The NAND Pain Is Temporary. Everpure's Recurring Engine Isn't.

The NAND Pain Is Temporary. Everpure's Recurring Engine Isn't.

The market spent the last three months punishing Everpure for a supply chain problem it didn't create and won't control for much longer. NAND flash prices surged four to ten times since mid-2025, crushing product gross margins and dragging free cash flow from a 27.2% margin down to 10.6% in Q1. The stock dropped nearly 15% after that report, even though revenue jumped 35% and management raised full-year guidance.

The headlines read like a margin crisis. The recurring revenue base says something different.

Subscription ARR hit $2 billion, up 19% year over year. Remaining performance obligations climbed 41% to $3.8 billion. These are not temporary metrics. They're the locked-in cash flow that sits underneath the NAND noise.

The setup over the next 12 months is straightforward: NAND supply tightness is expected to ease as 2027 progresses, hyperscaler deals with 75%–85% margins begin contributing, and Everpure's FCF margin re-expands from the Q1 trough toward its historical range. The market is still pricing the compression quarter while the revenue bridge to margin recovery is already built.

The NAND Squeeze — Painful, Not Structural

CEO Charles Giancarlo described NAND spot prices doubling every 18 days, with overall input costs surging 300% to 900% since mid-2025. The supply chain disruption is driven by AI demand pulling fab capacity away from NAND production. TrendForce estimates a 4%–5% supply deficit for 2026.

Everpure absorbed the pain at the gross margin level — product gross margin fell to 65.5%, the low end of its 65%–70% long-term range — while keeping price increases below actual cost increases to protect market share. That discipline is why FCF margin collapsed: revenue kept growing, but each dollar cost more to deliver.

The key detail the selloff ignored is the timeline. TrendForce projects NAND supply growth outpacing demand in 2027, with the supply-demand balance turning positive in the second half of the year. This isn't a permanent structural problem. It's a 12-to-18-month squeeze that's already peaking.

What the Recurring Engine Tells You

The subscription and backlog numbers don't care about NAND spot prices. Subscription services revenue grew 17% in Q1 to $476 million. ARR crossed $2 billion with 19% growth. The Evergreen//One storage-as-a-service model — where customers lock in multi-year subscriptions and Everpure handles hardware refresh — grew 73% year over year, as customers actively sought predictable pricing amid hardware cost volatility.

Remaining performance obligations at $3.8 billion, up 41%, are the forward contract base that tells you what revenue is already committed. A 41% increase in RPO on a $3.8 billion base means roughly $1.1 billion of new contracted revenue flowed in during the last 12 months. That's the part of the business that NAND can't disrupt.

Hyperscaler Deals — The Margin Recovery Lever

The second hyperscaler design win, announced in early August with a supply agreement attached, is the margin recovery catalyst. CFO Tarek Robbiati said hyperscaler contracts are expected to carry 75%–85% gross margins, well above the current product mix. He also noted the company expects "a multiple of the revenues we generated in FY26" from hyperscalers in FY27.

Hyperscale product revenue was intentionally minimal in Q1 as onboarding worked through supply constraints. The ramp is expected in Q3 and Q4. If even a fraction of that guidance materializes, it pulls total product margins meaningfully higher and rebuilds the FCF margin that collapsed in Q1.

The stock rallied on the hyperscaler news in August and is up 46% year-to-date. But the forward path matters more than the tape.

The Valuation Question

Here's where the story gets harder. At a $36 billion market cap, Everpure trades at roughly 9.2 times trailing sales and 200 times EV/EBITDA. Trailing FCF of $516 million implies a 70x multiple. By any traditional measure, the valuation is stretched.

But Q1 FCF was a trough, not a baseline. The $112 million of FCF that quarter was dragged down by elevated commissions tied to Q4 overachievement, merit bonus timing, and heavy capex — CFO Robbiati called these timing items, not structural erosion. Operating cash flow that quarter was still $180 million. If FCF margin recovers toward even the low 20s as NAND eases and hyperscaler margins flow through, annual FCF at the $4.4–4.5 billion revenue guidance range would be $880 million to $1.15 billion. That's roughly 31x to 41x on current price — still premium, but in a range where a company growing revenue at 20%+ with accelerating recurring revenue isn't automatically dismissed.

The 14% stock drop after a 35% revenue beat and raised guidance is what makes this worth looking at. The market punished a cash-flow blip while ignoring the fact that the backlog just grew 41%. Expectations reset on the wrong metric.

The Q2 Test

Everpure reports fiscal Q2 results on August 27. Consensus revenue sits near $846 million. Management's Q2 guidance from the Q1 call ranged from $1.095–$1.105 billion — notably higher. The discrepancy likely reflects consensus that hasn't fully caught up to management's own outlook.

What matters in Q2: - Product gross margin direction. Even if it hasn't recovered to 70%, any stabilization or improvement above the Q1 trough would signal the NAND pain is peaking. - Subscription revenue momentum. ARR growth above 19% and Evergreen//One continuing its 73% pace would reinforce the recurring story. - FCF margin. Any bounce above the 10.6% trough would confirm Q1 was a trough, not a new floor. - Full-year guidance maintenance or lift. The $4.41–$4.51 billion revenue range and $820–$860 million non-GAAP operating income range are the commitments. Defending them is the minimum.

Wedbush raised its price target to $127 from $105 today, citing pricing and demand strength and expecting a potential beat. That $127 target implies confidence that margin recovery is already building beneath the surface.

The Risk That Matters

The thesis breaks if NAND normalization is delayed beyond the second half of 2027, forcing another quarter or two of compressed margins while the stock trades at 70x trough FCF. At this valuation, patience costs money. The second hyperscaler deal could also underperform on margin if hyperscalers negotiate harder pricing than management's 75%–85% range suggests.

Conversely, if Q2 shows margin stabilization and the backlog keeps compounding, the current price reflects the worst quarter of FCF, not the direction of travel. The inflection is that NAND's a temporary bottleneck on a business whose recurring revenue is accelerating faster than its hardware cost problem is growing.

Tripwire: If full-year guidance gets cut below the current $4.41 billion floor and product gross margin stays at or below the Q1 trough through Q2, the margin recovery narrative is delayed enough to stay on the sidelines. The thesis requires NAND to ease and hyperscaler revenue to show up. If neither happens by the next earnings window, the patience bid turns into a value trap.