Microchip's Data Center Revenue Grew 98% Last Quarter. Here's Why It's Guiding to $1 Billion for the Year.

Microchip’s fiscal Q1 net sales jumped 38% YoY to $1.485B, topping the high end of guidance, with adjusted EPS rising 181.5% to $0.76 and GAAP net income of $202M (vs -$46.4M a year ago). Management guided September-quarter sales up 7%–9% sequentially (40.6% YoY at the midpoint) and expects its data center portfolio to grow 69.3% to $1B in calendar 2026. Shares reacted with an about 14% jump, supported by accelerating data center revenue (+97.8% YoY) and improving financial footing (net debt down ~$170M; dividends of $246.9M).

Analysis

The market is starting to price MCHP less like a late-cycle industrial recovery and more like a hidden AI-infrastructure supplier. That matters because the current profit step-up is being driven by fixed-cost leverage; if data-center mix keeps rising, incremental revenue should convert at a meaningfully higher margin than the legacy automotive/industrial base, supporting a rerating rather than just an EPS rebound. The key question is not whether the next quarter is fine, but whether management can keep converting design wins into backlog across multiple server generations.

Second-order beneficiaries are the broader AI hardware stack that sits around GPUs: connectivity, timing, storage, and power-management vendors with exposure to server content should see a rising TAM narrative, while purely industrial-chip peers may look comparatively ex-growth. The flip side is that this is still a small part of the company’s revenue base; if the AI build-out pauses or customers consolidate around fewer qualified vendors, the growth story can decelerate faster than consensus expects and the multiple can snap back toward a cyclical semiconductor valuation.

Near term, the stock likely needs time to digest the gap move; the better signal is the next 1-2 quarters of bookings and margin durability, not the headline revenue print. The contrarian view is that investors may be overestimating how quickly a few PCIe wins translate into durable share gains, especially with inventory still elevated versus pre-downturn levels. Falsify the thesis if data-center growth falls back toward mid-teens sequentially annualized, gross margin stalls below the low-60s, or guidance implies the rebound is merely restocking rather than end-demand.

Key Decisions for Investors

• Buy MCHP on a 5-10% pullback from the post-earnings spike; 6-12 month thesis is rerating if data-center revenue remains >50% y/y and gross margin stays above 63%.
• Pair trade: long MCHP / short TXN for 3-6 months to express AI-content optionality versus a more mature analog/microcontroller incumbent; exit if MCHP guides below 5% sequential growth or TXN reaccelerates faster than expected.
• For lower-risk exposure, use a bullish call spread in MCHP into the next earnings cycle; the setup is better for defined-risk upside than chasing the gap after a 14% move.
• Set a watch item on data-center design-win conversion and inventory days; if inventory stops falling or the PCIe win pipeline does not convert within 2 quarters, reduce exposure.