Level 5 "FAIRest of them all"
- Giovanni Nisato
Level | Nickname | Marketplace metaphor | Features | Picture |
5 | "FAIRest of them all" | ""Digital Online Store"
| It is aspirational and conceivable, but it still needs to be realized. Evidence exists of cross-organization standards and Interoperability. |  |
Table level 5 summary
Level 5 - Capabilities
Additionally to level 4 capabilities FAIR data principles are implemented across all key functions and organization domains. Data creation remains FAIR throughout its lifecycle, facilitated by scalable tools. Data can be exchanged and is interoperable across organizations in the life-science ecosystem such as pharmaceutical companies, CROs, regulatory bodies, and scientific data providers.
Machine actionability and automated operations are possible, leveraging self-describing digital objects. Knowledge-enabled citizens can utilize data products and robust analytical services, including AI linking provenance to machine-interpretable data. FAIR processes are integral to the business value generation, sustaining value creation throughout the data lifecycle. The maintenance costs of FAIR tools and infrastructure can be minimized and shared in the ecosystem, through community-established standards for data, metadata, models, and FAIR implementation profiles.
Level 5 - Business value
Ability to exchange and augment data sets across organizations. Cross-organizational insight generation. Machine actionability. Increased cost-effectiveness of FAIR infrastructure investments (cost-sharing across organizations). Secondary reuse of data (e.g., clinical trials). Real-world evidence data-mining and improved patient outcomes. Enhanced drug development efficacy (e.g. drug repurposing) and reduced time-to-market.
Truly data centric ecosystem. The fostering of a Data-Driven culture enabled through all the FAIR Principles inside and across organizations.
Level 5 - Questions to ask
Does the data and metadata make strict use of ontologies and vocabularies that are themselves, FAIR
Does the metadata for the data contain links that resolve to different data sources?
Does the data and metadata meet the standards required by the community or a particular application?
What are the mechanisms to maintain cross-company references, for example controlled vocabularies?
Can you elaborate on how data is maintained as FAIR throughout its lifecycle, (from collection to consumption and re-use).
How does C-level leadership oversee and hold themselves accountable for FAIR data strategies?
Can you describe the links between FAIR data practices and tangible business outcomes within your organization?
Can you describe the range of functional roles defined, recognized, and empowered within your organization?
Can you describe the world-class training programs established by your organization for FAIR data?
How does your organization actively share its FAIR learnings internally and externally through
communities of practice?
Can you describe the collaborative effort among various roles to ensure data is machine-actionable and compliant with industry standards?
What efforts does your organization make to build common and shareable FAIR data practices and platforms within the industry?
Level 5 - FAIR data
All organization functions and units adhere to FAIR data principles as much as relevant, emphasizing data Findability and Accessibility. The need for data to be Interoperable and Reusable across internal and external data standards, potentially driven by regulatory compliance, is fulfilled.
The norm is for data creation to remain FAIR throughout its lifecycle, including during processing via pipelines and APIs. Connecting these principles with scalable, fit-for-purpose tools ensures efficient data management, alignment with regulatory requirements, and ease of access, sharing, and use.
Several companies exist at this level, and some data is "FAIR" across organizations within a given ecosystem, including pharma, CRO, regulatory bodies, and consultancies.
Business concepts aggregate data, and users can navigate between concepts seamlessly.
Machine actionability and automated operations are possible. These operations may include machine-enabled AI and semantic solutions that can act directly on data sets without the need for interpretation, thanks to the availability of self-describing digital objects.
In addition to the features described in Level 4, we fully describe the data using a knowledge language or ontology of the types that carry GUPRIs (Role: Knowledge-enabled citizens).
FAIR data are a key strategic asset and advantage of the company. Based on expected business value, critical data assets comply with the FAIR data principle, and new data generated is "FAIR by design.
(Back to the FAIR Matrix).
Level 5 - FAIR leadership
The C-level leadership is pivotal in overseeing and accountable for FAIR data strategies. There is full buy-in by leadership from general management, business and IT roles alike.
The leadership demonstrates a profound awareness of the transformative impact of FAIR data on the organization.
The organization emerges as a notable leader in FAIRification, showcasing a tangible return on investment, setting industry standards, and influencing successful FAIR data patterns. It features a well-defined vision, actively engages leadership across the organizational hierarchy, empowers financially and culturally, and ensures enduring commitment to FAIR principles.
Precise alignment with FAIR practices underpins the organization’s approach to operational metrics, extending this perspective horizontally both within and outside the company.
Leadership transcends mere virtue signalling, opting for thought leadership by exemplifying FAIR principles and breaking down data silos. The organization enjoys international recognition, attracting collaborations leveraging FAIR data principle applications for mutual benefits.
(Back to the FAIR Matrix).
Level 5 - FAIR strategy
A company-wide vision and plan is supported and known and supported by all employees. FAIR is the norm, and everyone understands why FAIR data is needed. The company is also recognized for its FAIR data work and is a data-savvy company.
FAIR data implementation strategy presents several crucial elements. There are clear links between FAIR data practices and tangible business outcomes, including cost savings, efficiency gains, and the identification of, e.g. business or IPR opportunities.
Establishing Metrics and methodologies to measure these outcomes reinforces the strategy's connection to essential tools. Importantly, the organization comprehends the impact of FAIRness on its financial performance, prioritizing it as a part of its overarching business strategy, supported by empirical cost-benefit analyses.
A high emphasis on Data integration facilitates the discovery of synergies by linking data across various business functions.
The FAIR strategy is characterized by transparency, longevity, and effective measurement through metrics and feedback loops, aligning it with the organization's strategic imperatives.
Outreach efforts to diverse stakeholders, including pharmaceutical companies, CROs, regulatory bodies, and academia, are also evident when relevant: strategy embodiments are externally visible.
(Back to the FAIR Matrix).
Level 5 - FAIR roles
FAIR is the norm, and everyone understands why FAIR data is needed. Everyone who needs to have adequate skills has them. Everyone new also gets FAIR data training. FAIR data is a competence of all employees. There is recognition for the FAIR data work.
Knowledge enabled citizens are empowered with access to information enabling them to contribute effectively to the organization's goals and objectives by utilizing FAIR-enabled resources.
A range of functional roles are defined, recognized and empowered. These roles include data standard experts, data curators, semantics experts, knowledge engineers, data strategists, AI experts, data miners, communication experts, and various regulatory and decision-making positions.
Developing a structured and well-conceived role framework is emphasized, guided by experts versed in FAIR principles, such as a " FAIR" inter-organization architect or a bridge between the Chief Technology Officer and Chief Strategy Officer. The passage also underscores the need for strategic roles in promoting cultural change and maintaining FAIR practices while nurturing newcomers and recognizing the champions who initiated the FAIR agenda. Other companies ask FAIR Champions to share their best practices.
(Back to the FAIR Matrix).
Level 5 - FAIR processes
FAIR processes are integral across all data processes, ensuring proven value throughout the entire data lifecycle.
World-class training programs are established and disseminated through courses. Organizations employ adaptable FAIR data frameworks that provide common standards, facilitating seamless FAIR data exchange among entities via fit-for-purpose FAIR-enabled APIs or tools.
The organization actively shares its learnings internally and externally, among others, through communities of practices. Processes are in a position to align data harmonisation efforts beyond the organization, such as feedback loops with standard-setting bodies. Mechanisms exist to sustain metadata across organizations, including update and feedback processes between data vendors and consumers.
Organizational practices embed FAIR data principles compliance, with a commitment to maintaining this compliance in the future. This holistic approach ensures the embedding of FAIR principles in every facet of data management and exchange within and outside the organization.
A standard (organization-level) process for onboarding data/technology projects and the relevant resources/people (e.g., FAIR training at project start-up) exists.
There are planned processes to maintain FAIR implementation standards and tools (e.g., CV, ontologies) at the company and across data ecosystem partners.
Regulatory agencies (FDA, EMEA, etc.) are incentivizing the ecosystem with guidance concerning FAIR data in their regulatory process.
(Back to the FAIR Matrix).
Level 5 - FAIR knowledge
There is a company-wide clear understanding of FAIR data. The company has a "FAIR data academy" (or similar education program) with FAIR data materials: training, best practices, use cases, showcases, a calendar with events, and trainers.
Data contracts are in place, which involve a collaborative effort among data experts, curators, semantic web specialists, knowledge engineers, AI professionals, data strategists, regulators, procurement officers, and decision-makers. Their roles ensure data is machine-actionable, adhering to community-established metadata standards extending beyond an organization's boundaries. These experts may possess knowledge in employing, e.g., LLM and AI techniques. Furthermore, they understand regulatory requirements, enabling them to express them in machine-actionable profiles. This comprehensive approach ensures data management compliance with industry standards and legal regulations, promoting data interoperability and usability beyond organizational confines. Across organizations, there is management and knowledge of FAIR standards.
An industry-wide, cross-organization FAIR community of practice is established, with distributed leadership. Regular cadence of meetings with a standing agenda of project reviews, best practice development, real-world experiences, and sharing of learnings.
The FAIR community is actively driving a FAIR learning system approach to standards development, data governance, data stewardship, data product management, and other topics as a natural outcome of the community effort.
Regulatory agencies are part of the FAIR practice community and incentivise the ecosystem with guidance.
(Back to the FAIR Matrix).
Level 5 - FAIR tools and infrastructures
In this scenario, the design of platforms and suites of tools facilitate the automated creation of FAIR data while ensuring provenance capture and enabling FAIR data to remain such, thanks to appropriate "FAIR" processes.
Defined interaction mechanisms provide standardized interfaces serving FAIR data conformant to the (industry-wide) community-agreed metadata profiles for various consumption contexts. Furthermore, the organization is a leader in issuing recommendations for FAIR data implementation profiles.
The platforms may encompass various components, including:
GUPRI (Identifiers): The system supports using globally unique, persistent identifiers for data, both internally and externally, enhancing data discoverability.
FAIR Metadata Publication: The platform supports the publication of FAIR metadata internally and externally, with considerations for data access restrictions.
Authentication and Authorization Infrastructure (AAI): This ensures permissible accessibility for individuals and machines, aligning with industry standards.
Data Interaction Interfaces: The provision of interfaces to enable data usage, including analytics and data interaction methods, ensuring transparency and adherence to FAIR principles exist.
Data Exploration: The system offers tools to explore data with machines capable of interpreting metadata, which can also involve anonymized or aggregated data for tasks like cohorting and feasibility assessments.
Semantic Model Management: It manages semantic models and ontologies, including external and in-house ontologies/controlled terminologies.
FAIR Data Validation: Tools are in place to validate the FAIRness of data throughout its lifecycle.
Community Leader/Beacons: Additionally, the organization actively builds common and shareable FAIR data practices and platforms pre-competitive within the industry, encouraging other companies and vendors to follow suit. The platforms provide user-friendly tools for achieving comprehensive FAIR data management with robust documentation and subject matter expertise (SME) support. Aspirationally, an automated FAIRification tool, possibly utilising AI, is envisaged for all data types. Implementing FAIR API and protocols is considered for connecting with various data sources, such as HR and legal data, following cross-company standards. "Federation tools'' may enhance data sharing and collaboration across organizations.
(Back to the FAIR Matrix).
Findability:
Tool(s) or infrastructure component which contributes, enhances or enriches Findability:
Accessibility:
Tool(s) or infrastructure component(s) which contributes, enhances or enriches Accessibility.
Interoperability:
Tool(s) or infrastructure component(s) which contributes, enhances or enriches Interoperability.
Reusability:
Tool(s) or infrastructure component(s) which contributes, enhances or enriches Reusability.