Earnings reviews from this season:

My current portfolio.

Micron sells semiconductors for memory and storage. The firm's "NAND" chips offer non-volatile data storage, which maintains stored information when a system’s power is turned off. Separately, its Dynamic Random Access Memory (DRAM) chips offer volatile memory storage for personal computers, next-gen data centers, and more. “Volatile” means that storage isn’t maintained when a system’s power is turned off. DRAM helps processors access real-time data to alleviate processing latency, cost and other potential bottlenecks.

These chips provide the foundation for its solid-state drives (SSDs), which are used in computer data storage and things like USB flash drives. Micron sells standalone NAND/DRAM chips and also SSDs with their chips in them. SSDs replace hard disk drives (HDDs), as they’re more power efficient, durable and resilient. It provides basic memory cards for things like gaming devices and cameras as well.

Perhaps most interestingly, Micron offers a type of DRAM called high-bandwidth memory (HBM). This helps fulfill massive AI data processing needs. It sharply improves data processing capabilities and facilitates improved data sharing between CPUs & GPUs. Nvidia is a big customer, using Micron’s HBM in its Blackwell and future Rubin systems. Google and Amazon are big clients too. It also offers high-capacity SSDs to help with LLM storage.

As we work through this piece, keep in mind that Micron is a hyper-cyclical business. Demand fluctuates violently with changes in the macro environment. Margins do too, as pricing & utilization rates can experience hefty swings. Right now, the memory cycle is rocking in harmony with the overarching AI infrastructure cycle. Micron is a key enabler of this technological revolution.

Micron beat revenue estimates by 5.9% & beat guidance by 7%.

Micron significantly raised bit (unit of data storage) pricing for both DRAM and NAND. This powered Q/Q revenue growth. Bit shipments for DRAM rose “slightly” Q/Q while average selling price rose by 20% Q/Q. For NAND, bit shipments rose by around 7% Q/Q, with average selling price up around 15% Q/Q. They are flexing their pricing power muscles as the supply/demand backdrop remains convincingly favorable for the company.

Strong GPM expansion was helped by added fixed cost leverage from soaring demand levels and also disciplined cost growth. Additionally, price hikes helped GPM a lot.

OpEx rose by 27% Y/Y, which was much slower than revenue growth and amplified Y/Y EBIT gains.

For the full year, Micron now expects $20B in CapEx vs. last quarter’s $18B guide. The raise is to support strong demand signals as it struggles to keep up with customer needs. At the same time, management said their CapEx intensity (CapEx / revenue) is “dropping.” If they’re raising CapEx guidance by 11% and saying this? To me that means internal revenue expectations rose by more than 11% for the year. The 7% Q1 beat & 31% Q2 raise vs. analyst estimates bode well for that being the case.

They see results improving throughout fiscal year 2026, with strong demand and tight supply conditions lasting for at least another 3 quarters. The company is hard at work on securing additional (and large) multi-year contracts to keep this momentum humming. The team was noticeably excited about these prospects and about incremental customer interest in contracts that last longer than a year. If that momentum is strong enough, it could improve growth visibility for this cyclical business model. For now, conditions in 2026 should support strong growth at lofty margins. For calendar year 2025 (which has just 12 days remaining), Micron raised its DRAM bit (unit of data storage) growth estimate from ~18% to ~22%. It also raised its NAND bit demand forecast from ~15% to ~18%. For calendar year 2026, they expect bit shipment growth for both NAND and DRAM to be around 20% Y/Y (again with very tight supply conditions).

As its new DRAM and NAND nodes (manufacturing processes) ramp over the course of FY 2026, they expect a GPM tailwind to emerge. Leadership doesn’t expect the rapid pace of Q/Q GPM gains to be maintained (nor does anyone else), but does think GPM expansion can likely continue throughout the year. The large Q1 GPM and revenue beats, the larger Q2 beats and the consistent commentary on pace of GPM expansion helped profit estimates for the year spike higher.

Micron trades for 12x forward EPS. EPS is expected to grow by 280% this year, 18% next year and then fall by 10% the following year.

“Over the last few months, our customers’ AI data center build-out plans have driven a sharp increase in demand forecasts for memory and storage." – CEO Sanjay Mehrotra 

AI in Data Centers – A Massive Tailwind:

Micron sees its memory niche evolving from “system component to a strategic asset that dictates product performance.” As we talk about constantly, AI agents and models consume and process massive amounts of data. That requires a ton of memory to store data and make it cheaply available for usage. In turn, the evolving needs mean a lot more demand for things like Micron’s high bandwidth memory (HBM) offering and high capacity SSDs.

This leaves us with a compelling growth tailwind. Modern data centers need a lot more HBM to enable valuable, AI-powered work... and demand for these data centers is exploding higher. For evidence, server unit demand for calendar year 2025 should rise by nearly 20% compared to previous 10% growth expectations. For more evidence, the HBM market is expected to compound at a 40% clip through calendar year 2028 and reach $100B in total size two years sooner than Micron’s previous 2030 timeline. Truly amazing momentum right now for Micron as the memory cycle zooms.

The incredible momentum is why they’re so fixated on driving product leadership in this specific field. And while competitors would disagree, Micron feels like their DRAM and NAND offerings are both best-in-class in terms of cost per performance.

“This structural AI shift means that system capabilities heavily rely on advanced memory for real-time contextual processing. That is vital for achieving autonomous and intelligent AI.” – CEO Sanjay Mehrotra

Supply & Footprint:

As briefly mentioned, Micron now expects supply tightness for both DRAM and NAND to continue through calendar year 2026. DRAM is purely a byproduct of fantastic demand. The NAND tightness is via a combination of switching some capacity to fulfill more DRAM orders and also strengthening NAND demand.

The tightness will support a strong pricing environment for Micron for at least another 3 quarters as they struggle to keep up with customer interest. For context, HBM uses 3x the manufacturing capacity compared to its other DRAM-based offerings. That 3x will rise with future HBM generations, adding to capacity needs and overall demand levels.

And while this dynamic is great for forward-looking margins, they’d rather have all the capacity they need to fulfill all of the demand (hence the CapEx raise). They’re obtaining equipment, leases and power as quickly as they can to meet this moment, with encouraging progress including moving its Idaho fab completion schedule from 2H of 2027 to mid-2027. The second Idaho fab is still on track for 2028 production and its New York construction should begin early next year to support demand starting in 2030. They’re adding needed cleanroom space in Japan, HBM packaging capacity in Singapore for calendar year 2027 and a testing facility in India, which should be ready to roll next year. There are many projects being simultaneously executed, while the company continues to advance on schedule and as promised (if not better).

HBM subsection of DRAM:

Micron has locked in commitments and pricing for all HBM supply through 2026. This includes HBM4, which is on track for scaled deliveries next quarter. The “and pricing” part of that isn’t shocking, but still encouraging following considerable price hikes. By controlling the building materials and vertically integrating the wiring processes and the movement of data (via its own DRAM logic dies). HBM3 relied on Taiwan Semi for logic die, but Micron manufactured it on its own for HBM4. This has helped unlock what they view as industry-leading performance and a larger chunk of the margin for them to fetch for themselves.

Leadership was asked about Samsung’s recently invigorated push into HBM. They’re not seeing this impact demand at all, which makes sense considering they can’t make enough to meet customer orders.

“We are excited about our customized HBM4E (next iteration) customer engagements which offer further differentiation opportunities to us, and we continue to make excellent progress on our HBM road map.” – CEO Sanjay Mehrotra

Low Power DRAM (LP DRAM): 

While LP DRAM has been mainly for mobile use cases in the past, Micron is bringing this to the data center thanks to lower power and cooling needs. This isn’t as fast as HBM, but it’s still great for some inference workloads where customers are more sensitive to cost than overall performance. And it’s also still very fast in terms of data fetching, just not quite as fast as HBM4. They’re working on a new LP DRAM iteration (called LP SOCAMM2 but not important), which promises to deliver 50% higher capacity per unit, boosting efficiency and performance for this cheaper memory product even more.

DRAM & NAND Product Roadmap Schedules:

Going back to Micron’s obsessive mission to drive tech progress and leadership, its new 1-gamma node (node = manufacturing process) is on schedule. Mehrotra thinks this will likely be the 5th consecutive technology node in DRAM that “leads the industry.” Again… SK Hynix would disagree, but the progress Micron is making in this space is undeniably awesome and that couldn’t happen if their tech was inadequate. 1-delta and 1-epsilon nodes will come after 1-gamma.

On the NAND side, its new “G9” node and quad-level cell (QLC) NAND product set new revenue records this quarter. Mainstream storage (for broad-based use cases) and capacity storage (for packing as much data storage as possible for the lowest possible price) also both enjoyed strong quarters as Micron continues to advance both of those product roadmaps as well.

Using AI Internally:

Micron is voraciously consuming AI to improve internal operations as well. 80% of its workforce is using GenAI tools (+10X Y/Y). They’re cutting root cause analysis for manufacturing performance bottlenecks, boosting coding productivity by 30%, cutting design cycle times and more. They’ll keep leaning into this technological revolution for themselves, in addition to providing foundational hardware to fortify it. 

More on NAND:

Debuted the 6th generation of its Peripheral Component Interconnect Express (PCIe) SSDs. These can move 28 gigabytes (GB) of data per second vs. 14 GB for the 5th generation. Demand for this has been strong, while data center NAND revenue for the quarter crossed $1B and Micron rose to the 3rd-ranked SSD vendor by market share.

Non-Data Center Demand:

For the Personal Computer (PC) segment, Windows 10 end of life is driving strong PC-related demand. The AI-PC refresh cycle has been strong. They now see calendar year 2025 growth for this segment coming in above the ~5% expectation they laid out last quarter. They see strong 2026 demand, but cautioned everyone about expected supply-related issues for “some PC unit shipments.”

In mobile, just like on the PC side, AI is driving rising memory needs. Phones with 12GB of DRAM rose by 59% Y/Y this quarter, which compares to roughly 30% Y/Y growth for these types of models last year. AI is having the same impact on memory demand from the automotive segment, and is unlocking a larger robotics segment for Micron to pursue as well.

This was an excellent quarter. Micron is taking market share in the extremely important HBM arena while commanding a large portion of the (also important) high-capacity SSD opportunity. They’re doing this with multi-quarter demand visibility that is better than what Micron can typically provide investors (outside of the last handful of quarters). That’s emblematic of how strong the AI infrastructure cycle is for memory providers like this one. And the nice thing? They don’t really care if GPU leaders like Nvidia and AMD win more share… or if custom accelerator players like Google, Amazon and Broadcom win more share. Micron wins regardless. All of these chips will need more and more memory bandwidth at world-class performance, and Micron is happy to provide it. It looks like results will stay great for at least another 3 quarters and likely into FY 2027. From there, who knows what AI cycle longevity will look like, which matters a ton for this business model.

Beat revenue estimates by 1.9% & beat revenue guidance that called for a low single-digit Y/Y decline.

GPM is being greatly affected by ongoing inventory discounting and liquidation – especially in China. There are puts and takes here. On the takes side, they're doing less Nike.com discounting to improve brand perception, which helps margins a tad. At the same time, that decision lowers overall demand, which is negative for margins. Furthermore, there's so much low-quality inventory that needs to be cleared at razor-thin margins through wholesale channels. That headwind alone is much larger than the aforementioned Nike.com tailwind. Outside of this, tariffs are hitting margins as well. The gross Y/Y headwind was 320 bps, but leadership was able to reduce this to 120 bps through good supply chain work. In UCAN specifically, GPM fell by 330 bps Y/Y despite a 520 bps headwind from tariffs, meaning GPM would be expanding without this headwind. As long as tariffs don’t move even higher, this Y/Y comp headwind will go away towards the end of this fiscal year. The ex-tariff expansion is thanks to UCAN turnaround progress we’ll dig into later in the piece. Lastly, revenue declines and wholesale mix-shift drove the rest of the GPM pressure.

Nike guided to a low-single-digit Y/Y revenue decline next quarter, which missed 1% Y/Y growth expectations. Nike’s forecast includes 3 points of foreign exchange (FX) help, which leads to constant currency (CC) growth guidance of around -5% Y/Y. On the bright side, they’re increasingly confident in Y/Y wholesale growth for this fiscal year.

It also expects 200 bps of Y/Y GPM declines, while analysts were expecting 80 bps of Y/Y GPM margin expansion. Tariffs are a 315 bps headwind as they hold back Nike profitability, but that should have been fully understood and baked into analyst expectations at this point. The company sees a clear path back to a 10%+ EBIT margin over time.

Estimates will fall in the coming days. As of right now, Nike trades for 35x forward EPS estimates. EPS is expected to fall by 24% this year (fiscal year ends in May), rise by 51% next year and rise by 23% the year after.

Turnaround Priorities:

As a reminder, Nike’s priorities following Hill being named CEO were:

Starting from the top, classic shoe franchise revenue fell 20% Y/Y and lowered revenue by around $550M. Excluding this, CC revenue growth would have been 6% Y/Y. That’s a good indicator of where growth can go when they finish intentionally cutting these large, mature businesses.

For an idea of how material of a challenge this is, by the end of this fiscal year, Nike expects its classic footwear segment to be $4B smaller than the peak. They’ve been very aggressive here, as they see this as a vital prerequisite for driving more sustainable growth. Like LULU, this company relied on what made them special for too long. Nike didn’t look to cultivate other things and products that... well... keep making them special as preferences evolve and generations age. All of this has been painful for financials, but I agree with their decision.

For priority #2, $NKE has been intentionally discounting less aggressively on its digital storefront and 1st-party stores. Days of promotion, markdown rates and discounted sale rates all fell Y/Y. That has been another revenue concession they’ve needed to make in order to clean up an unhealthy marketplace. The brand prestige was waning and wholesalers were growing tired of Nike’s endless undercutting. The firm needed to pivot and they did.

Priority #3 (fixing wholesale relationships) is innately tied to #2. They fix things here by ending the perpetually aggressive direct-to-consumer discounting that effectively steals business and margin from its partners. Driving more full-price sales in direct channels is the best way it can fix indirect channels too. It’s also working on improving displays at 3rd-party stores and opening up higher value inventory for these retailers. This strategy is enabling strong Y/Y wholesale growth and positive wholesale order book gains for spring and summer 2026.

And finally, Nike has realigned their teams to “place sport at the center.” Groups were organized around individual sports rather than its 3 major brands and the juggernaut slashed several middle management layers to turbocharge innovation. They’ve gotten more focused on taking great care of athletes, improving messaging and showing up in big ways at global sporting events. This realignment unleashes what Nike calls “sport offense,” or a steady flow of relevant and well-placed product innovation driven by better organizational structure and priorities. Becoming this finely-tuned operational machine once more will allow Nike to turn its hefty comeback investments into durably strong ROI.

During the quarter, Nike also announced a couple of other changes. First, all geographic general managers (GMs) will report directly to Hill to drive better communication and faster innovation. Furthermore, Venkatesh Algirisamy was named the new COO. Nike made margin expansion a bigger theme of this call than any since Hill took over. They’re dedicated to making the investments needed to fix the business, but think they can do so in much more efficient ways that hurt margins less dramatically.

“I'd say we're in the middle innings of our comeback.” – CEO Elliott Hill

Signs of Progress – Running & North America:

The clearest signs of progress continue to be within Nike Running and the North American (UCAN) business. Running again grew by 20% Y/Y and enjoyed 10%+ growth across all of its channels and most geos. And as you can see in the chart earlier in this piece, UCAN was the sole source of its positive Y/Y growth. It overcame revenue declines in every other region and put the proverbial team on its back. This is encouraging. Nike decided to make these two major buckets its first two focus areas, with progress offering tangible evidence of its strategic playbook working. It gives me more confidence in results looking better and better as it implements the approach across more facets of its business. Early signs of this plan working in basketball, soccer and training provide more reasons for optimism. As Nike says all the time, this thing will take a while to fully recover. As we track the firm's progress, it's good to see the preliminary blueprint effectively taking hold.

“We are growing more confident in our ability to sustain the running and UCAN momentum as we look forward.” – CFO Matt Friend

Going back to UCAN momentum for a minute, the company now feels like its “inventory is in a healthy and clean position.” They’ve cleaned up a lot of stale assortment and gotten on a much better marketing rhythm.  They’ve also repaired key partnerships, paving the way for 20% Y/Y wholesale growth in this important region. The final UCAN inventory liquidations from the quarter helped this number, but so did strengthening underlying traction.

“The Win Now Actions and the Sport Offense are working, and it will lead us back to profitable, sustainable growth.” – CEO Elliott Hill

Various Phases of a Turnaround Elsewhere:

The positive note on UCAN inventory being in good shape is also true for Europe, the Middle East & Africa (EMEA). Together, those two categories make up more than 70% of Nike’s business. This should mean less discounting going forward. That would remove sizable margin headwinds, allowing other tailwinds like lower Nike.com discounting to become more apparent – especially once the tariff headwind is lapped.

They group Asia-Pacific and Latin America (APLA) together, but the two regions are performing at very different levels. The “AP” part is weak and the “LA” part is strong. Unlike UCAN/EMEA, inventory liquidations are ongoing, as they have more excess goods to shed before getting things to a healthier place in those regions.

China Recovery Lagging:

China is where things remain the weakest and most challenged. They’re trying to fix the business with more local messaging and relevant assortment but are disappointed in the pace of improvement. The company had gotten too reliant on promotions and distracted from more durable growth levers like taking good care of stores and delivering good products. They weren’t executing in the most basic of areas, with an uninspiring offering leading to heavier discounting and structural pressure on its margins there. Hill acknowledged this and told us China is why all geography-level GMs are now reporting directly to him. He also told us that the Chinese market is a different animal and that Nike hasn’t done a good job of adapting to that reality. They’re confident that the brand remains highly relevant and they will right the ship, but it will “take time” and big changes.

“The China reset requires a fresh way of thinking from our Nike teammates and our Nike store partners, and it will take time… it's clear that we need to reset our approach to the China marketplace. The priority is to create greater brand distinction through sport and innovation. We will leverage deep local insights in a premium, consistently managed and integrated marketplace across channels… I've done this before. I have a deep history in the market. We've diagnosed the problem, and we will return Nike to a beloved, premium, and innovative brand in China.” – CEO Elliott Hill

The quote above is a bit flowery and vague to me, but we thankfully did get more concrete color on their plan of action. They are ramping testing for their new store format, which is delivering better comp store sales vs. the overall base. They accelerated partner inventory returns, wrote all of that off and slashed their wholesale revenue plans for this spring and summer. These actions allowed inventory in that region to fall by ~15% Y/Y. They’re fixing a very messy situation, which, when done, should allow them to get more aggressive with growth and innovation plans in that region. For now, revenues are falling, store traffic growth is negative, EBIT declined by 49% Y/Y and the business will remain a mess.

Product Highlights:

This was another unsurprisingly underwhelming quarter. It may feel as though the progress Nike delivered 90 days ago stalled, but I don’t think that’s fair. UCAN and running (their first focus areas) were very healthy for a second straight quarter, which gives more evidence of their strategy being a good one. Basketball, training and soccer are also enjoying better traction, but need more time for that to build and become needle-moving for overall results. For now, ongoing inventory liquidations and severe challenges in China are overshadowing the modest progress Nike is making. The inventory headwind will soon go away, but China is probably going to take a while.

With all of that said, I think this team is doing all of the right things and their task at hand is absolutely daunting. Nike can’t fix everything at once. It rightfully focused on its largest segments first, and is now getting to everything else. Turning around giant and global enterprises that were horrendously run for a long stretch of time is hard and extremely messy. They’re getting through it about as well as I think anyone could have realistically expected. They’ve told investors to be patient and that the recovery won’t be linear or rapid, and there are signs of the initial work bearing considerable financial fruit. 

For me, despite being impressed with Hill, this is firmly in the too hard pile. I think Nike is in great hands and will recover eventually. I see Nike delivering mid-to-high single-digit revenue growth and margin expansion when the comeback is complete. I just think it will be an extremely noisy, chaotic and slow process. And I see other investments (including ONON) that I like more than this one.