MRAM Product +22% YoY With LEO Strength, Licensing Normalizes, Stock Reverses -17% on Crowded Long Positioning — Maintaining Hold

Results vs. Consensus

Year-over-Year Comparison

Quarter-over-Quarter Comparison

Revenue Assessment

The Q3 print revealed an important compositional shift in the revenue mix. MRAM product sales grew +22% YoY and +14% QoQ — the strongest sequential product growth in 4+ quarters, driven by LEO satellite + casino gaming + energy management strength layering onto the steady data-center / RAID franchise. This is the cleanest evidence yet that the company's design-win pipeline is converting to revenue at an accelerating pace. The 178 FY24 design wins are starting to ramp; the EM064LX HR / EM128LX HR engineering samples are progressing to production; the IBM FCM4 PERSYST 1Gb is contributing at a steady state; and the multi-customer LEO satellite ramp is becoming visible at the consolidated revenue level.

The licensing/royalty/patent/other line dropped to $1.4M from $2.1M Q2 — a 33% QoQ decline that warrants attention. The CFO's framing is structurally important: the line will "kind of continue to be around that range" going forward, "in the 10% to 15% range, probably more in the 10% range." Translation: don't model the Q2 spike as the run-rate. The Purdue contract has matured to steady-state delivery; QuickLogic continues on its ~2-year project arc; Frontgrade Phase 2 is contingent on AFRL renewal; TMR sensor foundry services remain ongoing but at modest scale. The structural licensing line is $1.4-1.6M per quarter, not $2.1M.

Margin Assessment

GAAP gross margin held at 51.3% — flat QoQ and +210bps YoY. The composition matters: with licensing declining as a percentage of total revenue (now 10% vs Q2's 16%), the implied product gross margin is improving. The CFO confirmed yield improvements driven by "process improvements developed in collaboration with our foundry partner" as a contributing factor — meaningful for the structural product margin trajectory. The cautionary framing remains: "we'll see — that's a good strong result for us. We do expect to kind of continue to be in that range overall" — i.e., 51-52% is a reasonable forward expectation but not a structural target.

OpEx of $8.8M is +1% QoQ on continued investment posture. The CFO's explicit "$7.5 million range going forward" comment on non-GAAP OpEx (vs $8.8M GAAP, with $1.3M SBC the bridge) confirms a disciplined cost structure with the SBC line as the main GAAP-to-non-GAAP reconciliation item. Q4 implied EBITDA margin works out to ~10-15% non-GAAP — consistent with the modest-but-improving operating leverage trajectory.

EPS & Cash Flow Assessment

Non-GAAP EPS of $0.06 vs guide of $0.02-0.07 lands at the high end for the second consecutive quarter, confirming the company's pattern of guiding conservatively and beating the midpoint by execution. The GAAP EPS of $0.00 (essentially breakeven; $54K net income) looks underwhelming versus the Q3 2024 $0.10 — but the YoY comparison is misleading. Q3 2024 included a step-up in DoD other income recognition that flowed directly to net income; Q3 2025's $1.2M is the ramp-into-Q4 step rather than a peak quarter. The CFO's explicit framing: "the decrease versus the year ago period was driven by lower other income stemming from lumpiness inherent in our DoD MRAM contract services as Q3 '24 required higher levels of activity upon initiation of the contract in that quarter."

Operating cash flow of $0.9M (down from $5.0M Q2) reflects AR timing on a distributor change rather than fundamental business weakness. The CFO explicitly attributed the sequential decline to "higher collections on receivables on a change in distributors" — meaning Q2 cash collections were front-loaded and Q3 is the normalization. The cash position of $45.3M (+$0.3M QoQ) confirms the business continues to self-fund operations + modest capex through cycles.

Segment Performance

MRAM Product Sales — $12.7M, +22% YoY, +14% QoQ (the print's standout metric)

MRAM product sales accelerated to +22% YoY — the highest YoY product growth rate in 5+ quarters. The CEO's framing identified three primary end-market drivers: (1) Low Earth Orbit satellite applications, where MRAM's reliability in harsh ambient + extreme temperatures is structurally differentiated; (2) Casino Gaming, a new strength callout this quarter where MRAM's persistent-data requirement aligns with regulatory compliance for game-state preservation; (3) Energy Management, an industrial application where MRAM's low-power persistent storage suits smart-grid and industrial-IoT deployments.

The LEO satellite framing is particularly important. The CEO disclosed: "LEO satellites have a short lifespan of 3 to 5 years, primarily due to atmospheric drag impacting the orbit. Design wins in this market with multiple Everspin MRAM parts per satellite is expected to translate into meaningful revenue for Everspin as this market grows." The structural implication: as the LEO satellite market scales (SpaceX Starlink, Amazon Kuiper, Astro Digital, Blue Origin, plus dozens of commercial constellation operators), Everspin's per-satellite content + replacement-cycle revenue compounds materially. The Astro Digital + Blue Origin design wins disclosed in Q2 are now starting to ship parts.

"Our performance this quarter was driven by strength across all products, specifically in Low Earth Orbital or LEO applications, Casino Gaming and Energy Management. In addition, our data center business remains strong with continued demand for our Toggle MRAM products for redundant array of independent disks or RAID from a broad selection of data center customers, including Dell, Supermicro and others." — Sanjeev Aggarwal, CEO

The IBM FlashCore Module 4 PERSYST 1Gb deployment continues at steady state — important for thesis stability but no longer a growth driver in 2025. The next leg from IBM (FCM5, disclosed in Q4 2025) is the bridge to incremental contribution in 2026.

Assessment: The product line at +22% YoY is structurally healthy. The diversification across LEO + casino gaming + energy management + data center reduces single-customer concentration risk. We model the product line at $48-52M for FY25 (vs FY24 ~$42M), implying +14-19% YoY growth — well above the consolidated growth rate because licensing normalization is netting against it. For FY26, $54-60M product revenue is plausible given the design-win pipeline + xSPI HR full production + early LEO satellite ramp.

Licensing/Royalty/Patent/Other — $1.4M, -18% YoY, -33% QoQ (structural rebase)

The licensing line declined sequentially as the Q2 Purdue ramp normalized to steady-state delivery and Frontgrade Phase 2 awaits AFRL funding renewal. The CFO's explicit framework: "we'll probably expect that to kind of continue to be around that range, right — kind of in the — again, we've been in that 10% to 15% range, probably more in the 10% range as we go forward." Translation: $1.4-1.5M per quarter is the structural run rate, not the Q2 $2.1M peak.

"These projects, right, and that kind of bucket of nonproduct revenue encompasses a lot of different things, including license revenue, any engineering service revenue, different foundry services that we provide as well. And so really on that particular piece of our revenue, that can be, I'll call it, somewhat lumpy as we go forward in time and even some of the initiatives that we have... those projects tend to be anywhere between 1 year to 18 months type — 2 years type projects typically. So they do wrap up." — Bill Cooper, CFO

The strategic value of the licensing portfolio remains intact — Purdue STT-MRAM for energy-efficient AI is progressing through development phases (the company "developed materials with higher percent MR and characterized devices using these new materials"); the QuickLogic AgILYST MRAM technology continues on its ~2-year arc; the TMR sensor foundry services project provides recurring contribution. But the financial expectations should reset.

Assessment: Licensing at 10% of revenue ($1.4M quarterly at Q3 run-rate) is the new baseline. Any reacceleration depends on new contract wins replacing expiring ones. The newly-announced Quintauris partnership (RISC-V-based reference designs) is one potential incremental contributor; the broader Microchip MPU ecosystem qualification (announced Q4 2025) is another. We model licensing at $5.5-6M for FY25 (down from Q2's implied $8M annualized rate), with FY26 contingent on new contract conversion.

DoD Sustainment Contract — $1.2M Q3 ($8.5M cumulative of $14.6M)

The DoD sustainment contract recognized $1.2M in Q3 (up from $0.8M Q2), bringing cumulative recognition to $8.5M of $14.6M (58%). Management reiterated the Q4 acceleration commitment: "We continue to expect this business to pick up meaningfully in the fourth quarter." With $6.1M of contract value remaining and Q4 typically the heaviest-activity quarter in the contract's design, Q4 recognition could reasonably land in the $3-4M range — the load-bearing forward EPS lever.

The contract's strategic significance compounds beyond the dollar value. The U.S. government's designation of Everspin as a domestic MRAM supply-chain priority sets up the broader portfolio of follow-on opportunities: the Microchip Foundry Services Agreement (announced later in Q4 2025) and the $40M defense-prime subcontractor agreement (announced in Q1 2026) are direct outgrowths of the DoD relationship that started with this contract.

Assessment: The Q3 step-up to $1.2M (from $0.8M Q2) is the early evidence that Q4 acceleration is real. If Q4 lands at $3.0-3.5M, the consolidated FY25 EPS picture brightens meaningfully — driving non-GAAP EPS toward $0.13 high-end of Q4 guide. The structural compounding of DoD relationships into broader defense / aerospace contracts is the multi-year tail.

Key Topics & Management Commentary

Overall Management Tone: Management was measured and consistent — the same calm, execution-focused posture as Q2. The CEO led with end-market strength (LEO + casino + energy management as new callouts) and the structural growth thesis (LEO satellite market is "expected to grow rapidly in the coming years"); the CFO was disciplined on gross margin, OpEx, and cash flow. The new content of the quarter was the explicit framing of the licensing line as ~10% of revenue going forward (resetting Q2 enthusiasm), the disclosure of the Quintauris partnership for RISC-V automotive applications, and the continued commitment to Q4 DoD ramp. No defensive callouts, no downbeat commentary — measured execution.

1. LEO Satellite — Structural Growth Vertical with Multi-Year Tailwind

The CEO's most expansive end-market commentary of the call addressed the LEO satellite opportunity. The market is "expected to grow rapidly in the coming years" with Everspin's MRAM "ideally suited" for the deployments. The structural attributes: short lifespan (3-5 years) creates replacement-cycle revenue; multiple parts per satellite creates per-unit content density; harsh ambient + extreme temperature requirements favor MRAM over alternative memory technologies.

"The LEO satellite market is expected to grow rapidly in the coming years. Everspin MRAM with its reliability at extreme temperatures and harsh ambient is ideally suited for these deployments. As mentioned in our last call, we are seeing good traction in this market with announced design wins with Astro Digital and Blue Origin. It is our understanding that LEO satellites have a short lifespan of 3 to 5 years, primarily due to atmospheric drag impacting the orbit. Design wins in this market with multiple Everspin MRAM parts per satellite is expected to translate into meaningful revenue for Everspin as this market grows." — Sanjeev Aggarwal, CEO

The PERSYST EM064LX HR and EM128LX HR (the high-reliability xSPI STT-MRAM variants announced in Q1 2025) are being engineering-sampled to multiple LEO satellite customers in Q3, with full-production target of Q4 2025. The structural visibility into FY26 LEO revenue is improving — the design-win-to-production cycle is on track.

Assessment: LEO satellite is the longest-runway revenue lever in Everspin's portfolio. We model the contribution at $1-3M in FY25 (engineering samples + early production); $5-10M in FY26 (full production ramp); $15-25M by FY28 if the satellite-market growth projections hold. The lever is not currently material but is the bull-case driver for the 3-5 year revenue acceleration story.

2. Casino Gaming and Energy Management — New End-Market Callouts

Casino Gaming and Energy Management were called out for the first time as explicit Q3 strength contributors. Casino Gaming is a niche specialty market where MRAM's persistent storage requirement aligns with gaming regulators' game-state preservation requirements — every spin/hand must be auditable post-power-loss. Energy Management is a broader industrial application encompassing smart-grid, industrial-IoT, and predictive-maintenance applications where MRAM's low standby power + persistence + endurance combine attractively.

Neither was framed with quantified contribution dollars, but both were called out as material Q3 drivers — suggesting collectively several hundred K of incremental revenue. The strategic value is end-market diversification: as MRAM penetration expands beyond the historical core (industrial automation PLCs, RAID memory), Everspin's revenue mix becomes less cyclically sensitive to any single end-market.

Assessment: The end-market expansion is structurally healthy. Each new vertical (LEO, casino, energy management, automotive, defense) represents a 10-25%+ growth-rate market with MRAM-specific technical advantages. We expect Q4-Q1'26 disclosures to continue surfacing new vertical strength callouts as the design-win pipeline matures.

3. Licensing Line Normalizes — "10% of Revenue" Steady State Framework

The CFO's most explicit forward-framing of the quarter addressed the licensing/royalty/patent/other revenue line. Q2's $2.1M was the upside outlier driven by Purdue ramp; Q3's $1.4M is the steady-state baseline. The CFO's framework: "we've been in that 10% to 15% range, probably more in the 10% range as we go forward."

"On that particular piece of our revenue, that can be, I'll call it, somewhat lumpy as we go forward in time and even some of the initiatives that we have that — where we do work for other groups that we've mentioned in the past, those projects tend to be anywhere between 1 year to 18 months type — 2 years type projects typically. So they do wrap up, right? And then on the second part of your question, I think as we see sort of this level of product — nonproduct revenue, we'll probably expect that to kind of continue to be around that range." — Bill Cooper, CFO

The implication for valuation: any Street model that extrapolated Q2's $2.1M as the run rate is overestimating FY26 revenue by $2-3M annualized. The structural product growth (15-22% YoY) does not get amplified by licensing growth at the consolidated level — total revenue growth tracks product growth.

Assessment: The licensing normalization is the most important Street-modeling reset of the call. We model the line at $5.5-6M for FY25 (vs Q2-annualized $8M); $5-7M for FY26 contingent on new contract wins (e.g., Frontgrade Phase 2, Microchip MPU ecosystem revenue, new Quintauris-related deliverables).

4. Quintauris Strategic Partnership — RISC-V Automotive Reference Designs

Everspin announced a strategic collaboration with Quintauris (a joint venture of major semiconductor companies focused on RISC-V automotive architectures) for next-generation reference designs combining MRAM with RISC-V cores. The partnership focuses on automotive, industrial, and edge applications where data persistence + integrity + low latency + security are critical.

"As we announced last month, we entered into a strategic collaboration with Quintauris to strengthen the reliability and safety of RISC-V-based platforms with our MRAM offerings. This partnership is focused on automotive, industrial and edge applications where data persistence, integrity, low latency and security are critical. The goal is to jointly build reference designs that would lay the foundation for scalable, reliable platforms for these applications." — Sanjeev Aggarwal, CEO

The RISC-V automotive ecosystem is in early development but has structural tailwinds — open-standard ISA, no licensing fees, growing adoption among automotive OEMs seeking alternatives to ARM-based architectures. Quintauris's joint-venture structure (with backing from major semis including likely Bosch, NXP, Infineon, others) provides Everspin with broad ecosystem exposure.

Assessment: The Quintauris partnership is strategically important but financially immaterial in FY25-26. The reference-design phase is multi-quarter; OEM design-in is multi-year. We treat this as a 2027+ revenue optionality lever, not a near-term contributor.

5. Gross Margin Sustainability at 51%+

The opening analyst question addressed the structural sustainability of the 51-52% gross margin range — three consecutive quarters above 51% despite the CFO's "45-50% range" framing. The CFO's response acknowledged yield improvement initiatives and factory utilization benefits but stopped short of committing to 51%+ as the new structural level.

Q: "Three quarters in a row now of non-GAAP gross margin over 52%. Just curious, I guess, how sustainable you think this is going forward?"
— Neil Young, Needham & Company

A: "I think we saw some improvement this quarter, Neil, on the gross — the product gross margin specifically as well based on some of our yield improvement initiatives and our factory utilization. But to answer your question directly, I think we'll see — that's a good strong result for us. We do expect to kind of continue to be in that range overall."
— Bill Cooper, CFO

The "in that range overall" framing is deliberately ambiguous — it doesn't commit to 51% as the new floor or signal compression risk. The implication: don't model further expansion, but don't model compression either. We model FY25 gross margin at 50.5-51% (slight decline from Q3 as the year completes) and FY26 at 50-52% range.

Assessment: Gross margin sustainability at 51%+ would be a meaningful positive surprise to the structural margin model. The CFO's measured framing protects against over-expectation but the data is the data — three consecutive quarters at 51%+ on product mix that includes lower-margin Toggle MRAM. If FY26 sustains this level, the EBITDA-margin profile expands materially.

6. OpEx Discipline Holding

The CFO confirmed non-GAAP OpEx of ~$7.5M was the steady-state run rate (GAAP at $8.8M with $1.3M SBC the bridge). The framework: "we continue to sort of manage through on OpEx... we've sort of indicated that we are going to continue to sort of move toward product development type costs. But for Q4, you're going to see a lot of consistency."

Q: "On OpEx. So it was flat again in the quarter on a non-GAAP basis. Should we sort of assume that it stays in that $7.5 million range going forward?"
— Neil Young, Needham & Company

A: "Yes, Neil, that's a safe assumption. We're going to — we continue to sort of manage through on OpEx, and we've been pretty consistent throughout this year. Now again, we've sort of indicated that we are going to continue to sort of move toward product development type costs. But for Q4, you're going to see a lot of consistency."
— Bill Cooper, CFO

Assessment: OpEx discipline at the $7.5M non-GAAP / $8.5-8.8M GAAP quarterly level is the structural foundation for operating leverage. With revenue growing 7%+ QoQ and OpEx essentially flat, the incremental margin on revenue growth flows to EBITDA. This is the cleanest operational-discipline signal of the call.

7. DoD Contract Q4 Acceleration — "Continue to Expect Meaningful Pick-Up"

The CEO reiterated the Q4 acceleration commitment on the DoD MRAM sustainment contract: "We continue to expect this business to pick up meaningfully in the fourth quarter." Q3 recognized $1.2M (vs $0.8M Q2), suggesting the step-up curve is starting. Cumulative recognition at $8.5M of $14.6M (58%) leaves $6.1M of remaining contract value. If Q4 lands at $3.0-3.5M and Q1 2026 at $1.5-2.0M, the contract completes in mid-2026.

"With respect to below-the-line items, we recognized $1.2 million in other income in the third quarter and $8.5 million to date from the $14.6 million contract we have with a DoD contractor to develop a sustainment plan for our MRAM manufacturing facilities to provide continuous onshore MRAM capabilities to their aerospace and defense customers. We continue to expect this business to pick up meaningfully in the fourth quarter." — Sanjeev Aggarwal, CEO

Assessment: The Q4 DoD ramp is the central FY25 EPS lever. The Q3 step-up validates the trajectory; Q4 execution validates the magnitude. We model Q4 DoD at $3.0-3.5M; if it lands above $3M, Q4 non-GAAP EPS lands at $0.10-0.13 (within guide); if below $2M, EPS lands at $0.06-0.08 (low end of guide).

8. IBM FCM4 Steady State + Lucid Gravity Continued Shipments

The IBM FlashCore Module 4 PERSYST 1Gb STT-MRAM deployment continued at steady-state contribution in Q3. The CEO reiterated: "We continue to anticipate revenue from this project to remain at this level for the remainder of the year." This confirms the IBM contribution is not a growth driver in 2025; the next leg from IBM (FCM5, FCM6) is the 2026 catalyst.

Lucid Motors Gravity SUV PERSYST MRAM shipments continued in Q3 with expected volume increases as Lucid ramps Gravity production. No specific revenue contribution disclosed; the strategic-validation value (automotive OEM-spec design win) exceeds the dollar value.

Assessment: Both IBM and Lucid are validation references rather than near-term growth drivers. We model both as flat through FY25 with modest growth in FY26 as Lucid Gravity production ramps and IBM FCM5 contribution begins.

Guidance & Outlook

The Q4 guide implies sequential revenue growth of $0.4M (+3% at midpoint) — modest by face but consistent with management's pattern of guiding conservatively to mid-range and executing toward the high end. The composition: continued product growth ($0.5-1M sequential) offset by licensing normalization ($0.1-0.3M sequential decline). The DoD contract Q4 acceleration is the load-bearing EPS lever — if recognized at $3M+, non-GAAP EPS lands in the upper half of the $0.08-0.13 range.

Implied FY25 trajectory: Q1 $13.1M + Q2 $13.2M + Q3 $14.1M + Q4 ~$14.5M = $54.9M FY25 revenue (+7.6% YoY). Non-GAAP EPS ~$0.21 FY25 (vs ~$0.10 FY24). Street at: Consensus is thin given limited coverage; we model Q4 at $14.7M and non-GAAP EPS at $0.10-0.11. Guidance style: Pattern continues — conservative guide; execution toward high end; +$0.2-0.5M revenue beat vs guide midpoint typical.

Analyst Q&A Highlights

Gross Margin Sustainability at 51%+

The opening question of Q&A addressed the structural gross margin question — three consecutive quarters above 51% despite the CFO's prior framing of 45-50% as the structural range. The CFO acknowledged yield improvements and factory utilization benefits but maintained a measured forward outlook.

Q: "First question, so three quarters in a row now of non-GAAP gross margin over 52%. Just curious, I guess, how sustainable you think this is going forward?"
— Neil Young, Needham & Company

A: "I think we saw some improvement this quarter, Neil, on the gross — the product gross margin specifically as well based on some of our yield improvement initiatives and our factory utilization. But to answer your question directly, I think we'll see — that's a good strong result for us. We do expect to kind of continue to be in that range overall."
— Bill Cooper, CFO

Assessment: The "in that range overall" deliberately doesn't commit to 51% as the new floor. The implication is that yield improvements are real but mix shifts (less licensing, more product) could offset them. We model FY25 GM at 50.5-51% — a slight decline from Q3 as licensing share moderates.

Licensing Line Normalization and Forward Steady State

The follow-up question pressed on the sequential decline in licensing/royalty/patent/other revenue — the most-watched metric of the quarter given Q2's spike. The CFO's framework reset Street expectations: "10% range" as the new normal, with project-based lumpiness creating quarterly variability around that mean.

Q: "The sequential decline in licensing, royalty, patent and other, I know you talked about it a little bit, but I was just hoping you can maybe provide some more detail just on the sequential decline and then sort of, if possible, where you think that is going in 4Q?"
— Neil Young, Needham & Company

A: "These projects, right, and that kind of bucket of nonproduct revenue encompasses a lot of different things, including license revenue, any engineering service revenue, different foundry services that we provide as well. And so really on that particular piece of our revenue, that can be, I'll call it, somewhat lumpy as we go forward in time and even some of the initiatives that we have that — where we do work for other groups that we've mentioned in the past, those projects tend to be anywhere between 1 year to 18 months type — 2 years type projects typically. So they do wrap up, right? And then on the second part of your question, I think as we see sort of this level of product — nonproduct revenue, we'll probably expect that to kind of continue to be around that range, right — kind of in the — again, we've been in that 10% to 15% range, probably more in the 10% range as we go forward."
— Bill Cooper, CFO

Assessment: The CFO's "10% of revenue" framing is the structural reset. The Q2 $2.1M was the outlier; $1.4M is the new baseline; quarterly variability of $0.3-0.5M around that level is expected. Street models that extrapolated Q2 must rebase. This is the most important Street-modeling reset of the call.

OpEx Discipline — $7.5M Non-GAAP Steady State

The follow-up question confirmed OpEx discipline — non-GAAP OpEx in the $7.5M range as the steady state. The CFO confirmed the assumption and noted Q4 will show "a lot of consistency" with prior quarters.

Q: "Just a question on OpEx. So it was flat again in the quarter on a non-GAAP basis. So similar, should we sort of assume that it stays in that $7.5 million range going forward?"
— Neil Young, Needham & Company

A: "Yes, Neil, that's a safe assumption. We're going to — we continue to sort of manage through on OpEx, and we've been pretty consistent throughout this year. Now again, we've sort of indicated that we are going to continue to sort of move toward product development type costs. But for Q4, you're going to see a lot of consistency."
— Bill Cooper, CFO

Assessment: OpEx discipline at $7.5M non-GAAP / $8.5-8.8M GAAP is the structural foundation for operating leverage. With revenue growing 7%+ QoQ and OpEx flat, every $1M of revenue growth flows ~50%+ to non-GAAP operating income. This is the structural operating-leverage engine.

What They're NOT Saying

1. FY25 revenue or EPS guidance: Management continues not to provide formal FY25 guidance — only quarterly. The implicit FY25 framework (~$55M revenue, ~$0.20 non-GAAP EPS) is derivable from Q4 guide + Q1-Q3 actuals.
2. FY26 revenue trajectory: No formal commentary on FY26 — management defers all forward-year framing. Design-win pipeline and DoD contract relationships imply FY26 acceleration but are not quantified.
3. Specific LEO satellite revenue contribution: Astro Digital + Blue Origin design wins disclosed; no per-customer or per-program revenue contribution.
4. IBM FCM4 revenue dollar value: Steady-state confirmed; no dollar disclosure. We continue to estimate $1.5-2.5M per quarter.
5. Lucid Motors Gravity SUV per-vehicle MRAM content: No disclosure; estimated at $5-20 per vehicle.
6. Q4 DoD contract specific dollar target: "Meaningful pick-up" maintained; no dollar target. We infer $3-3.5M based on the Q3 step-up trajectory.
7. Quintauris partnership revenue model: Strategic collaboration disclosed; no licensing structure or per-design-win economics.
8. Design-win count update: No 2025 YTD design-win count disclosed; FY24's 178 was the most recent number. Management is deliberate in not surfacing the metric quarterly — likely because individual quarter swings are noisy.
9. Frontgrade Phase 2 status: No update on AFRL funding renewal for the contingent Phase 2 of the Frontgrade radiation-hard MRAM project. The omission is itself a soft negative signal.
10. NOR Flash supply-chain pressure: Not surfaced on this call. The industry NOR Flash → DRAM capacity reallocation dynamic (later disclosed in Q4 2025) exists but management didn't reference it yet.

Market Reaction

• Pre-print setup (Nov 4 close): $9.85, +53.7% YTD from $6.41 (Jan 2, 2025), +63% from Q2 print T+1 close $6.03. The stock had nearly doubled between the Q2 and Q3 prints. 30-day pre-print -11.0% (from $11.07 Oct 6), suggesting a modest pullback into the print as some positioning had reduced. Average daily volume ~200-300K (elevated from Q2's 80-100K), reflecting growing institutional interest as the design-win + LEO + DoD narrative built.
• Day-of action (Nov 5 — print released after-market): Regular hours close $10.53 (+6.69%) on volume of 364K (~1.5x trailing avg). The +6.7% gain into the print reflects buying anticipation; after-hours initial reaction was modestly positive on revenue in-line + non-GAAP EPS at high end. Post-call drift turned negative as the licensing-line decline and the Q4 EPS-guide-midpoint setup were digested.
• T+1 (Nov 6): Open $9.50 (gap DOWN -9.8% from Nov 5 close); intraday high $9.76; intraday low $8.30; close $8.72 (-17.2% from Nov 5; -8.21% reported). Volume 564.7K shares (~3x trailing 30-day average) — the highest single-day volume since Aug 7 Q2 print T+1.
• T+2 (Nov 7): Open $8.62; close $9.01 (+4.52%); volume 270K. Partial recovery on bargain-hunting; failed to reclaim the pre-print $9.85.
• Sell-side reaction: Needham (Neil Young) maintained Hold; no PT change. Craig-Hallum did not engage on the Q3 call. Sell-side coverage materially unchanged.
• Peer reaction: Semiconductor sector flat to modestly up Nov 6 on broader chip tape; MRAM did not track sector. The stock traded as an idiosyncratic name.

The Q3 reaction pattern is the structural feature of a microcap with crowded long positioning entering an in-line print. The +63% stock run from Aug 7 to Nov 4 had embedded high expectations on both licensing strength and DoD acceleration; the licensing normalization (-33% QoQ) was the proximate trigger for institutional repositioning. The -17% intraday move on 3x volume is consistent with several institutional accounts trimming or exiting, not a fundamental thesis break.

The T+2 partial recovery to $9.01 (still -8.5% from pre-print) signals the market has settled into a new equilibrium pricing the licensing normalization while preserving the structural bull case. We do not view this as thesis-changing — the operational data is consistent with execution; the price action is positioning-driven.

Street Perspective

Debate: Is the licensing line normalization a structural ceiling or a Q3 one-off?

Bull view: The Q3 $1.4M licensing line is the temporary trough between the Q2 Purdue ramp and Q4-Q1 2026 expected contract additions. As Quintauris ramps, Frontgrade Phase 2 potentially renews, Microchip-related licensing contributes, and new DoD-adjacent contracts emerge, the line can return to $1.8-2.2M quarterly by FY26 H2. Plus the strategic value of licensing IP (validation, customer relationships, defense-supply-chain positioning) far exceeds the dollar value.

Bear view: The CFO has explicitly framed 10% of revenue ($1.4-1.5M quarterly) as the steady-state — that's the new model baseline. Project-based licensing is structurally lumpy with frequent contract completions and contingent renewals (Frontgrade Phase 2 depends on AFRL funding, which is outside Everspin's control). The Q2 $2.1M is the upside outlier; modeling above $1.5M quarterly creates earnings-disappointment risk.

Our take: The bear case is more probable for FY25-26. Licensing at $5.5-6M annualized is the modeling base; upside to $7-8M is contingent on multiple discrete contract conversion events that are individually low-probability. We don't assume the Q2 spike repeats.

Debate: Does the Q3 stock reversal signal thesis weakness or positioning normalization?

Bull view: The -17% T+1 move is positioning-driven, not fundamentals-driven. Q3 product growth +22% YoY, gross margin held at 51.3%, and DoD Q4 acceleration on track are all positive signals. The stock had run +63% from Q2 print into Q3 — too much, too fast. The reversal is the necessary positioning reset that creates the next leg-up opportunity if Q4 delivers on DoD acceleration.

Bear view: The -17% T+1 reaction reflects emerging concerns that institutional investors are seeing more comprehensively — the licensing normalization is the surface; deeper concerns include design-win-to-revenue lag, microcap liquidity limits, and the structural difficulty of MRAM achieving cost competitiveness with NOR Flash at scale. The Q3 stock action is the early warning signal of a multi-quarter derate.

Our take: Probability is ~65/35 positioning vs fundamentals. The product growth trajectory + DoD ramp signal + gross margin stability supports the bull positioning view. The licensing normalization is real but doesn't structurally break the thesis. We are at Hold not Outperform because the upside-vs-downside math at $9 is closer to symmetric than at $6 (when we initiated coverage).

Debate: Can Everspin sustain $0.10+ non-GAAP EPS through Q4 + FY26?

Bull view: The Q4 non-GAAP EPS guide of $0.08-0.13 with DoD acceleration provides the visibility line. The high end of $0.13 implies $3M+ Q4 DoD recognition + product growth + gross margin hold + OpEx discipline. Annualizing, that's $0.30-0.40 FY26 EPS run rate — implying current valuation at ~25-30x forward earnings, reasonable for a 15-20% revenue-growth specialty-semi.

Bear view: The Q4 EPS guide is meaningfully back-half-loaded — without the DoD Q4 ramp, the non-GAAP EPS lands at $0.05-0.08. FY26 EPS depends on continued DoD contract recognition (the $14.6M contract completes in mid-2026), then a gap until follow-on contracts (the eventual $40M defense agreement announced in Q1 2026) ramp. The FY26 EPS picture has a "DoD gap" that the bull case glosses over.

Our take: Q4 non-GAAP EPS at $0.10-0.12 is achievable. FY26 EPS at $0.25-0.35 is plausible if (a) DoD follow-on contracts ramp in H1, (b) product growth sustains 15%+, (c) gross margin holds 50%+. Below $0.20 FY26 EPS would be a meaningful downside surprise.

Model Update Needed

Valuation impact: At Nov 6 close of $8.72 and ~23.1M diluted shares, market cap is ~$201M; cash $45.3M, debt $0, EV ~$156M. On FY25E revenue of $54.5M and non-GAAP EPS of $0.20 ($4.6M net income), MRAM trades at 2.9x EV/Sales and ~34x EV/NI. On FY26E revenue of $62M and non-GAAP EPS of $0.25 ($5.8M net income), MRAM trades at 2.5x EV/Sales and 27x EV/NI. The valuation has expanded materially from the Q2 print (1.7x EV/Sales, $6 stock) reflecting the +50% run on positive structural narrative. The current multiple is fair but no longer cheap. Our 12-month fair-value estimate: $9-11 (3-26% above current); the range tightens as the structural valuation appropriately compresses to reflect the de-risked thesis.

Thesis Scorecard Post-Earnings

Overall: Q3 was a "confirmation-not-acceleration" quarter. The MRAM product growth at +22% YoY is the standout positive; the licensing normalization is the structural reset to expectations. The DoD contract Q4 acceleration remains the central forward EPS catalyst. The Q4 print on Mar 4, 2026 is the structural setup test.

Action: Maintaining Hold. Q3 doesn't move the needle on the rating — the operational picture is fine, the licensing line is structurally smaller than Q2 implied, and the stock has run +50% from Q2 print levels. Position sizing remains conservative given microcap liquidity. Upgrade trigger: Q4 prints >$15M revenue with non-GAAP EPS >$0.13 (above guide high end), DoD recognizes >$3M in Q4, AND FY26 guide implies >15% revenue growth. Downgrade trigger: Q4 prints below $13.5M revenue OR DoD Q4 acceleration falls short of $2M OR gross margin compresses below 48%.