Digital agriculture systems can deliver timely, science-based information to farmers on topics such as weather forecasts, pests, soil chemistry, new seed varieties, and agricultural practices. These systems can also strengthen agricultural markets by linking farmers with input suppliers, traders, cooperatives, and other market actors. Evidence from multiple randomized trials shows that farmers adjust their behavior in response to digitally delivered agricultural information, and digital advisory services can improve market linkages and complement traditional agricultural extension systems. One meta-analysis of six SMS-based agricultural advisory programs estimated that benefits at scale would exceed costs by 46 to 1 (Fabregas et al. 2025).

Most smallholder farmers currently lack access to reliable agricultural information. Governments have traditionally relied on in-person extension services, but farmers often outnumber extension agents by more than 1000 to 1, limiting the reach of these programs (Fabregas et al. 2019). Surveys in India illustrate these gaps: only 6 percent of farmers reported receiving advice from an extension agent in the previous year (Cole and Sharma 2018), and many farmers report limited trust in extension recommendations (Eishman et al. 2016). As climate change alters rainfall patterns and pest pressures, timely information on crop choices, input use, and farming practices is becoming increasingly important for agricultural decision-making.

Digital agriculture systems can address these information gaps by delivering low-cost advisory services through mobile phones and digital platforms. Such systems can provide localized recommendations tailored to agro-climatic conditions, which is important because soil characteristics and input profitability vary widely across farms (Marenya and Barrett 2009; Suri 2011). Digital tools can also reduce the costs of experimentation and learning for farmers, who otherwise must rely on costly trial-and-error or limited peer networks for information (Hanna, Mullainathan, and Schwartzstein 2014; Chandrasekhar et al. 2022).

Advances in artificial intelligence could further expand the capabilities of digital agriculture systems. Machine learning tools and large language models can help customize advisory services and make them more accessible through voice interfaces, images, or conversational tools. For example, an experiment in India found that personalizing the timing of digital extension messages based on farmers’ characteristics increased message engagement by 2.6 percentage points, an 8 percent increase over baseline rates (Athey et al. 2023). As these technologies improve, digital agriculture platforms could increasingly deliver tailored advice at scale, providing farmers with actionable information to improve productivity and resilience to climate change.