---
title: "A Billion Dollar Future: Building China’s Agricultural New Infrastructure"
description: Horizon Data redefines Chinese agriculture by building robust data infrastructure, enabling banks insurers and government to make informed decisions
author: Darie Nani (Editor-in-Chief)
date: 2025-07-31T10:24:05.000Z
updated: 2026-02-26T18:02:15.006Z
canonical: https://www.sovereignmagazine.com/article/a-billion-dollar-future-building-china-s-agricultural-new-infrastructure
image: https://cdn.nanimediahouse.com/calvin-image-2.png
categories: Science & Tech
content_type: Spotlight
region: China
publication: Sovereign Magazine
---

When Calvin Cai left the mobile industry, few expected him to dedicate his next chapter to [farming](https://www.sovereignmagazine.com/article/from-science-to-sea-how-hatch-blue-picks-start-ups-that-might-actually-scale). Yet his latest venture is less about fields or farmers and more about the backbone that could support agricultural modernisation in China.

While headlines celebrate autonomous tractors and farming drones, a different kind of agricultural revolution is taking shape. This one happens in server rooms rather than fields, building the invisible data infrastructure that banks, insurers and governments need to make sense of Chinese agriculture. It’s the kind of foundational work that rarely gets attention but could reshape how food systems operate at national scale.

## From Mobile Growth to Data Infrastructure

Cai built his reputation scaling mobile distribution platforms across Southeast Asia, connecting app developers with advertisers through high-velocity systems. The transition to agriculture might seem unlikely, but his approach remains the same: focus on the systems that make everything else possible.

‘The future of food is a data problem before it’s a technology problem,’ Cai says. His Beijing-based company, Horizon Data Technology, maps over 16 million acres of Chinese farmland, but not for the reasons most people expect. Rather than selling directly to farmers, Horizon builds what Cai calls the ‘data rails’ – the underlying systems that other players can plug into.

This outsider perspective matters in Chinese agriculture, where [fragmented land ownership](https://www.csis.org/analysis/chinas-food-security-key-challenges-and-emerging-policy-responses) creates thousands of small family farms that complicate technological adoption. Climate volatility adds another layer of complexity, with droughts, extreme heat and floods destabilising production across regions.

## What Horizon Data Actually Does

Most AgTech companies compete on dashboards, sensors or autonomous equipment – the visible parts of modern farming. Horizon Data takes a different approach. Rather than building another farm app, they’re constructing the data foundation that sits beneath multiple applications.

The company combines satellite imaging, edge devices and institutional partnerships to create standardised profiles of agricultural regions. This isn’t about helping individual farmers check soil moisture. It’s about giving financial institutions, insurance providers and government agencies clean, structured data they can trust for major decisions.

‘You can’t optimise what you can’t measure,’ Cai explains. The company’s internal systems integrate weather patterns, machinery operation records, planting data and satellite observations to build what he describes as ‘live profiles’ of farming clusters.

## Why Agriculture Needs This Kind of Infrastructure

China’s agricultural sector suffers from problems that technology alone cannot solve. [Land fragmentation and dispersal](https://www.nature.com/articles/s41598-024-70605-1), particularly in hilly regions, obstructs effective coordination and modernisation efforts. Traditional recordkeeping systems struggle to track productivity across millions of small plots.

The Chinese government recognises these challenges. The [National Smart Agriculture Action Plan](https://www.dcz-china.org/2024/10/31/china-releases-smart-agriculture-action-plan/), launched in October 2024, commits to building a national agricultural big data platform by 2028. The plan emphasises standardised data systems and unified resource coding – exactly the kind of infrastructure work Horizon Data specialises in.

Standardisation matters because it enables trust. Banks need reliable data to assess agricultural loans. Insurance companies require accurate risk models for crop coverage. Government agencies must understand which regions need targeted support or subsidies. All of these decisions depend on having consistent, verified information about what’s actually happening in fields across the country.

## Competing on Credibility, Not Product Demos

While competitors focus on user interfaces and feature sets, Horizon Data competes on something harder to demonstrate: reliability. ‘We’re not trying to be the best dashboard – we’re trying to be the best system behind the dashboard,’ Cai says.

He draws parallels to how [payment infrastructure platforms](https://stripe.com/resources/more/what-is-fintech-a-guide-to-financial-technology) like Stripe revolutionised financial services. Most businesses never see Stripe’s interface, but they trust it to handle complex payment processing reliably. Similarly, Horizon’s customers don’t need to see the underlying data collection systems – they just need confidence that the information they receive is accurate and comprehensive.

This approach requires a different kind of patience. Backend infrastructure companies build trust through consistency over time rather than impressive demonstrations. The [Agricultural Bank of China](https://fbr.springeropen.com/articles/10.1186/s11782-021-00098-6) has already begun using digital technologies and big data for better credit operations and risk management, showing how financial institutions are embracing data-driven agricultural services.

## Building a Utility, Not a Product

Cai frames Horizon Data as a utility rather than a traditional software company. Like electricity or telecommunications infrastructure, the goal is to serve multiple stakeholders simultaneously. Banks use the data for loan assessments. Insurers develop better risk models. Government agencies identify where subsidies should be targeted. Agricultural corporations plan supply chain operations.

This multi-stakeholder approach reflects broader changes in how [rural finance and agricultural development](https://www.nature.com/articles/s41598-024-57091-1) intersect in China. Digital technologies are breaking geographical limitations and enabling more sophisticated risk assessment for agricultural lending.

The strategy comes with challenges. Agricultural data infrastructure is capital-intensive and politically complex. Building trust with government agencies, financial institutions and farmers requires different approaches for each group. The fragmented nature of Chinese agriculture means solutions must work across widely varying conditions and practices.

[Specialised agricultural data platforms](https://www.sovereignmagazine.com/article/indoor-agricultural-data-tools-drive-sector-resilience-amid-challenges) are proving their worth across different farming sectors by helping stakeholders navigate market volatility and investment decisions. Early traction suggests Horizon’s approach is working. The company has partnered with major Chinese agri-finance players, policy pilot zones and agri-industrial areas looking to digitise their risk models.

## The Long Term Prize

The opportunity extends beyond immediate revenue streams. ‘Whoever standardises agricultural data at scale doesn’t just make money – they shape how entire food systems function,’ Cai says. As climate change increases uncertainty around food production, [data becomes the bridge](https://www.sovereignmagazine.com/article/ai-for-the-paddock-why-algorithms-keep-stalling-before-reaching-australian-farms) between policy decisions, productivity improvements and resilience planning.

China’s [10 trillion yuan agricultural investment package](https://moderndiplomacy.eu/2025/01/29/chinas-agricultural-priorities-in-2025/) and focus on digital villages suggest the government recognises infrastructure’s importance. The National Smart Agriculture Action Plan aims to complete foundational digital data management systems by 2025, creating opportunities for companies that can deliver standardised, reliable information.

The work isn’t glamorous. Building data infrastructure requires meticulous attention to accuracy, consistency and scalability – [qualities that don’t make for exciting product launches](https://www.sovereignmagazine.com/article/pinpointing-progress-how-skylink-s-micro-positioning-tech-gets-saudi-infrastructure-moving). While [niche data platforms](https://www.sovereignmagazine.com/article/where-big-tech-doesn-t-bother-the-rise-of-profitable-niche-data-platforms) across other industries prove that specialisation and accuracy can build profitable businesses, [foundations matter precisely because they’re invisible](https://www.sovereignmagazine.com/article/the-regenerative-ai-factory-model-and-the-push-for-on-site-renewables).

Cai sees a future where Chinese agricultural decisions are made with the same data confidence that characterises other developed industries. Banks can assess agricultural loans as reliably as commercial real estate. Insurance companies can price crop coverage with actuarial precision. Government agencies can target support based on evidence rather than estimates.

That future depends on layer upon layer of careful infrastructure work. Cai’s job is to [build those layers](https://www.sovereignmagazine.com/article/trial-grants-on-farm-data-for-ag-input-makers-winners-and-runners-up-alike), one invisible system at a time.

[food security](https://www.sovereignmagazine.com/article/australia-s-agricultural-crisis-why-skills-training-could-save-the-nation-s-food-security) that banks, insurers and governments need to make sense of [Chinese agriculture](https://www.sovereignmagazine.com/article/precision-agriculture-takes-centre-stage-how-next-gen-gps-systems-are-transforming-modern-gra).

This one happens in server rooms rather than fields, building the invisible [data infrastructure](https://www.sovereignmagazine.com/article/can-data-models-really-change-filmmaking-lessons-from-the-front-line) that banks, insurers and governments need to make sense of Chinese agriculture.