This is a summary of an article by James Henderson, published in issue no. 33 of CULTIVAR.
Read the article in full here.
Across global farms and forests, a quiet revolution is unfolding. Satellites scan canopies, sensors track soil moisture, drones map crop health, and digital platforms crunch vast datasets to offer real-time advice to farmers. This digitalization has generated an unprecedented volume of agricultural data, accelerated by recent advances in artificial intelligence (AI). Yet for all this technological progress, one fundamental question persists: Is this data helping to make better decisions and for whom?
The critical insight: data alone creates no value. Without effective data use, even the most comprehensive datasets remain inert. Data must be transformed through active use into useful information and into applicable knowledge that serves farmers, policymakers, and food systems alike. The gap between data collection and data use represents one of the most significant missed opportunities in agricultural transformation today.
The data paradox: Abundance amid scarcity
Despite the explosion of agricultural data globally, a troubling paradox exists: many countries, especially low- and middle-income countries (LMICs), still lack the most basic and up-to-date agricultural statistics needed for effective decision-making. The digital revolution has created an illusion of data abundance while masking critical gaps.
What's often missing is timely and accurate agricultural data at subnational levels. While we have access to petabytes of satellite imagery, turning that into actionable insights requires skills, context-specific data, and ground validation. Without this, we still struggle to answer critical questions: How many farmers are growing which crops, where? What yields are they achieving? What inputs are they using? What prices are they receiving? Satellite data can help fill many, but not all, of these gaps but only if paired with the right expertise and supporting systems. These data gaps are particularly severe in LMICs, where agricultural statistical systems have faced chronic underinvestment. Equally problematic is the erosion of trust in official agricultural statistics due to these very real shortcomings. Many potential data users from ministries and researchers to agribusinesses and farmer organizations express skepticism about the accuracy and timeliness of national census results. This leads to a vicious cycle: when agricultural censuses aren't trusted, stakeholders develop ad-hoc or parallel systems for decision making, further fragmenting the data landscape.
Fragmentation across national and international actors
Even when the will to improve agricultural data systems exists, fragmentation across both national institutions and international partners undermines progress. Ministries, national statistics offices, and external funders often operate in silos, each guided by different mandates, timelines, and accountability frameworks. This results in overlapping surveys, duplicated investments, and parallel systems that fail to communicate. In many countries, development partners launch uncoordinated data initiatives that crowd the landscape with dashboards, registries, and pilots, often without a long-term sustainability plan or connection to national priorities.
The platformization of agriculture: Risks, gaps, and the role of public systems
The agricultural sector is widely misunderstood. Many people still hold a romanticized view of farming, imagining pastoral landscapes, traditional practices, and a way of life rooted in community and land stewardship. For many producers, this remains true: farming is fundamentally a home and livelihood. For others, however, agriculture has evolved into a high-tech enterprise, managed through computers, sensors, and smartphones, integrated with global supply chains, and optimized through data analytics. Both are farmers, but the realities they navigate are vastly different.
Today, Big Tech and agribusinesses are developing integrated digital platforms that bundle advisory services, inputs, finance, and market access. These systems blur traditional boundaries and increasingly serve as agriculture’s organizing structure.
The intelligent use of data holds immense promise. It can reduce production costs, optimize resource use, predict harvests and price trends, and minimize losses. For many farmers, this can lead to greater profitability, competitiveness, and sustainability.
Yet the platformization of agriculture brings serious risks. As these systems consolidate, they tend to entrench the market dominance of incumbents, increase dependency on closed ecosystems, and contribute to fragmented, siloed data landscapes. Critically, these privately managed data systems often bypass national statistical systems altogether.
Global Data Power Dynamics
While national systems are where policy decisions are made, global frameworks play a critical role in setting priorities and shaping how resources are allocated. UN agencies such as FAO, IFAD, WFP, and the UN Statistical Commission provide essential guidance on food security, agriculture, and rural development, helping countries align with international standards and goals.
At the global level, FAO serves as the UN’s normative authority for agricultural statistics. It collects data from countries through standardized questionnaires and works to ensure international comparability across diverse contexts. In cases where data are missing or outdated, FAO may rely on publicly available sources or model-based estimates to maintain continuity in global datasets..
To improve timeliness and coverage, FAO is also expanding its use of non-traditional data sources, including satellite imagery, web scraping, and social media analytics. These innovative methods offer valuable new insights, especially where conventional data are scarce. However, they also raise important questions about validation, transparency, and accountability. Who determines what gets published? How are these new sources assessed for reliability? These are not just technical considerations, but political ones, tied to questions of representation and trust.
Three priorities for agricultural data governance
To unlock the full potential of agricultural data while avoiding the risk of deepening inequality, three priorities stand out:
- Reinforce data sovereignty and system resilience: National governments must be empowered to steward their own data ecosystems - hosting data locally, ensuring interoperability, and building infrastructure that can withstand political or funding shocks. This is essential for long-term sustainability and sovereignty over national decision-making processes.
- Shift toward strategic, domestic-led investment: Traditional donor-driven data projects are often short-term and narrowly scoped. Moving forward, investment models must prioritize long-term, nationally driven strategies. Countries should lead in defining their own priorities, supported by technical partners who can bring coherence, flexibility, and scale. Domestic political leadership should anchor both planning and financing efforts to ensure accountability and alignment with national development objectives.
- Reduce fragmentation and foster alignment across systems: Fragmented tools and uncoordinated investments weaken the public value of data. Strengthening foundational infrastructure, such as land registries, farm-level survey systems, and integrated dashboards, requires coordination among public agencies, development partners, and the private sector. This alignment is key to enabling innovation, ensuring relevance, and avoiding duplication.
Read the article in full here.
