of data governance and lend credibility to research focus areas. Data quality and how the data are captured, conjoined with ongoing management and assurance, would lend veracity to the research process. Documentation and describing the data, in particular, must be sufficient to enable multiple researchers' viewpoints and investigative tasks. Attitudes towards data and data sharing must be articulated, especially when dealing with primary data. Reflecting on the total scientific data process, researchers must recognise their role in holding and providing for future researchers. Long-term viability and the documentation of research data with regard to appropriate non-proprietary data formats, such as common modes of exchange, and the potential to take full advantage of external standards is particularly important to aid in data reuse and evaluation.

4.Challenges in Implementing Data Management Programmes

Challenges, as perceived by the different discipline subject librarians, in offering the Data Management Programs were discussed through in-person consultations and feedback sessions and included the following points. Subject librarians have a full-time commitment to discipline, a barrage of responsibilities, and recommended collections, while the installation of a data management program will add to their portfolio. The implication was that other tasks would suffer, and the feeling was that these sessions were to be perceived as another area of expertise expected from subject librarians. The recommended collection development add-on to their duties was considered a major issue; although somewhat exaggerated, the perception of a full-time productivity robot added to the perception of subject librarians as generalists. As a result, the logistical effort of coordinating these sessions with subject librarians might have been met with resistance. In addition, the learning curve for first-time topics was a deterrent. The introduction and propagation of these programs, in general, were thus seen as a potential source of confrontation.

The fact that they were placed by the administrative library had caused students to assume that they were just another class and treated them as such. Some students asked for verification from the professors of the significance or viability of their attendance, with students in general not displaying a solid grasp of their specific research assignment. In essence, what are subject librarians doing with data? Were questions asked? Moreover, the expectation that students would readily adopt data best practices was a bit naive since a clear articulation of the demand for data management from lecturers was absent, which created a leap for learning without an understanding of the subject. After an introduction to the program and session explanation, students chose not to attend and did not appreciate the value of the data sessions. Since the learning curve is so broad, focusing on a different section of this offer might have been a more suitable option. Other ramifications for the day-to-day topic and the necessity of other disciplinary-level programs could be considered or emphasised to the academic audience in general, with structured co-teaching with librarians. Finally, academic awareness for these discipline-specific data training classes was a continual message, and triple redundancy in communication was the key to the success of the sessions.

4.1.Lack of Institutional Support

Higher education institutions have a strong commitment to research and science. Yet this commitment extends to a range of activities, many of which are administratively managed, particularly as funding bodies tighten their focus on the management of research grants and the accountability for public funds. The administration of awards, intellectual property licensing and patenting, copyright, and research commercialisation are all areas of academic life that often attract a much higher level of institutional commitment. Data management has usually not been integrated into the systems and practices of a higher education institution, so early discussions in this area will need to give department, faculty, and institutional management advice and perhaps suggest policy changes.

It was found that this lack of institutional policy increased the resistance to change, so implementation was prolonged. Much of the data is unstructured or semi-structured and difficult to store accurately, making it difficult for third parties to develop business intelligence systems and normal procedures. It was found that the longer the delay in the development of appropriate data structures or reporting tools introduced, the stronger the resistance to implementing data structures in the first place. The biggest challenge was to establish a single source of truth that was acceptable to all parties, including the research officers, who were required to provide annual statements of compliance for more than 7,500 awards. No policy existed at the time, and little help was available from the research office. The academic community was firmly committed to a policy of response on demand wherever possible. This involved, for negotiation purposes, details of grants and contracts that were not available in a straightforward retrieval scenario from existing systems.

4.2.Resource Constraints

Resource constraints: Constraints of resources surface in various ways across all activities in the data management program. Projects in the research life cycle that can potentially benefit from the outcomes of institution-wide data management services include teaching, research, and service. It may not be possible to match the short-term critical needs of a large number of research projects that would benefit from support, such as assisting with data management plan requirements for grant applications, with the available staff time. Library personnel are classified as civil servants and are not eligible for additional compensation resulting from the increase in workload. Consequently, library faculty and staff are required to accomplish all their work in a 40-hour week. At the library, all data services are provided with existing staff resources, a practice that is atypical in the field. During technology development, resource shortfalls can limit the scope, extent, and sophistication of the services that the library is able to offer. Skill development among faculty, the research community, and the library staff is similarly subject to the cap of education budget limitations. Furthermore, a lack of funding also constrains the library's ability to formulate, support, and promote activities and research within the institution. The acquisition of capabilities as a part of the data management program will be seriously affected due to a lack of monetary funding.

4.3.Data Silos and Fragmentation

Data is stored in numerous data silos across an institution and is often fragmented. With multiple silos, it is difficult to find the exact data that is required for a particular audience as information is spread out in many different places. These silos are not only organisational artefacts but may exist at the information system level. Three possible causes of data silos based on organisational structure are the organisational structure of the institution and responses to the environment, scenarios for innovation in business and IT functionality, and factors relating to the management of data or the perceived value afforded to data. Organisational structure could inhibit efficient data integration. Business and IT vision, along with management commitment, may not be aligned. Organisations that do not value the need for data integration might not be committed to providing data across the enterprise.

The causes for data silos can be divided into two types—functional and dysfunctional. While functional silos are necessary for efficiency, dysfunctional silos create barriers among different parts of the organisation, which impede internal communication and sharing of business intelligence for decision-making. It is argued that businesses need to balance the efficiency benefits derived from functional silos against the coordination cost of information sharing. It also highlights the importance of reconceptualising data integration strategies, which is that technology should not be focused too much on physical integration mechanisms.

4.4.Resistance to Change

People generally resist change. We, therefore, need to prepare people for change by providing an analysis of the problems, by frequent reportbacks by the people responsible for identifying, exploring, and instigating change, and by maintaining open lines of communication between those affected by the change process and the agents and instigators of change. Resistance to change was therefore noted as a sub-theme. Staff must be assisted through the process. Communication and reportbacks should be a priority activity at any stage of the project. Failure to report or communicate often resulted in confusion, stagnation, and uncertainty, which led to resistance to change. Staff must trust the process. Many processes would take unsolicited comments and use them effectively. Often, myths were debunked to the satisfaction of the person creating the myth.

Reports were made which confirmed or denied the myth, and in this way, change was accepted. Change should be small, should be perceived as small, and should be quickly demonstrated. Change was manageable, one step at a time, but the crucial motivational element was then towards larger or longer-term benefits. Data management as a project, however large or small it may appear to be, must itself be broken down into manageable and identifiable subprojects. Subprojects basically follow the principles for the management of a data management system: people must be involved in identifying and understanding the issues; the scope of each subproject must be defined and clarified; responsibility for each subproject must be identified; objectives must be set and reviewed; reporting of progress is of critical importance; and results must be communicated. Progress in implementing change will occur only when enough pain is felt from the existing situation.

Succeed or fail, change could be painful. Employees and management understood the potential painfulness of discussions about an integrated approach to data management. They wanted a smoother, less painful working environment, and therefore, the consensus was to develop a data management system to improve the management of data. While symptom relief provided a release of tension, it was anticipated that permanent relief should result from developing an understanding and, thus, developing a data management system. Change should be incremental – one step at a time, but change should be quick to demonstrate benefit.

4.5.Budgets Constraints Against Value Perception

It is a well-known fact that universities are under severe budget constraints . This is further aggravated by the overall reduction in financial resources due to a depressed global economy and other priorities. Despite numerous budgetary constraints, data-intensive research demands have become both more pressing and common, especially with advances in big data techniques and analytics. These factors suggest that investments in data management will be both necessary and required if research performance is not to be degraded. The problem that exists is that, at this stage, the perception that value is added in data management holds more theoretical than practical weight.

Budgetary constraints may impact the delivery and success of implementing data management programs. Data management expenses can vary widely depending on research intensity, discipline diversity, and research practice volume. The need is to develop best practices to mitigate costs without sacrificing the production of high-quality research data outputs and data management successes. Examples include quick guides, templates, training materials, and other educational materials that can be reused internally. A sustainable support environment with staff that has vested interests across all stakeholder levels potentially means more success at the coalface in implementing a data management program, all other things being equal.

5.Opportunities and Benefits of Effective Data Management

Area, location, and time-specific data series can contribute to process improvement, organisational modelling, and other applications in support of the institution's strategic and operational missions, such as the management of performance and accountability by management, funding bodies, and regulators. Many administrative processes generate data, and the more effectively and efficiently organisations can manage data, the lower their costs are likely to be, and the less weight is given to other performance drivers. With well-managed data, a quality and timely response to nearly any data request becomes more feasible. High-quality institutional data is also critical for the increasing number of external systems and reporting agencies that use this data to calculate critical organisational metrics or benchmarks that are essential to the continued funding success of institutions of higher education.

As a result, funding agencies and state governments usually dictate standards for most of this data. The national data are aggregated into a variety of federal databases. Responsiveness and attention to annual and periodic requests are non-negotiable, as these data collections are usually titled as statutory obligations. Getting data right the first time means that an institution will likely have fewer data corrections as well as less data tracking and reallocation of resources—especially important for smaller institutions or those that experience significant staff turnover. With increased competition for research, the ability to validate and verify research data ensures that every component of the research data lifecycle can pass federal regulations and, audits, other statutory obligations and that security is linked to the privacy of research subjects and the safety of the public.

5.1.Enhanced Decision-Making

Data analytics has a significant role to play in enhancing decision-making across universities. Institutional research, business intelligence, and data analytics are widespread across the sector, and they are used for strategic and operational decision-making as well as to meet reporting requirements. These needs continue to develop; however, data management processes must also be adjusted accordingly. Data quality and completeness remain critical to the accuracy of these outputs, and this supports a need for detailed discourse to occur and for the growth of cross-functional data management governance as higher education institutions continue to evolve and change, whether as a result of wider policy changes, financial challenges, or changes in student expectations, the need for analytics will also evolve. While the availability and use of data are becoming more institutionalised, this, in its own right, presents a number of challenges about how these will be used and the decisions that they will frame. Furthermore, as these evolve, this is going to put continued pressure on the institutional data management frameworks and what they need to deliver in terms of data collection, data quality, and data governance. Institutional research reports are one small part of demonstrating the impact that data has on this agenda item.

5.2.Improved Institutional Efficiency

Through a coordinated approach between a DMP and institutional strategic plans and initiatives, the DMP can help to reduce costs, increase financial efficiencies, and enhance the environment. By having accurate and transparent data usage, student research, employee finance, and facility management, an institution is better positioned to respond to changing circumstances. Additional savings and potential income enhancement can be used to help strengthen education, research, governance, accountability, and engagement. Enhanced analytics and evidence-based decision-making can help an institution align institutional data with its mission, goals, and strategies across levels and units to streamline operations and maximise existing and potential resources.

By adopting excellence in governance and accountability to achieve better public relations and engagement, in addition to improved economic value and development, an institution can help engage constituents and the community who support public and private institutions. After all, an institution’s employees, students, alumni, research, interface partners, and various communities have data at the heart of their interactions, shared and created data, and data that must be responsibly managed throughout academia. Data of varying exit status, value, sensitivity, and contractual obligations exist in multiple locations, most of which are not actively managed by the university. Existing institutional data management practices and policies are also not comprehensive. Actively managing university data is aimed at mitigating risks and enforcing laws, maintaining value, unlocking opportunities, and ensuring adequate controls are needed to improve efficiency, ease compliance concerns, and mitigate potential legal risks.

5.3.Support for Research and Innovation

Data management programs in research institutions seek to respond to the call for leadership and support for their researchers to take responsibility for the management of their data. The funding demands of both grant awardees and research programs all need appropriate levels of attention. Indeed, investment in the Netherlands will include the costs for digital repositories as selection criteria, given the recommendation above. However, not only should recipient institutions have an infrastructure in place, but their review board should also have a standardised set of data management practices against which they judge an institution's research practices. To aid institutions in setting up a data management program, data management plans have been prepared for several research disciplines, each with clear guidelines and case studies, making it attractive to use and freely accessible. To become a best practice, short versions of data management plans have also been developed as guidance for the assessment of data management plans. Useful tools in the form of case studies, including the results of a year-long program of open data sharing in the health industry, are also available. In recognition of the value of open data, mandates cover data, including associated metadata, that underpin the research results in scientific public research. In addition to these pilots, the Future and Emerging Technologies program will also use open data.

5.4.Compliance with Regulations

Compliance with regulations is an important aspect of research data management. Requirements come from various places and can include:

Funder requirements: Many funders already require researchers to prepare data management plans as part of their grant proposal, which includes information on how the data generated will be prepared for and made accessible to other researchers. Ethical and legal regulations for collecting and sharing data, including but not limited to sensitive personal data, human biological samples, and/or commercially sensitive research output. Gaining consent for use or sharing data, restricting or adequately securing certain data during storage and transmission. Funder and/or publisher requirements: Some journals aggregate data availability statements, only accepting a response with no restrictions. The purposes above are illustrative and not intended to be fully exhaustive. It is the responsibility of both researchers and institutions to be aware of and interpret the specific compliance requirements for their context. It is the responsibility of repositories to provide clear guidance on how to comply with these regulations where relevant.

6.Best Practices for Implementing a Data Management Programme

Keeping in mind the culture of the organisation and best practice principles, following a systematic and evidence-based approach while implementing a data management program is extremely important. Higher education institutions have finite resources, and ensuring that an enterprise-level data management program provides the best return on investment will require some critical forethought. The primary focus of this section, identifying best practices for developing, implementing, and assessing organisational data management programs, is informed by current data management literature, as well as good practice cases. It is recommended that, after reviewing these practices, higher education institutions choose a manageable number of the best practices initially and modify the list as needed based on their organisational culture. When implementing a data management program, an effective practice to ensure clear roles and responsibilities is to create a council at the enterprise level. This group should be programmatically placed to manage the data quality, data security, and business intelligence needs of the institution. To address the data requirements and capabilities of the entire institution, data management needs to be centralised. Successful data initiatives will help move decision-making to the right place using appropriate data-informed guidance. It is not just about the technical aspects that ensure the base architecture of data operations. It also requires promoting the availability of data to all employees. By sharing reporting assets and the corresponding systems, the cultural emphasis is shifted to a data-driven decision-making environment. Data security and data governance need strategic enforcement mechanisms, which is best achieved when devising a strategy and architecture for enterprise data management. Understanding organisational data needs and reaching an agreement on organisational data strategies will increase the overall value of data. With enterprise governance, data maturity is encouraged through matriculation and incentive plans. The incentive plans should focus on making an organisation's investment in a comprehensive data management strategy real and clearly not lip service.

6.1.Establishing Clear Policies and Procedures

Establishing clear policies and procedures is important in securing the delivery of a comprehensive data management service. Data management policies should set out the university's expectations regarding the management of research data and the responsibilities of the research community and should provide information on required standards. Data management policies or other related data policies or policies related to the protection of data may exist at the faculty or department level or higher. The existence of clear institutional policies, endorsed at senior levels of the university, will provide the authority to underpin important data management decisions, e.g., decisions regarding who has the power to enter into agreements with data repositories to maintain access to data and decisions about whether access to data should remain open or be restricted.

From experience, a one-size-fits-all approach may not be suitable, and thus, a continuum of policy development might be more appropriate, e.g., high-level policies from central administration to policies more closely related to specific disciplines. It is noted that policies endorsed by funders and publishers are also important in driving data management service provision; these should align with institutional policy to avoid conflict and allow for the efficient implementation of a data management program. In addition, the withdrawal of funding or reduction in publication opportunities for failure to meet required standards will not appropriately underscore the importance of good data management. The development of data management procedures should be led by the respective resource centre as part of the central research services or other specific data management services for research data management and education. These procedures should provide support to the research community in the delivery of their obligations as outlined in the policy statement and should incorporate best practices to ensure research compliance with an increasing number of funder and publisher requirements. The repository forms part of the procedural landscape, and its documentation should define deposit procedures and be continually revised to mitigate known data management risks.

6.2.Building a Collaborative Data Culture

Another aspect that must not be neglected in the development of an effective DMP is the creation of a collaborative data culture, where the different stakeholders accept the importance and requirements of proper data management and are motivated to work together to reach the proposed goals. This culture must build upon the existing one and adapt to the changes and challenges of a rapidly evolving environment. Additionally, it must foster an open and transparent approach to data management, encouraging the sharing of resources and information and the improvement and reutilization of existing data in solutions that explore collective intelligence. The development of a data culture is not an easy task, and progress will not happen overnight, but the benefits for the university far outweigh the development costs. The first step in the adoption of a data culture in higher education institutions concerns the definition of a strategic approach at all management levels, from university governance to library management. This top-down process assures that the institution's strategic data management goals are disseminated and understood throughout the organisation. However, the development of a collaborative data culture also involves a bottom-up process, with activities focused on the individual stakeholders, where they can develop the necessary skills and motivation to assume an active role in the implementation of the institutional DMP. This is an effective means of building ownership and may generate positive network externalities when people motivate peers to join in. Since the library has long been involved in collecting, organising, and preserving information within the institution, its promotion of DMP and data curation services is essential for the success of the initiative. Data stewardship and data management skills are indispensable in the digital age and must be developed, nurtured, and adapted to address the specific needs of different groups of stakeholders in various stages of the data lifecycle.

6.3.Investing in Data Management Tools and Technologies

Investing in data management tools and technologies often rises to the top of the list of activities to pursue in order to lay the foundation for data management. Conversations with the participants confirmed the importance and often the urgency felt for investing in these tools and technologies. Participants indicated that different skill sets are required for selecting and developing data management tools and collecting data tools and technologies. In addition, it was clear that the selection of different types of data tools and technologies required support from central resource providers. Participants also indicated that it was sometimes difficult to understand which tool would be appropriate for which context – which tool was scalable, which tool should be planned as a pilot, for example. They also expressed concerns that 'we need to get our house in order fast.' Some of the participants were not building or negotiating their business cases clearly; they were prioritising solutions that would provide quick results. Several critical success factors were combined with issues such as timing, control, and complexity. Participants freely acknowledged underestimating the effort required and failing to plan for change or overlooking some basic activities or the importance of maintaining a holistic view of data maintenance or the data demanded, built, and subsequently used. The governance capabilities of universities were also often not strong enough to support a mature data management program that would enable architectural development for long-term data capabilities. The higher education sector appeared to have room for improvement in terms of data architecture and the selection and capability of data management.

6.4.Providing Ongoing Training and Professional Development

Implementing a DMP requires considerable expertise in the requirements of data management and research data sharing and storage infrastructure. It is, therefore, essential that this expertise is made available to researchers in an ongoing and sustainable way. Research data management support will need to be professional, consistent, and reliable. Trained staff must be available to consult with researchers during the proposal-writing process. Implementation must cover all the topics included in typical DMP courses developed at many HEIs. The topics presented here may be included in courses for which some of the content is missing, but ideally, researchers should receive a comprehensive education in data management, as is the case in the best programs. Courses must also address the differences in data requirements between basic and applied research; for example, mandated researchers often partner with external stakeholders, so their data may be of a different type and format.

Data management support staff must be trained to educate researchers about the importance of data management and sharing early in the research process. Special emphasis must be placed on the importance of training regarding ownership, file format, standards, metadata, file naming, storage, and data sharing, including spatial data. The vital atypical functions of the data management support staff that are often underrepresented include database design and management, spatial data management and support, and training regarding data rights during the proposal preparation and execution phase. The last two, although equally as important as those in the Toolbox for the responsible sharing of digital data, are mentioned only in a few guides and courses designed by specific research data management support staff, and currently, they are not universally relevant to a wider academic audience. It would be beneficial if there were webinars and other resources and courses available and created based on the needs of those interested in starting new DMP services. The availability of these services is important for the future development of successful DMP programs in the idle stage as well as in those of the late adopter to advocate and support change within their units. It is essential to make the data management and sharing support offered to researchers user-friendly.

6.5.Choosing no-code Tools for Rapid Automation and Requirements Formulation

At a time when the need and demand for data management by higher education institutions are growing continuously, the involvement of no-code tools in automation and data management processes seems to address directly prolonged issues, such as the "skills" and "lack of staff" challenges relating to the data integration task, and concerns relating to data privacy.

Using no-code tools like Notion and Coda to merge many Excel sheets and CSV files. This allows start automating your data gradually. These tools enable rapid prototyping and iterative development of data workflows without requiring profound technical expertise.

??????????????? Combine essential data scattered in various solutions into interactive dashboards quickly using Looker Studio. This helps cut down on silos and makes your data visible in one place.

??????????????? Enable agile development by starting with small-scale automation projects whose scope will be expanded with increasing clarity to avoid overkill.

??????????????? The engagement of data management in aiding non-technical staff in the process can reduce the load on professionals and other data officers.

A major complication in data systems implementation is the lack of clear automation requirements from the outset.

??????????????? When there are no clear requirements, data professionals spend time analysing issues and iterating, which may be vague or may evolve. They can’t get on with their job.

??????????????? When a buyer has ambiguous needs, it’s hard to scope a data system properly. This may lead to inefficient system sizing and budgeting.

??????????????? Implementing projects in the wrong way will consume more time and money.

No-code tools help ease and solve this issue.

??????????????? The requirement refinement would be flexible as the stakeholders interact with and validate preliminary workflows in real-time.

??????????????? Testing on a small scale allows you to assess the practicality and resources needed before implementing more expensive and larger systems.

??????????????? Promote stakeholders’ engagement, facilitate collaboration between technical and non-technical staff and build a shared understanding of data requirements.

So long as universities opt for tools like Notion or Coda, they will be able to phase their automation according to their REAL needs. With this, they can expect to develop something scalable over a period of time. This strategy helps improve decision-making and prevents delays in implementing solutions.

7.Conclusion and Future Directions

This article has considered the challenges and opportunities for implementing a data management program within higher education. Universities are being driven by various social, economic, and political imperatives to be better stewards of their data, and this has created the conditions for researchers, administrative staff, and IT professionals to take on specialised data stewardship roles. Yet, the cultivation of novel approaches to data management within higher education is a complex process which calls into play the connections between local and international politics, institutional structures, professional practices, and individual motivations. Implementing a data management program is important to meet the requirements of the research excellence agenda and foster a wide range of data-driven innovations. It is hoped that this paper will assist those undertaking or contemplating similar initiatives elsewhere and foster collaboration between geographically dispersed projects. These data-driven innovations have the potential to transform higher education by supporting sophisticated management and governance systems, allowing institutions to manage access to resources, supporting informed decision-making, and enabling novel forms of collaboration. However, to achieve this potential, universities need to implement processes to manage their data assets, matching these policy initiatives with strategic data management processes and effective governance. While universities are starting to support researchers in complying with sponsor data management obligations, they require support of their own because there are substantial costs associated with implementing a data management program at an institutional level. Many of the associated factors are hindered by a climate of ambiguity about the future.