Opportunities in Digital Agriculture: Investment and Entrepreneurship

Preface

Traditional farming puts people at the center of breeding, irrigation, fertilization, feeding, disease control, transport, and sales—relying on experience and craft. The result: low efficiency, high volatility, and uneven quality. It’s hard to solve today’s challenges by adding traditional inputs; we must shift decision-making from people to data and use digital agriculture to drive precise control.

Digital agriculture is a key part of the digital economy. With 5G and AI, many believe it’s entering a window of opportunity. The U.S. (Big Data R&D Plan), UK (Agritech Strategy), and Germany (Agriculture 4.0 Framework) all prioritize it. China’s policies—Digital Village Strategy, Next-Gen AI Plan, and the Digital Agriculture and Rural Development Plan—set quantitative targets.

Indicator	2018	2025	CAGR (%)	Type
Agriculture digital economy as % of value-add	7.3	15	10.8	Target
Online ag product retail as % of total	9.8	15	5.5	Target
Rural internet penetration (%)	38.4	70	10.5	Target

Source: Ministry of Agriculture and Rural Affairs (MARA), Digital Agriculture and Rural Development Plan (2019–2025)

Digital economy and industrial structure

The digital economy—popularized by Negroponte’s Being Digital (1994) and the G20 Hangzhou initiative (2016)—relies on digital tech, networks, and integration with the real economy, reshaping growth and governance.

Structure (by tech/production progression): foundation, core, application.

Foundation: systems (research projects, institutes, incubators, investors) and infrastructure (networks—broadband/NB/4G/5G; chips—EDA; software—OS, languages, DBs).

Core: industries of digital tech/products (six pillars):

• Software: software, big data, cybersecurity, cloud, e-commerce, AI
• Semiconductors: design, manufacturing
• Networks: internet, IoT, mobile, wireless, satellite, navigation
• Electronics: computers, phones, smart home, devices, auto electronics, wearables
• Digital content: film/TV, animation, online fiction, games
• Automation: sensors, instruments, control systems, robots, industrial software

Application: use of digital tech/products/methods in agriculture, industry, services.

How digital economy drives growth

Built on ICT, it connects products and services via the internet, enabling digital management of households, cities, and nations. In 2018, the U.S. digital industry hit $1.5T; China’s digital economy exceeded RMB 31T.

Year	Digital economy (RMB T)	% of GDP
2018	31	33
2017	27.2	32
2016	22.6	30
2015	18.6	10

Source: MARA, China Digital Village Report

Accenture: +10% digitalization → +0.5–0.62% GDP per capita. By 2025, digital economy may exceed 50% of global GDP.

What is digital agriculture?

Academic terms—smart, precision, facility, data agriculture—share common traits: build intelligent models from experiments/experience; collect real-time sensor data; analyze/predict via models for better decisions. Digital agriculture is the clearest: using digital data as a new production factor, integrating internet/IoT/cloud/big data/AI/smart equipment with modern agriculture to enable sensing, quantitative decisions, intelligent control, precise input, and personalized services—boosting chain efficiency and resource allocation.

Farmers will predict/prevent crop disease; view soil/crop status in near real time; automate irrigation/feeding. Sensors get smaller, smarter, cheaper; networks more capable and secure. The future is connectivity and data to maximize efficiency and output.

Digital economy in rural China

Infrastructure: Broadband China and universal service pilots pushed rural fiber to 96% of villages and 4G to 95% by 2018.

Rural internet users: 222 million (2018); penetration 38.4%, +3 pp YoY.

Digital demand: 290k village info centers built, 625k info agents trained; 71M public services, 222M convenience services; e-commerce turnover RMB 178B.

Public ag data: Systems for ag operators, resources, and major bio-resources; full-chain data for key products.

Planting digitization: National ag info dispatch platform; richer guidance; admin platform for planting data, tech release, input tracking.

Livestock digitization: “Scale Farming Cloud” and “Digital Dairy Cloud”; QR code tracing for vet drugs; animal ID and epidemic traceability.

Fishery digitization: Dynamic monitoring, digital gear pilots, national germplasm platform.

Seed digitization: China Seed Big Data Platform; world’s #2 crop germplasm DB; commercial breeding IT platform “Golden Seed”.

Equipment digitization: Beidou-based precision ops; integrated “sky-ground” monitoring; precision management; ag production public services. “Farm Machinery Express” platform/app; “Didi for farm machinery”; sensors on tractors/subsoilers; 90% info-based subsidy monitoring; unified equipment filing platform; online repair-skill learning.

Processing digitization: Databases for quality of bulk/ag-specific processing; monitoring/early-warning platform.

New operators: National registry for family farms (590k); monitoring for coop demonstrators (6,800+).

Rural e-commerce: E-Commerce Law (2019) promotes ag e-comm and poverty alleviation. 2018 rural e-comm: 9.8M firms; RMB 1.37T retail (+30.4%); ag e-comm RMB 554.2B (9.8% of ag trade).

Rural e-government: Innovation in environment clean-up, party building, digital governance, asset/land transparency.

Digital services: Digital preservation of intangible heritage, online education, telemedicine, inclusive finance.

Opportunities and challenges

Digital ag is the digitalization of biological/environmental factors, processes, and rural governance—a deep revolution. IoT/AI/Big Data/cloud drive change in production/life, reshape industries, and cement consensus on digital economy. Progress exists, especially e-comm, but contribution to ag value-add remains low.

Drivers

Environmental: Global warming and water stress—by 2050, major soy/corn regions may lose 18–23% yield; Africa may lose 15–30% by 2080–2100.

Demand: UN projects 9.1B people by 2050; food output must rise 70% (ex-biofuel). Grain to ~3B tons; meat to 470M tons. 80% of gains from yield/multiple cropping, 20% from more land.

Biofuels: Policies drive biofuels; demand for sugar/corn/oilseed feedstocks will keep rising, pressuring food prices.

Efficiency: Digital tech cuts costs and labor; optimizes seeds/fertilizer/pesticide/labor; lowers energy/fuel; balances time/resources for max yield.

Constraints

Fragmented market: Ag solutions are small and siloed; single-link tools struggle for scale. In developing countries, integrated service providers are scarce; growers buy point products.

Immature tech: High returns to ag R&D are proven, yet many countries underinvest. Digital ag faces global tech/application bottlenecks; weak innovation, lagging core tech, few ag sensors, low-fit robots/machinery.

Capex: Transforming farms for efficient, sustainable digital ecosystems needs big upfront spend. In China/Brazil/India, many farmers can’t justify it. China’s land-contract system dampens long-term investment appetite.

Data gaps: Data underpins digital tech; natural-resource and production data are missing. Government must push standards for collection and use to ensure adoption.

End-user hurdles: 60%+ of farmers cite low coverage and high cost as main issues; also long deployment cycles and bandwidth limits.

Opportunities

Smartphones + internet: More farmers rely on smartphones for info and knowledge-sharing; multi-language resources spread best practices. Ag agency sites (e.g., MARA) see 3M+ daily clicks.

Public-private partnerships: Governments/SOEs seek deeper ties with processors, finance, restaurants, e-comm, and tech firms to boost yield, safety, efficiency, and markets. Such programs provide advanced tech/management and new revenue; broad adoption will accelerate digital ag.

Tech reliance: Farmers lean on connectivity (LPWA, Zigbee, Wi-Fi) and wireless sensing to plan procurement, inventory, planting, harvest.

Digital ag value chain

Stakeholders: device makers, network providers, app providers, hosting/analytics, wireless connectivity, mobile operators, system integrators—together advancing digital ag.

Field sensors collect data; LPWA/mobile networks transmit; integrators/solution providers process and deliver via apps.

Commercial applications and cases

Precision agriculture

IoT + ICT to optimize yield and conserve resources; real-time data on field/soil/air to balance environment and profitability.

Variable rate tech (VRT)

Combines variable controllers with equipment to apply inputs at the right rate/time/place for each plot.

Smart irrigation

IoT-driven, measuring humidity, soil moisture, temp, light to calculate water needs and improve efficiency.

Ag drones

Use cases: crop health, acreage, variable-rate application, livestock management; low-cost monitoring and rich data via sensors.

Smart greenhouses

Monitor temp/humidity/light/soil; automate responses to keep optimal climate with minimal human intervention.

Yield monitoring

Track flow, moisture, total yield; inform decisions; cut costs and raise output.

Farm management systems (FMS)

Sensor and tracking data for collection/management; stored/analyzed for complex decisions; identify best practices and delivery models. Benefits: robust financial/production management; better risk mitigation for weather/shocks.

Hangzhou Yunhe Zhilian builds grower-centric services, merging ag tech and digital tech: crop/commercial insight–based plans; balance time/resources; lower cost; boost capability.

• Planting: reduce labor dependence via integrated environment monitoring, crop models, and precise control; IoT+AI for smart scheduling/monitoring/operations; differentiated production.
• Management: big data+AI for “digital” decisions—predictable/adjustable, with yield/quality focus.

Soil monitoring

Track and improve soil quality; prevent erosion, compaction, salinization, acidification, toxins. Monitor physical/chemical/biological indicators (texture, water holding, infiltration) to reduce risks.

Precision livestock

Real-time monitoring of reproduction, health, behavior to maximize profit; data-driven decisions to improve herd health.

Pest/disease detection

Timely monitoring and early warning are prerequisites for control. Traditional scouting is subjective/lagged. Modern tech is key across data capture, transmission, and processing.

Tech description

Artificial neural networks (e.g., BP) with backprop handle nonlinear pattern recognition/prediction. Input/hidden/output layers mimic biological nets. Nodes connect across layers; weights enable distributed processing.

Pest drivers (physical/environmental) interact nonlinearly—hard for traditional stats. Neural nets’ self-learning/adaptation/fault tolerance suit nonlinear problems.

Examples: Jin Ran et al. used BP to predict wheat aphid incidence (1980–2006 data) with 96.09% accuracy (2007–2011). Klem et al. used weather/soil temp to predict cabbage root weevil at 97%. Li Bo et al. used PCA spectra + probabilistic NN to identify rice nematode/leafroller at 95.65% accuracy.

Drawbacks

BP convergence is slow; hidden-layer sizing is ad hoc; gradient descent may trap in local minima—needs optimization/combination.

Examples: Zhang Fangqun et al. used PLS-GA-Elman hybrid NN for corn borer in Shaanxi (1988–2013); 5-year relative error 0.0661–0.1222%. Cao Zhiyong et al. used PSO-optimized hybrid NN for rice blast with max error <0.001.

NNs can pair with math models: Yang Shuxiang used SPSS stepwise regression + BP for larch caterpillar distribution/density; Wen Zhiyuan combined fuzzy logic+NN for navel orange pest ID; Tan Wenxue used deep learning with momentum for real-time fruit disease warning and diagnosis; multi-channel NNs detect cucumber mosaic virus; Kouakou used optical fingerprints + multi-channel NN for cucumber virus.

Monsanto + DataRobot built image recognition for pest/disease with 95.7% accuracy—faster and better than experts. Resson’s AI monitors pests/diseases and trends for alerts. Yunhe Zhilian uses UAV + multispectral imaging + image recognition to detect crop pests/health—30 minutes for 300 mu at 95% accuracy.

Breeding by design

Seed is strategic. Design breeding fuses genetics with bioinformatics, big data, and AI; molecular design is the main battleground.

Use genetic tests and variation data to quickly mine trait genes and predict phenotypes; use gene editing to create new traits (stress resistance, yield); use AI to design breeding schemes combining superior alleles.

Ag robotics

• Grafting robots: For crops like watermelon/tomato, grafting is key to avoid replant disease but labor-intensive and variable. Japan’s TGR Institute built a robot for cucurbit grafting that auto-detects suitable seedlings and skips defects; 98% success.
• Weeding robots: Weeds compete and host pests; herbicide overuse causes compaction/resistance. Vision-based robots segment soil/plants, separate crop/weed, locate targets, and mechanically remove weeds nonstop.
• Harvest robots: Manual fruit picking is costly, with labor shortages in peak season. Belgium’s OCTINION strawberry picker (Dribble platform) navigates greenhouses without retrofits, judges ripeness via vision, and picks in 3 seconds—on par with skilled workers.
• Autonomous tractors: Case upgraded Magnum T8 in 2016 to autonomous—works alone or with traditional machines; uses radar/LiDAR/cameras to detect/avoid obstacles; plans routes; remote control; auto-return for refuel/seeds.
• Seeding robots: Iowa inventor David Dorhout’s “Prospero” uses sensors for soil info, algorithms for optimal density, and autonomous seeding; multiple units can swarm for efficiency.
• Nondestructive testing: Image-processing–based QC of produce without damage—measure size/shape/color to grade quality.
• Plant factories: IoT captures greenhouse data; big data+AI control climate/nutrients for higher yield/quality, less labor, better economics. Essential for future off-planet habitats (e.g., Mars bases).
• Livestock: Cainthus (Canada) uses computer vision on farm cameras for cow face/body to gauge emotion/health. Netherlands’ Connecterra uses wearables + fixed sensors for health and estrus detection.

Digital ag innovators

Nano Ganesh (India): Water Pump Control 23—cellular remote control/alarm for pumps; addresses power fluctuation, harsh terrain, wildlife damage, danger, exposed wiring, shock, erosion. Rural water systems also struggle with tank/source coordination.

Benefits: saves 180k m³ water, 1080 MWh power, 180 m³ fuel, 18 m³ land; $720k labor saved annually.

Colombian telcos: Teléfonica/Movistar/Claro/Tigo support banana farm monitoring. Challenges: floods, low soil oxygen, high humidity, low temps.

Benefits: +15% yield; better environmental/ag sustainability; better traceability.

VinaFone (Vietnam): Operators (Viettel, MobiFone, VinaFone) support data from wireless sensors to platforms to cloud. A major fish farm stats before tech:

• Fry in tanks: 2,000 kg; harvest after 6 months: 30,000 kg; price $1.5/kg → revenue $45k.

Benefits: mortality cut 40–50%; harvest 42–45k kg in 6 months; revenue $63k–67.5k; savings $18k–22.5k.

Telefónica (Spain): Launched automatic irrigation via GPRS linking valves/meters/level gauges across >21k ha farms—manual control impossible. With ABB, built a remote irrigation system combining computers/phones for proper schedules using GPRS and remote meters.

Benefits: saves 47 hm³ water/year; farm profit +25%; power cost -30%.

References

1. CAICT: Digital Economy White Paper, 2017
2. Digital Economy Federation: Digital Economy Blue Book, 2019
3. FAO: How to Feed the World in 2050, 2009
4. Huawei: The Connected Farm, 2017

Published at: Nov 20, 2025 · Modified at: Jan 14, 2026