SF Express's early-incubated retail locker business is about to go public.

```

On June 10, Shenzhen Fengyi Technology Group Co., Ltd. (operating brand "Feng e Foot Food", hereafter referred to as Fengyi Technology) submitted its prospectus to the main board of the Hong Kong Stock Exchange. Agricultural Bank International acts as the sole sponsor, while China International Capital Corporation and CMB International serve as overall coordinators.

The story of Fengyi Technology began in 2017 as an internal incubation project by SF Holdings. At that time, unmanned shelves were trending, and SF, relying on its dense courier network and vast corporate client resources, entered the office snack shelf business.

As the industry receded and business models evolved, Fengyi Technology gradually shifted its focus from high-loss-rate open shelves to intelligent retail cabinets, and introduced external capital for independent development.

Historical financing information shows that SoftBank Asia Venture Capital once led a 300 million RMB Series A round of financing, with follow-on investments from CICC Capital and Shenzhen Capital Group. After this round, SF affiliate Mingde Holdings and SoftBank Asia became core shareholders.

The latest disclosed prospectus data indicate that SoftBank currently holds more than 25%. From an edge business spawned by a logistics giant to independently seeking a listing, this is not only the conventional path for capital exit, but also reflects the strong demand for funds in the secondary market for asset-heavy direct operation models.

In terms of revenue, Fengyi Technology’s income structure is extremely single.

The prospectus shows that during past performance record periods, about 99% of the company’s revenue directly came from food and beverage sales from intelligent retail cabinets to end consumers.

This structure means it lacks high-margin financial hedges such as advertising display or equipment leasing, and profit squeezing relies entirely on expansion of scale and refined supply chain circulation.

From the basic financial perspective, the company achieved synchronised growth in revenue and site scale. From 2023 to 2025, Fengyi Technology's operating income grew from 1.244 billion RMB to 2.009 billion RMB, with a compound annual growth rate of 27.1%; during the same period, the overall gross profit margin increased slightly from 52.4% to 55.8%.

Of particular note is performance on the profit side. Given the industry characteristics of asset-heavy investment and high fulfilment costs, Fengyi Technology recorded an adjusted net profit (Non-GAAP) of 118.6 million RMB in 2025.

Against a backdrop where most intelligent cabinet peers are stuck in loss, this marks the company having achieved initial scalable profitability under the direct operation model.

But on the flip side, as the equipment network expands, the pressure from depreciation and amortization caused by future large capital expenditures will continue to erode the actual profit space in the long term.

Traditional unmanned retail usually competes in core business districts, transport hubs, and other “heavy scenarios”, exchanging high rental fees for absolute foot traffic. The differentiation of Fengyi Technology lies in the downward shift of site focus, targeting offices, industrial premises, inner commercial zones, logistics warehousing facilities, and other “light scenarios”.

As of December 31, 2025, Fengyi Technology's direct-operated intelligent cabinet count reached 183,989, covering 72 cities nationwide, with a compound growth rate exceeding 30% compared to 106,000 at the end of 2023.

The characteristic of "light scenarios" is low single-point traffic ceiling and fragmented demand, but the core advantage is relatively low site rental cost and stable daily instant consumption repeat purchase rate among the customer group.

This downward strategy avoids direct rental wars but also results in exponential complexity for back-end scheduling. When the terminal network approaches 200,000 units, traditional manual station inspection and replenishment models cause fulfilment costs to soar, eventually eating up the already not-high per-cabinet profit.

To counteract operating losses caused by high-density "light scenarios", Fengyi Technology made substantial capital investments on the technology side.

The prospectus mentions that the company began algorithm-driven operations in 2022 and launched the FLOW Pilot intelligent system composed of four intelligent agents in 2024, covering the entire chain of site expansion, product mix, inventory planning, and task scheduling.

From the perspective of industrial digitalization, this is essentially a management tool that algorithmizes the logic for replenishment scheduling and site swap-out. Among tens of thousands of discrete sites, the system relies on real-time operational data feedback to make autonomous decisions, greatly reducing the proportion of manual judgment in site evaluation and supply chain scheduling.

The prospectus, citing Frost & Sullivan data, notes that during past performance periods, Fengyi Technology's annual R&D expenditure exceeded 60 million RMB. In a retail sector with already thin gross margins, this expense can be seen as the fixed tech cost required for the company to reduce long-term fulfilment costs and maintain direct operation network functioning.

Overall, Fengyi Technology’s prospectus showcases a business sample that achieves book profitability by relying on "light scenario direct operation + algorithmic scheduling". But under macro conditions where the downward dividend in retail terminals is nearing its ceiling and the slope of offline consumption recovery remains uncertain, to what extent its fulfilment network made up of 180,000 intelligent cabinets can resist potential price wars in the market remains a key metric that secondary markets need to continue to scrutinize.

Risk warning and disclaimerThe market has risks, investment needs to be cautious. This article does not constitute personal investment advice and does not take into account the special investment objectives, financial conditions, or needs of individual users. Users should consider whether any opinions, views, or conclusions in this article fit their particular circumstances. Investment made based on this article is at your own risk. ```