FAQ

The FAIR Process Framework (FPF) is a six-step approach developed in collaboration with over 32 global partners. It offers a series of practical, bite-sized activities that guide donors and grantees in seamlessly embedding FAIR and responsible data practices from the initial concept through to project implementation.

Embedding FAIR principles ensures that data and resources generated by an investment are easy to find, access, and reuse. This not only increases the visibility of the investment and strengthens collaboration but also amplifies the impact of research outputs by fostering transparency and openness.

The FAIR Process Framework (FPF) is designed specifically to support Program Officers (POs) at the Bill & Melinda Gates Foundation (BMGF) and their grantees in developing effective data strategies that adhere to the principles of FAIR. By equipping BMGF projects with robust data management and sharing practices, the FPF enables greater impact and collaboration across initiatives focused on global health, agriculture, and beyond. However, the FPF extends its utility beyond BMGF stakeholders. The resources, templates, and guidance it provides are valuable to the wider agriculture and data communities. Researchers, policy makers, development organizations, and data stewards across sectors can also leverage these tools to enhance data transparency, improve knowledge sharing, and foster innovation. This dual-purpose design makes the FPF a powerful resource not only for BMGF-aligned projects but also for any organization committed to building a more connected and data-driven agriculture ecosystem.

The FAIR Process Framework (FPF) was developed through years of research, collaboration, and hands-on experience in addressing barriers to data sharing within national systems. From 2018 to 2019, CABI conducted in-depth research in select Gates priority countries to identify factors hindering effective data sharing. This research highlighted several non-technical challenges—such as policy, capacity, and resource limitations—that impact data sharing practices. Additionally, the study examined the roles, capacities, and capabilities of Program Officers (POs) in overcoming these challenges, focusing on ways to leverage or enhance their competencies to drive effective solutions. From further funding, CABI also explored  understanding the roles, capacities, and capabilities of Program Officers (POs) in addressing these barriers, as well as identifying ways to leverage or strengthen their competencies. 

Building on this foundation, in 2021, CABI introduced the Data Sharing Toolkit, a resource with modules designed to demystify how to collect, use, and share Findable, Accessible, Interoperable, and Reusable (FAIR) and safeguarded data. Drawing from the insights gained over the years, CABI has now developed the FAIR Process Framework—a structured, step-by-step approach that enables POs and grantees to address barriers to FAIR data early in an intervention. The FPF offers practical activities, templated resources, research-backed guidance, and three real-world use cases to support embedding FAIR data practices effectively. This comprehensive toolkit provides everything needed to implement responsible, FAIR-aligned data management strategies in agricultural development initiatives.

The time required to implement the FAIR Process Framework depends on the investment’s size, complexity, and its current alignment with FAIR principles. For smaller or more focused projects, implementation may take less time, while larger or more complex investments may require a phased approach to fully integrate FAIR practices.

The FAIR Process Framework stands out because it was developed specifically with the needs of donors, program officers, and grantees in mind, especially those involved in agricultural development. The FAIR Process Framework emphasizes a collaborative approach, allowing stakeholders to adapt FAIR practices progressively throughout the project lifecycle. It includes tailored resources, templates, and real-world use cases that make it easier for projects to align with FAIR principles regardless of their starting point. This framework is also designed as a living document, meaning it can be continually updated to reflect evolving needs and new insights, making it highly adaptable across diverse project contexts.

Unlike other FAIR frameworks, which may focus primarily on technical solutions, the FAIR Process Framework provides a step-by-step, practical approach that addresses both technical and non-technical challenges—such as policy, capacity, and resource limitations.

In the context of Bill & Melinda Gates Foundation AgDev grants, Program Officers (POs) are responsible for initiating grant-making processes that incorporate the FAIR Process Framework. This includes using tools like the Data Access Checkbox in INVEST, developing Data Management and Access Plans, and ensuring data-sharing and management requirements are embedded in grant agreements. POs can refer to the Job Aid: Foundation Grant-Making Tools to Enable Data Access in AgDev for guidance on implementing these practices.

Grantees are responsible for utilizing relevant resources within the FAIR Process Framework to support FAIR-aligned data practices throughout their projects. This includes managing data-sharing agreements, ensuring data cataloging, and meeting data access milestones. As FAIR principles are community-driven, everyone involved has a role in improving data quality, metadata standards, ethical data use, and reusability. This collaborative approach builds trust, promotes traceability, and fosters a culture of responsible data stewardship among all data users and contributors.

While activities in the FAIR Process Framework are in presented in a sequential order of steps, you can absolutely integrate elements from the FAIR Process Framework, even if your project is already underway.  The framework is designed to be flexible and can be applied at various stages of a project’s lifecycle. By incorporating specific FAIR practices now, you can enhance data quality, improve accessibility, and align with best practices for data stewardship. 

For instance, you might start by focusing on enhancing metadata standards, setting up data-sharing agreements, or improving interoperability with other data sources. Implementing these elements can still add significant value by making your data assets more discoverable, reusable, and better aligned with FAIR principles, ultimately maximizing the impact of your project. They are not fixed and elements from the steps can be used when your project is ongoing. But to maximize the potential of the framework in your project, following the linear approach will return maximum benefit. 

The FAIR Process Framework offers structured, step-by-step guidance, templates, and resources to help teams new to FAIR. Additionally, CABI is available to provide support as a sub-awardee for training and implementation assistance.

Yes, CABI is available to offer further support and can be engaged through future contracts as a sub-awardee to assist with implementation and other Framework-related needs. 

FAIR data is often mistakenly equated with open data, but the two are not synonymous.   Data can adhere to FAIR principles while still having restricted access and reuse. For instance, business assets can be made FAIR by ensuring they are discoverable, accessible within the organization, interoperable with various systems, and reusable with defined agreements or licenses.  Applying FAIR principles to proprietary data involves: 

• Implementing robust metadata structures to make assets easily searchable and discoverable. 

• Establishing clear access controls and permissions to safeguard sensitive information while allowing appropriate access. 

• Adopting standardized formats and protocols to ensure compatibility across different systems. 

• Comprehensive documentation to facilitate efficient reuse and repurposing, with clear agreements in place. 

By following these practices, businesses can optimize asset management, foster collaboration, enhance decision-making, and maximize the value derived from their assets throughout their lifecycle. 

The FAIR Process Framework can aid in understanding the enabling environment that can influence your project, highlight the stakeholders vital to your project, the power dynamic/relationships that can enable/disable your project, and an appreciation of culture and context.   

Even though the full benefits of implementing FAIR principles may take time, the FAIR Process Framework offers some quick, valuable gains: 

• Understanding the Enabling Environment. The FAIR Process Framework aids in analyzing the broader environment influencing your project, identifying key stakeholders, understanding the power dynamics and relationships that could impact project success, and appreciating cultural and contextual factors. 
• Improved Data Organization. Early adoption of the Framework’s guidance on metadata standards and data cataloging practices helps create a structured, organized dataset, making it easier for teams to locate and access data quickly. 
• Enhanced Collaboration. By clarifying data-sharing expectations and responsibilities through tools like Data Sharing Agreements, the Framework facilitates smoother collaboration among project stakeholders, reducing delays due to misunderstandings or access issues. 
• Risk Reduction. Implementing responsible data practices from the start helps identify and address potential risks around data privacy, security, and compliance, reducing the likelihood of data-related issues later. 
• Better Decision-Making. With clear, standardized data practices in place, teams can make more informed decisions based on reliable, easily accessible data. 
• Capacity Building. The framework provides templates, guidance, and structured steps that build capacity among team members, enhancing data management skills and establishing a foundation for long-term FAIR practices. 

These quick gains set the stage for greater data accessibility, reusability, and impact as the project progresses, offering tangible benefits early in the process. 

The FAIR Process Framework demystifies FAIR implementation by breaking down complex concepts into simple, everyday language, making it accessible to a wide range of users. Its step-by-step activities, practical exercises, and real-world case studies offer clear guidance, helping users understand and apply FAIR principles in a straightforward way. By focusing on practical, incremental actions, the framework provides insight into manageable approaches for FAIR implementation, reducing the perceived complexity and enabling users to adopt FAIR practices with confidence.

The FAIR Process Framework includes steps to ensure data practices are aligned with relevant data privacy laws and regulations, such as GDPR. It provides guidance on securing permissions, handling data ethically, and meeting compliance requirements in different jurisdictions. 

A data governance policy under the FAIR Process Framework includes: 

• Purpose and Scope. Defines objectives and outlines the types of data covered. 

• Roles and Responsibilities. Identifies roles (e.g., data stewards, owners) and clarifies responsibilities. 

• Data Management Standards. Sets standards for data quality, security, and lifecycle management. 

• Data Access and Sharing Policies. Establishes rules for access, sharing, and ethical use of data. 

• Compliance and Accountability. Ensures alignment with regulations and designates accountability for policy adherence. 

These components create a structured approach to managing data responsibly and in alignment with FAIR principles.

A DMAP should outline data collection methods, storage solutions, access and sharing policies, data security measures, metadata standards, and plans for data preservation and archiving. It ensures that data is managed responsibly and remains accessible and usable over the long term. Step 5

The DMAP is intended to be a living document, meaning it should be revisited at key project milestones or when significant changes occur. Regular updates ensure that the data management strategy remains relevant and aligned with project needs.

Data management refers to the day-to-day activities of collecting, storing, and maintaining data. In contrast, data governance involves setting the policies, standards, and roles that guide these data management activities, ensuring data quality, security, and compliance over time.  Step 5

Best practices for creating a data inventory include identifying and cataloging all datasets, assigning ownership, documenting metadata, setting update schedules, and ensuring accessibility within the organization. This inventory acts as a comprehensive resource for managing and locating data assets. Step 3

The choice between a Data Sharing Agreement (DSA) and a license depends on the level of control and specificity you need for data use. 

• Data Sharing Agreement: A DSA is a legally binding document typically used when sharing data with specific parties under agreed-upon terms. It is suitable when you need detailed control over how the data is accessed, used, shared, and protected, and when there are privacy or confidentiality concerns. DSAs often outline responsibilities, data usage limits, and any necessary compliance with data protection regulations. 

• License: A license is generally used for broader data sharing with a wider audience, often under standardized terms. It defines permissions and restrictions on data use but is less restrictive than a DSA. A license is suitable when you want to make data available to the public or a large group with clear usage. 

Data sharing agreements (DSAs) are not one-size-fits-all documents; they should be tailored to fit the specific needs and requirements of the project, the type of data involved, and the roles of the various parties. Each DSA typically outlines data-sharing terms, responsibilities, and any access restrictions or licensing requirements, ensuring that all parties are clear on how data will be used, managed, and protected.

The responsibility for drafting a DSA usually falls to the grantee, with guidance and oversight from the Program Officer (PO). Grantees can refer to Step 5.3 within the FAIR Process Framework to ensure all necessary aspects—such as access permissions, usage limitations, and data stewardship responsibilities—are appropriately addressed. The Grantee should adapt the agreement to their specific data-sharing context. By using a customized DSA, all parties involved can ensure clear expectations, foster trust, and promote responsible data sharing practices that align with both FAIR principles and proprietary needs.

Metadata provides context about data—such as its source, structure, and intended use—making it easier for others to find, understand, and reuse the data. Good metadata enhances data discoverability, supports interoperability, and ensures that data can be accurately interpreted by others. Step 3

A FAIR Enabling Resource (or FER) is described as ‘any digital object that provides a function needed to achieve some aspect of “FAIRness” and is explicitly linked to one or more FAIR Principles’.

The FIP is a collection of FAIR implementation choices made by a community of practice for each of the FAIR Principles. Community specific FAIR Implementation Profiles are themselves captured as FAIR datasets and are made openly available to other communities for reuse. (reference: GO FAIR website).

Metadata improves data sharing by providing essential context, such as the source, structure, and purpose of the data. This additional information makes data easier to find, understand, and reuse, allowing researchers and stakeholders to quickly assess the relevance of a dataset and utilize it effectively.

Machine-actionable metadata allows computers to process and interpret data automatically, which is crucial for efficient data management and analysis. In FAIR data practices, machine-actionable metadata helps streamline data retrieval, integration, and analysis by making it easier for software systems to understand and use the data without human intervention.

A DOI is a permanent, unique identifier for digital content, such as datasets or publications. By providing a stable link, DOIs make it easier to find, cite, and access digital objects reliably over time. In FAIR data practices, using DOIs enhances data accessibility, ensuring users can consistently locate and reference specific datasets.

The value of FAIR principles in AI lies in promoting transparency, accountability, and trustworthiness in AI systems. By ensuring that AI artifacts—such as models, datasets, and results—are findable, accessible, interoperable, and reusable, stakeholders can more easily locate, interpret, and validate these resources. This transparency fosters collaboration, reproducibility, and knowledge sharing within the AI community. Moreover, applying FAIR principles supports responsible AI development by encouraging ethically sourced datasets, interpretable models, and efforts to mitigate bias. Ultimately, FAIR principles help create AI systems that are more reliable, equitable, and aligned with societal values, contributing to the development of AI that serves the public good. 

Data ecosystem mapping helps visualize the flow and connections between different data sources, stakeholders, and tools within a project or organization. It provides a clear picture of where data is created, how it is used, and who manages it, helping teams identify potential gaps, streamline processes, and improve data sharing.