Megmeet's Q1 Revenue Meets Expectations; AI Server Power Supply Volume Boosts Earnings Forecast, Maintaining Neutral Rating

Overview

This report provides commentary on Megmeet's first-quarter results for 2026. Although overall revenue grew 20% year-over-year, meeting expectations, operating profit (OP) fell below Goldman Sachs' forecast due to margin pressure. The key highlight is the rapid expansion of the AI server power supply business, which has secured bulk orders for NVIDIA's GB200/GB300-related products and sees opportunities arising from the Vera Rubin architecture. Given this business's strong growth potential, Goldman Sachs has raised its earnings estimates and target price, yet maintains a 'Neutral' rating, arguing that the current stock price reflects a fairly balanced valuation considering both risks and rewards.

Core views

In terms of financial performance, Megmeet reported Q1 revenue of 2.788 billion yuan, up 20% year-over-year; however, operating profit stood at only 113 million yuan, down 32% year-over-year—below Goldman Sachs' (-5%) and Visible Alpha consensus (-45%) expectations. Gross and net margins declined by one percentage point each, to 22% and 4%, primarily due to RMB appreciation, rising raw material costs, and intense price competition. The AI server power supply segment was the primary driver of growth this quarter. Power supply sales surged 67% year-over-year to 817 million yuan, with server power supplies leading the way. The company has secured bulk orders for 5.5kW power supply units (PSUs) destined for NVIDIA's GB200 and GB300 systems, with orders and shipments continuing throughout 2026. Management noted that 18.5kW PSUs for Vera Rubin NVL72 racks and 110kW power shelves are currently in the sampling phase, expected to become significant growth drivers in late 2026 through 2027. Consequently, Goldman Sachs significantly revised upward its revenue projections for server power supplies: 1.4 billion yuan in 2026 and 4.9 billion yuan in 2027 (previously 1.2 billion and 4.2 billion), projected to account for 11% and 28% of total revenue, respectively, and 18% and 41% of total gross profit. Traditional core businesses showed mixed results. Home appliance control sales grew modestly by 3% due to high base effects and macroeconomic uncertainties—such as Middle East conflicts impacting emerging market energy supplies and domestic real estate weakness. Growth in new energy vehicle and rail transit components slowed sharply to 12% (down from 109% in 2025), largely dragged down by weak domestic electric vehicle demand. By contrast, industrial automation remained robust, expanding by 20%. For non-AI core businesses, Goldman Sachs expects 18% revenue growth and a 22.1% combined gross margin in 2026. Regarding valuation and forecast adjustments, given the strong prospects of the AI power supply business partially offsetting macroeconomic headwinds (currency and cost pressures) on margins, Goldman Sachs raised its 2026–2030 net profit forecasts by an average of 4%. Accordingly, the target price was increased from 98 yuan to 112 yuan. Valuation methodology employed a 32x P/E multiple for 2028, discounted back to 2027, with a cost of capital (CoE) of 11%. Current share prices correspond to approximately 105x/46x P/E ratios for 2026/2027, well above the median for China's industrial technology sector, thus justifying the maintenance of a neutral rating.

Analysis framework

Goldman Sachs' analytical framework follows two main threads: 'structural transformation validation' and 'valuation anchoring.' First, analysts dissect growth across business segments, identifying the company's transition from a traditional automation and control specialist into a global competitor in AI server power supplies. They closely monitor order fulfillment for AI power supplies—such as bulk orders for GB200/GB300—and track R&D progress on next-generation products like 800V DC and solid-state transformers (SST), using these indicators as key measures of future revenue elasticity. Second, when assessing overall profitability, analysts apply a 'add-and-subtract' logic: assigning higher revenue weights to fast-growing AI segments while factoring out macroeconomic headwinds and currency-related cost pressures affecting traditional core businesses. Finally, at the valuation level, rather than relying solely on static P/E multiples for the current year, analysts employ a multi-period discounting approach—discounting forward-looking 2028 P/E ratios back to 2027—to smooth short-term volatility and capture long-term growth value. This method acknowledges the incremental opportunities brought by AI while cautioning against overheated valuations through elevated multiples, ultimately arriving at a balanced 'Neutral' conclusion.

Methodology notes

• Industry/sector analysis frameworkPenetration Rate S-Curve

  Technology iteration and product lifecycle analysis

  The report analyzes how data center power architectures are evolving from traditional AC toward 800V DC and solid-state transformers (SST). Analysts note that although newer technologies represent the future direction, large-scale adoption will take time due to supply chain, safety, and integration challenges, leaving AC still dominant today. This reflects an assessment of where different technological stages fall along the S-curve.

• Valuation methodDCF Cash Flow Discounting

  Multi-stage valuation model (implicit)

  The report calculates the target price by discounting forward-looking (2028E) P/E ratios back to the near term (2027E), incorporating a cost of capital (CoE of 11%). Essentially, this is a simplified application of discounted cash flow (DCF) or residual income models, aiming to determine intrinsic value by accounting for time value of money and long-term growth rates, rather than relying solely on current earnings multiples.

Asset mapping & comparison

Structured mapping from thesis to named assets (strengths, weaknesses, peers, risks).

• Megmeet (002851.SS)

  Benefit Logic: AI server power supply business secures bulk orders for NVIDIA's GB200/GB300, with next-gen 18.5kW PSU and 110kW power shelf samples leading the way. Revenue from this segment is expected to surge in 2026–2027, optimizing the overall revenue structure.

  Strengths

  Solid R&D pipeline; development of 800V DC and SST products aligns with global peers, even surpassing them in certain projects; entry into NVIDIA ecosystem and recognition from North American CSP customers.

  Weaknesses

  Still lags behind top-tier competitors like Delta Electronics in high-efficiency product portfolios; power supply orders are project-based, resulting in volatile monthly flows and low visibility; traditional core businesses heavily impacted by macroeconomic headwinds.

  Comparison

  Compared to industry leader Delta Electronics, Megmeet has gaps in high-end, efficient product offerings; compared to other newcomers, it enjoys certain first-mover advantages in certification and sample phases, but faces intensified competition due to blurred supply chain boundaries.

  Risks

  Slower-than-expected market share gains in AI server PSUs; delays in validating ASIC hardware supply chains; underperformance in large-scale production execution; extended R&D cycles for 800V DC products; declining sustainability of global AI capital expenditures.

Key data

• Q1 2026 Revenue2.788 billion yuanUp 20% year-over-year, in line with expectations
• Q1 2026 Operating Profit113 million yuanDown 32% year-over-year, below expectations
• Q1 2026 Gross Margin22%Down one percentage point year-over-year
• Q1 2026 Server Power Supply Revenue Forecast1.4 billion yuanRaised from 1.2 billion, accounting for 11% of total revenue
• Q1 2027 Server Power Supply Revenue Forecast4.9 billion yuanRaised from 4.2 billion, accounting for 28% of total revenue
• Adjusted Target Price112.0 yuanRaised from 98.0 yuan
• Current Valuation Multiples105x / 46xCorresponding to 2026E/2027E P/E ratios

Impact & implications

For Megmeet, the successful ramp-up of AI server power supplies marks the establishment of its second growth curve, significantly boosting the company's revenue scale and market visibility. However, since this business still accounts for a relatively small share (estimated at only 11% of revenue in 2026) and faces fierce competition—still lagging behind industry leaders like Delta Electronics—the short-term impact on overall profit margins remains limited, potentially even suppressing gross margins during early capacity expansion phases. For investors, the current stock price already reflects high growth expectations (105x P/E), leaving limited room for further upside unless the company achieves breakthroughs in ASIC power supplies or 800V products; otherwise, the stock may remain subject to broader market fluctuations or wait for clearer earnings realization.

Risks

• Slower-than-expected market share gains in AI server PSUs
• Delays in validating ASIC hardware supply chains
• Underperformance in large-scale production execution
• Extended R&D cycles for 800V DC products
• Slower-than-expected industry adoption of 800V DC distribution architectures
• Unsustainable growth in global AI capital expenditures and shifting AIDC supply-demand dynamics
• RMB exchange rate volatility and rising raw material costs

What to watch

• Mass production yield and delivery schedules for AI server power supplies
• Progress in developing next-generation 800V DC products and solid-state transformers (SST)
• New order acquisition status among North American and Chinese cloud service providers (CSPs)
• Signs of recovery in traditional home appliance control and new energy vehicle businesses