Thesis
Applied Materials is the broadest high-quality way to own rising semiconductor manufacturing complexity. The company sells the deposition, etch, implantation, metrology, packaging, and service solutions that let chipmakers actually move from roadmap slide to high-volume production. That matters because the most important AI-era inflections are happening at the process level: gate-all-around transistors, HBM, hybrid bonding, and more complex 3D integration all require more materials engineering and more process steps per wafer. Applied is positioned across several of those inflections at once rather than relying on one hero tool.
The business is already proving that this is more than a theme. In fiscal 2025, Applied generated a record $28.37 billion of revenue and record non-GAAP EPS of $9.42. Semiconductor Systems alone delivered $20.80 billion of revenue, while Applied Global Services contributed $6.39 billion. Management said in November 2025 that it was preparing operations and service organizations for higher demand beginning in the second half of calendar 2026. The real investment question now is whether AI-related logic, memory, and advanced packaging demand is strong enough and broad enough to keep Applied in a richer cycle than investors usually grant a WFE leader. If yes, the stock can continue working as both a quality compounder and an AI infrastructure beneficiary. If not, it falls back into the box of 'great company, ordinary semi-equipment cycle.'
Applied is clearly positioned at several major AI-era device inflections. The real question is whether logic, DRAM, advanced packaging, and services stay strong enough together to keep this a richer cycle than a normal semiconductor-equipment upturn.
Valuation and financials
The 4Ps
Gary Dickerson has spent years positioning Applied around the next major device transitions rather than just defending legacy share. The message in 2025 and early 2026 was consistent: the company is aiming at the highest-value inflections in logic, DRAM, and advanced packaging. That strategy only works if R&D dollars keep landing in the right places, but Applied has earned some credibility there.
Applied matters because it sells into many of the places where semiconductor complexity shows up: front-end process steps, metrology, packaging, and service. The 2025 product cycle made that concrete. Kinex targets hybrid bonding for logic and memory, Xtera helps enable 2nm-and-beyond GAA structures, and PROVision 10 is aimed at metrology challenges in complex 3D logic and memory. That is exactly the kind of product breadth that makes one cycle feel larger than one tool win.
Applied does not need one market to explode. The better bull case is that leading-edge logic, DRAM/HBM, advanced packaging, and services all stay active enough at the same time to keep the revenue base elevated while maintaining healthy margins. If several of those layers stay strong together, the company can earn more than a normal WFE leader in a simple one-node upgrade cycle.
Applied is still cyclical, but it has more stabilizers than many equipment peers. The service business is large and recurring, the product portfolio spans multiple process steps, and the customer base tends to telegraph major technology ramps well in advance. That does not eliminate volatility, but it does make the business easier to underwrite than a narrower equipment story.
Portfolio manager lens
Starting point: Applied is the broadest quality way to own semiconductor complexity without needing to guess one specific device winner.
What is in the stock: strong 2025 results, a very large service base, and the idea that AI-era logic, memory, and advanced-packaging transitions stay favorable into late 2026 and beyond.
What can still surprise upside: hybrid bonding and HBM becoming bigger revenue contributors, services staying strong, and customers pulling Applied deeper into roadmap-critical process steps.
What changes the view: a narrower-than-expected spending cycle, export-control or customer concentration pressure, or new technologies that stay strategically important but fail to translate into meaningfully better earnings power.
Trade framing
Applied is not early. The market already knows it is well positioned. The better setup is around confirmation that the company is entering another period where breadth and services matter more than simple cycle beta.
The next checkpoints are straightforward: does revenue track around the fiscal first-quarter 2026 guide of $6.85 billion plus or minus $500 million, do customers keep signaling stronger demand into the second half of calendar 2026, and do the advanced-packaging and HBM-related collaborations move from strategic headlines toward visible shipped business? If yes, the stock can continue compounding from a stronger base. If not, it stays a great company in a more ordinary cycle.
What matters now
What matters now is whether advanced packaging, DRAM, and services stay strong enough together to keep this cycle richer than a normal wafer-fab-equipment upturn. The checkpoints are HBM-related spending, leading-edge logic intensity, and whether service revenue keeps smoothing the earnings profile.
Key questions
Applied sells semiconductor manufacturing equipment and services across the chipmaking flow. Financially, the company is easiest to understand through its reporting segments. In fiscal 2025, Semiconductor Systems generated $20.798 billion of revenue, Applied Global Services generated $6.385 billion, and Display contributed about $1.06 billion within Corporate and Other.
The practical point is that Applied is not one machine. It is a materials-engineering platform. It helps customers deposit, remove, modify, inspect, connect, and service the materials that become a modern chip. That breadth is why the company shows up repeatedly whenever the industry hits a new complexity wall.
Thesis last reviewed April 2, 2026. Live data updates automatically.