MU Q3 2026 Earnings Analysis

Yeah, "don't sound real" is right. Let's just start with the top line — $41.5 billion in revenue. That's up 74% sequentially and 346% year-over-year.

DRAM did most of the heavy lifting — $31.3 billion, up 67% sequentially, three quarters of total revenue. But NAND actually grew even faster percentage-wise, up 99% sequentially to $9.9 billion. Prices in NAND jumped mid-80s percent.

These aren't small beats over guidance either — Sanjay Mehrotra said revenue, gross margin, and EPS all exceeded the high end of guidance. And the data center number is the real story underneath all this — data center revenue exceeded $25 billion in the quarter, annualizing above $100 billion. Data center SSDs alone more than doubled sequentially to over $5 billion.

This is the part I think investors are going to spend the next few quarters trying to fully digest. These are take-or-pay, multi-year deals — five years for most, three years for automotive — running roughly calendar 2026 through 2030. They cover about 20% of Micron's DRAM volume and a third of NAND volume over that period.

That's the headline for me. Historically Micron's margins have been this brutal boom-bust cycle — peaks in the low 60s percent gross margin, then crashing during downturns. If the floor on these new contracts is above the old ceiling, that's a structurally different business.

And this is where the Q&A got really interesting. Analysts pushed hard on what that $100 billion actually represents. Tim Arcuri from UBS did the math — $100 billion over roughly five years is about $20 billion a year, which is well below Micron's current $40-plus billion quarterly run rate.

Mark Murphy also fielded some good detective work from Morgan Stanley's Joe Moore on the cash deposits — Micron's getting about $22 billion in deposits and financial commitments tied to these deals, $18 billion of that in actual cash. Analysts wanted to know: is this a prepayment? Collateral?

Which tells you these customers — and remember, we're talking four very large customers, three medium ones, plus a batch of smaller auto suppliers — are willing to lock up real capital just to guarantee supply. That's a pretty strong signal about how tight this memory market actually is.

The reasons are structural too — greenfield fab construction takes years, there's a skilled labor shortage, permitting is complex, and technology transitions themselves are yielding slower bit growth per node. Plus HBM's trade ratio — it eats disproportionate wafer capacity relative to the bits it produces, which squeezes supply of everything else.

Now, guidance. Fiscal Q4 revenue guide is $50 billion, plus or minus a billion — another record. Gross margin guided to about 86%, and EPS guided to $31 a share. But Mark Murphy did flag that the rate of price increases is moderating, so don't expect gross margin expansion at the same pace going forward.

So stepping back — what does this mean going forward? The bull case here is that Micron is trying to convert from a classic cyclical commodity business into something with contracted, multi-year visibility and margins that don't collapse in the next downturn. The SCAs are the vehicle for that. The open question analysts kept circling back to is just how much of future revenue actually falls under these protective floors versus how much is still exposed to spot pricing that could eventually soften.

Before we wrap, one more required note: everything discussed is AI-generated analysis for educational purposes. Past performance doesn't guarantee future results. Please do your own due diligence.

Thanks for tuning in to Beta Finch — we'll catch you next time.