Micron Deep Dive: What a 41.5% Net Margin and 194% Revenue Growth Really Mean

Three Rounds Micron Crushed

What pulls me toward Micron isn't the headline finish — it came second in my 6-round comparison. It's how Micron won the three rounds it did win.

Net margin: 41.5%. Forward revenue growth: 194.1%. Profit-adjusted PE: 0.27.

I've been analyzing semis for long enough to know that those three numbers don't usually live in the same company. High-growth names trade at premium multiples. High-margin names are mature and grow slowly. Micron is breaking both patterns at the same time, and I think that's the part the market still hasn't fully priced in.

What a 41.5% Net Margin Tells Me

I genuinely can't remember the last time I saw a memory company posting 40%+ net margins. For most of my career I've used single-digit memory margins as 'normal' and double-digit margins as a cycle peak warning.

That framework just doesn't work in this cycle. 41.5% means Micron has effectively become a price-maker in HBM. When NVIDIA needs to align its next-gen GPU launch to Micron's HBM3E production schedule, the conversation isn't 'can you discount this batch.' It's 'what do we need to do to secure allocation.'

For context: SanDisk came in second at 34.2%, ASML third at 29.7%, Seagate at 21.6%, AMD last at 13.4%. AMD posting roughly a third of Micron's net margin tells you something important — AMD is positioned as the GPU challenger, but the actual pricing power in this cycle sits one layer down, in the memory.

Decomposing the 194% Revenue Growth

A tripling of revenue is a forecast that deserves scrutiny. Here's how I think it actually adds up.

Three forces stacking simultaneously. First, HBM ASPs are roughly 5-7x standard DRAM per wafer area. Same fab footprint, dramatically more revenue per square millimeter.

Second, HBM content per data center GPU keeps climbing. H100 to B100 to B200 — each generation adds more HBM stacks per accelerator.

Third, HBM demand isn't NVIDIA-only anymore. AMD MI300 and MI350, plus the custom silicon programs at the hyperscalers, all consume HBM. The combined backlog locks in multi-year visibility.

When these stack, 194% might actually be the conservative number.

The Weakness: Cash ROIC and FCF Margin

Micron ranked 4th and 5th on cash ROIC (14%) and levered FCF margin (17.7%). That's why it lost the overall fight to SanDisk.

I read this as a phase difference, not a structural weakness. Micron is mid-ramp on HBM3E and next-node DRAM. Capex is heavy, and quarterly cash generation gets compressed. ASML wins these metrics easily (FCF margin 26.6%, ROIC 40.1%) because its business is mature — capex is a smaller share of revenue.

The relevant question is when the capex cycle stabilizes. My best guess is 2027-2028, when ramp matures and the margin/cash-flow gap closes.

The 0.27 PE That Made Me Stop

Round 5's profit-adjusted PE is the data point I keep returning to. Micron 0.27. SanDisk 0.61. ASML 1.39. Seagate 2.42. AMD 3.89.

You're paying roughly 14x less per unit of profit for Micron than for AMD. Same AI infrastructure theme. Same hyperscaler customer base. Wildly different market valuation.

I want to be honest about the limits of PE comparisons. Memory is cyclical. AMD has a more diversified revenue mix. So a 14x ratio doesn't mean 14x of alpha is sitting there waiting.

But here's the question I keep asking myself: if Micron's net margin compresses from 41.5% back down to a 'normal' 30%, and revenue growth comes in at half the forecast — call it 100% — does this valuation still make sense? My math says yes. It says Micron is still cheap under significantly more conservative assumptions than the consensus.

How I'd Approach the Trade

Two markers I'm watching. First, the peak of the capex cycle — when capex-to-revenue starts declining. Second, actual HBM3E shipment data versus guidance.

Both confirm the same thing: when does cash ROIC and FCF margin start expanding? That inflection is the moment the market should re-rate the PE. Until then, Micron stays the most asymmetric setup in this group.