XAG (极飞科技) — the Chinese agri-drone maker with regulated financial disclosure, and the numbers that qualify the sector story

East-Asia (China primary); export to the Americas, Eastern Europe and Southeast Asia

Content

XAG is China’s second-largest agricultural drone maker and the only Chinese agri-drone company with regulated financial disclosure — it refiled for a Hong Kong listing on 26 March 2026, having withdrawn a Shanghai STAR Market application in 2022 and first filed in Hong Kong on 25 September 2025. That makes its prospectus the corpus’s best hard-numbers source for Chinese agricultural drone economics.

Financials (prospectus, FY2023–FY2025).

The overseas gross margin is nearly three times the domestic one: export is XAG’s profit centre, not its volume centre. Unit growth without revenue growth (units +22.76%, revenue +9.23%) means falling average selling price — consistent with the price competition visible across the Chinese drone market.

Product portfolio — no longer only drones. Four segments: agricultural UAVs (P-series: P200, P150 Ultra, P120 in 2026; P150 Pro, P60 Pro in 2025); agricultural UGVs (R100/R200 from 2025, RM80); APC-series autopilot retrofit kits for tractors, transplanters and harvesters; and smart-farm IoT (cameras, valves, fertigation). The XA1 dock (10 kW/18 kW mains supply) enables unattended autonomous operation. XAG’s own positioning has shifted from “agricultural drone company” to “agricultural robotics company”. Note the UGV line: 584 units in FY2025 against 26,296 drones — ground robotics is a real but small and very early line.

The penetration figure that matters. Co-founder Gong Jiaqin (龚槚钦) states that only around 8% of Chinese farmland has drones attached (up from 3–5%) — more than 90% of Chinese farmland is unserved. This is the most useful single number in the unit, because it comes from a company with an incentive to overstate, and it undercuts the “China is drone-saturated” reading. XAG claims 1,000+ Chinese counties covered and channel reach in about 70 countries.

Market share — and the correction it forces. The market-share figures circulating for XAG (global agri-drone 17.1%, China agri-robotics 18%, China agri-drone 20.8%) come from a Frost & Sullivan study commissioned by XAG and reproduced in the prospectus; the same source puts DJI at 36.6% of global agri-robotics and ~59% of global agri-drones. The corpus’s previous claim that XAG accounts for “more than half of Chinese agricultural drone sales” is contradicted by XAG’s own commissioned data (20.8%) and should not be used. DJI publishes no Chinese market share at all, so no independent Chinese share figure exists in either direction.

Deployment model — the service-provider pattern, in three verified modes.

1. OEM-founded service organisation — XAG Agricultural Services, running since 2015.
2. Purchase-and-join — an individual buys a drone from a dealer and joins a plant-protection team (植保队) or flying-defence alliance (飞防联盟) that handles training and dispatch, bidding for government coordinated pest-control contracts (统防统治) and cross-region work.
3. Cooperative in-housing — e.g. Nanjing’s Liyou machinery-service cooperative runs a drone team serving 120 farmers over ~6,500 mu.
4. Emerging: app-based dispatch (“flying-defence dispatch system”), farm-booking on a ride-hailing model.

Founder narrative (Peng Bin, via SCMP 2018). The origin story — observing unprotected elderly farmers spraying in Xinjiang in 2012 — frames the technology as labour-saving and farmer-safety first: drone spraying limits farmers’ chemical exposure and does the job in a fraction of the time. This remains the corpus’s clearest Chinese instance of worker conditions as the named primary purpose.

What this unit is doing in the taxonomy

Anchors the robotics — aerial × on-farm-production-open-field × worker-conditions cell with the corpus’s only tier-2 (regulated disclosure) Chinese agritech financial dataset. It is the evidential counterweight to the DJI unit: same market, same country, but with numbers that a regulator has read.

Distinct from:

• dji-agriculture-global-export.md — the volume leader, whose environmental and scale claims are unaudited and internally inconsistent.
• naio-technologies.md — French ground robotics, direct sale; XAG’s drone line sells through service providers and its new UGV line is sold into its existing channel.
• john-deere-see-and-spray.md — US equipment-integration model; XAG retrofits autopilot kits rather than selling integrated machines.
• china-smart-agriculture-action-plan-2024-2028.md — the state framework; XAG is a commercial actor inside the subsidy regime, not a state instrument.

Why it matters for talks

• The 8% farmland penetration figure reframes the Chinese drone story. The sector’s own co-founder says more than nine-tenths of Chinese farmland has no drone. Use this before any claim about Chinese drone saturation.
• Export margins are double domestic margins. The best evidence in the corpus that Chinese agritech exports for profit, not for market share, and that the export push is financial rather than strategic.
• The service-provider model is the Chinese deployment pattern: most drones are bought by service companies and cooperatives, not by farmers. Any comparison with direct-sale models (Deere, DJI’s farmer channel) should start here.
• The “more than half of Chinese sales” claim was wrong and came from a company competing with XAG. Worth naming as a correction, because it entered circulation through syndicated Western trade press.

Critical context

• Prospectus figures are regulated, not audited for accuracy. They are the company’s own accounts, sponsored and reviewed as part of a listing process; the Frost & Sullivan market shares inside are commissioned. maturity-verification: V2 reflects “reputable regulated disclosure”, not independent verification of the deployment claims.
• Unit economics are still thin. FY2025 gross margin of 35.67% and adjusted net profit of CNY 138 m on CNY 1,166 m revenue is a modest business — the “Chinese agritech giant” framing in trade press overstates it.
• The 1.2 million farmers figure and the 27,000 drones in operation from the earlier corpus version are superseded by the prospectus’s 26,296 units sold in FY2025 and by the counties/countries channel figures. Keep the units-sold figure as the primary one.
• The STAR Market withdrawal (2022) is worth remembering. XAG’s first attempt at public capital failed; the Hong Kong route succeeded on a second attempt. Capital access for Chinese agritech has not been frictionless.
• Market-share claims about this market are structurally weak in every direction. The only commissioned study comes from one competitor; the market leader publishes nothing. Treat all share figures as contested.