NVE
NVE makes tiny magnetic sensors and data couplers in its own Minnesota factory, for heart implants, factory robots and industrial equipment.
The scores
Growth3/4 →
Revenue sat near $26M a year for a decade, rose 81% in June's quarter, then fell 47% in September.
Business quality3/4 →
Implant sensors are locked in by FDA approval and margins near 80% hold, but industrial parts face Allegro and Analog Devices on price.
Financial position4/4 →
$43.9M of cash and bonds, no debt and $2.3M of liabilities; the only drain is a dividend above earnings.
Execution record3/4 →
Margins near 60% for years and a finished fab expansion, but revenue did not grow for a decade and the chief executive just changed.
Per share capture4/4 →
The share count has not moved in six years and the dividend hands shareholders about everything the business earns.
Potential2/4 ↑
Heart implants and factory robots are large markets, but NVE's parts are a tiny slice of each and new customers start slowly.
Risk profile3/4 →
Abbott is 37% of sales on an agreement ending in December 2027, and defense swings; the balance sheet absorbs a bad year.
Valuation3/4 ↑
18.5 times trailing earnings against a six-year median of 20.6, and about 26 times the September quarter run for a year.
Evidence quality3/4 →
Results, the Abbott agreement and FDA records are public; the defense split and new-product volume rest on management's word.
The Verdict
We rate NVE neutral. It is a strong business: about 80 cents of each sales dollar is left after production costs, and its sensors inside heart implants are hard to replace because changing them needs the regulator's approval. Sales fell 47% in September, back to $5.8M, after June's $11.0M, and twice before a big quarter faded. But the shares have fallen 33% since the results to $68.36, about 18.5 times the last year's profit, near their usual level, so the price no longer counts on June returning.
The long caseRobots and implants need tiny, reliable magnetic sensors, and NVE makes some of the smallest in its own fab.
The evidence under this - 6 claims
- Abbott filed a 2025 supplement changing a magnetic sensor process in an ablation catheter 'to support more flexible sourcing', so implant makers do qualify second sources when they choose to.
- Customer A's share rose only from 35% to 37% in a year when product sales rose 2.4%, so the barrier holds the business without growing it.
- On the six months' earnings of $1.97 a share run for a year the price is about 17 times, and the dividend yield is near 5.9%, so the price is cheap if September was only a timing change.
- Costs are small and mostly fixed, so a recovery in revenue lifts profit faster.
- 2026-10-21editedre-struck at the 20 October close and the trailing earnings to September 2026
- 2026-10-23editedre-struck at the 22 October close: 18.5 times trailing earnings, from 27.7 at $102.05, after the shares fell 33.7%
The company
NVE makes sensors that notice a magnet and parts that pass data between circuits without a wire joining them. Both use films a few atoms thick, made in its own Minnesota clean room.
Heart implant makers use the sensors to switch a device into a safe mode when a doctor holds a magnet over it. The older choice, a reed switch, is a tiny mechanical contact that can stick: in 2024 one maker corrected 81,606 nerve stimulators for that fault and moved to a sensor like NVE's.
The evidence under this - 4 claims
No customer for the new sensors has been named and none appears in a filing.
How it makes money
Nearly all sales, 95.8% in fiscal 2026, are parts sold to device makers directly, through distributors and under partners' brands; the rest is paid research, mostly for defense agencies.
Because the films are made once and the factory is small, about 80 cents of each sales dollar is left after production costs. Last year NVE paid shareholders $4.00 a share, more than it earned, and the share count has not changed in six years.
The evidence under this - 4 claims
Where the money comes from, stream by stream
| Stream | Share of sales (FY 2026) | How it earns it |
|---|---|---|
| Product sales | 95.8% | Sensors, couplers and power parts sold to device makers directly, through distributors, and under partners' brands. |
| Contract research and development | 4.2% | Paid development work, mostly for government and defense agencies. |
Who runs it
Peter Eames, at NVE since 2003 and head of research and defense products, became chief executive in August 2026. Daniel Baker, chief executive since 2001, stayed as chairman.
Directors and officers own 2.1%. After the jump, Baker sold 48,371 shares outside any pre-set plan. Pay is mostly salary plus a small share of operating profit and of its increase.
The evidence under this - 4 claims
Why it wins, and who could beat it
Why customers come back
In a heart implant a sensor is hard to swap out. US rules require the regulator to approve any change to a component that could affect the device's safety, before it is made. Abbott filed for approval just to change the test range of the sensor in its pacemakers in 2009.
So once NVE's part is inside an approved device, a rival must persuade the maker to pay for testing and a new filing. That is why Abbott, about 37% of sales, has kept buying since 2006.
The evidence under this - 6 claims
The concentration note calls the customer 'Customer A' and never names it, and the risk factor names 'certain distributors' beside Abbott, so a distributor could in principle be the 37%.
- Abbott filed a 2025 supplement changing a magnetic sensor process in an ablation catheter 'to support more flexible sourcing', so implant makers do qualify second sources when they choose to.
- Customer A's share rose only from 35% to 37% in a year when product sales rose 2.4%, so the barrier holds the business without growing it.
The threat
Outside medicine the barrier is lower. Allegro MicroSystems sells sensors built on the same physics, and Analog Devices sells data couplers with the same pitch. Both are many times NVE's size. NVE's own annual report says competition is rising, prices are under more pressure and rivals may be closing its performance lead in factory equipment.
The evidence under this - 3 claims
- Gross margin recovered to 81.3% in Q1 fiscal 2027 while volume rose 82%, so the pricing pressure did not show then.
- Fiscal 2026 product sales rose partly on price increases, which a business with no pricing power could not make.
The rivals: ALGM, ADI, LFUS.
Who else does this, rival by rival
| Ticker | Rival | How far along | Priced against it |
|---|---|---|---|
| ALGM | Allegro MicroSystems XtremeSense TMR sensorsPriced against it: Different product | On sale | Different product |
| ADI | Analog Devices iCoupler isolatorsPriced against it: Different product | On sale | Different product |
| LFUS | Littelfuse reed switchesPriced against it: Different product | On sale | Different product |
How big it could get
How big this could get
NVE sells into two large markets. A world survey counted about a million pacemakers and 328,027 defibrillators implanted in 2009, each a possible home for a sensor. Factories installed more than 600,000 industrial robots in 2025, and each robot hand and joint can use several sensors.
No one publishes how many magnetic sensors those markets buy a year, so the size of NVE's slice cannot be checked from outside.
The evidence under this - 2 claims
Who has signed
Abbott buys under a supply agreement that began with St. Jude Medical in 2006 and has been amended 12 times. It runs to 31 December 2027. Every other customer orders as it goes.
The evidence under this - 2 claims
The results
The numbers
Sales grew 24% over the twelve months to Jun 2026, faster than the 1.8% in the year to Mar 2026.
Yearly sales growth has ranged from −22% to +42% over the last 4 years, with sales falling in 2 of them.
The share count barely moved in 4.3 years.
All the numbers
Filed figures from the company's 10-K and 10-Q, read 2026-10-21. Fiscal years end in March.
| Mar 2022 | Mar 2023 | Mar 2024 | Mar 2025 | Mar 2026 | 12 mo to Jun 2026 | |
|---|---|---|---|---|---|---|
| Size and growth | ||||||
| Sales | $27.0M | $38.3M | $29.8M | $25.9M | $26.3M | $31.3M |
| Sales growth | - | +42% | −22% | −13% | +1.8% | +24% |
| Profit | ||||||
| Gross margin | 77% | 79% | 77% | 84% | 79% | - |
| Operating margin | 61% | 67% | 62% | 62% | 60% | - |
| Net margin | 54% | 59% | 57% | 58% | 58% | - |
| What growth costs, per sales dollar | ||||||
| Research and development | 11% | 6.8% | 9.2% | 14% | 12% | - |
| Selling, general and admin | 5.4% | - | - | - | - | - |
| Pay in shares | 0.3% | - | - | - | - | - |
| Capital spending | 1.8% | 2.4% | 0.1% | 4.9% | 8.3% | - |
| Cash | ||||||
| Cash from operations | $12.5M | $19.1M | $18.2M | $14.3M | $16.7M | - |
| Free cash flow | $12.0M | $18.2M | $18.2M | $13.1M | $14.5M | - |
| Balance sheet | ||||||
| Cash and short-term investments | $31.3M | $17.2M | $22.2M | $21.7M | $19.8M | $21.8M |
| Returns and shares | ||||||
| Return on capital | 21% | 33% | 23% | 21% | 23% | - |
| Shares (diluted) | 4.84M | 4.83M | 4.84M | 4.84M | 4.84M | 4.85M |
What growth costs
Growth costs NVE very little. Research and administration were about 18% of last year's sales, and in Q1 fiscal 2027 costs rose 50% while sales rose 81%. More sales therefore turn into more profit at a higher rate, and the reverse holds if sales fall back.
The evidence under this - 3 claims
The last report (22 Jul)
Quarter ended June 2026, filed figures.
In the three months to 30 September 2026 sales fell to $5.83M, 47% below June's $11.03M and 8% below a year earlier. NVE says one customer rescheduled orders because its stock needs changed for now, and that defense sales fell. It gives neither the customer nor the sizes.
Gross margin held at 78.2%, but operating profit fell to 54.9% of sales. Net income was $0.65 a share, and the $1.00 dividend again exceeded the quarter's earnings.
On the call, management said orders were pulled forward into June and the second half should match the first, about $8.4M a quarter. It tracks no backlog.
What it did to the thesisIt failed the test we set and fell below the line that would make us doubt the case. The business is intact, but the price counted on June lasting. We cannot yet tell a pause in orders from a fade.
The quarter in detail
Sales growth from the previous quarter, filed figures.
Earnings per share $1.32; $1.32 as filed, with no adjustments.
The evidence under this - 5 claims
Six-month revenue is still up 35.5%, and the 10-Q calls the fall a timing change at one customer, which would let sales recover.
- 2026-10-21editedthe October print answered the open question: the jump did not repeat
- 2026-10-22editedthe call gave a second cause, a pull-forward of orders into June, and a second-half outlook
- The 10-Q gives a different cause, a customer's temporary change in inventory needs, and the transcript's '95% decrease in defense product sales' appears in no filing, so the size of each cause is still unstated (S20, S23).
- With no backlog tracked, the second-half forecast cannot be checked against orders on hand (S23).
What management did with the money
NVE spent about $3.4M over two years expanding its factory, finished in March 2026, and pays almost everything else out. Last year's $19.35M of dividends came partly from maturing bonds, since it was more than profit.
Insiders in 2026Q3: 11 purchases and 18 sales, from their Form 4 filings.
The evidence under this - 2 claims
Capital spending fell to $57K in Q1 fiscal 2027, and how much capacity was added is not disclosed in any filing; 'doubled' was an analyst's word.
The price, and where we disagree
In short. Fair. The price no longer needs June's level to return, only a recovery to the first half's.
The shares rose 55% the day after the June results. After the September results they fell 33.7% in a day, to $68.36 on 22 October, from $103.07. Most of the year's gain came from buyers paying more for each dollar of sales, and much of it has now gone. The retiring chief executive sold most of his shares into the rise, and short sellers hold 11.6% of the shares available to trade.
The next test is December's quarter, due about 20 January: it says whether September was a pause or the fade.
The evidence under this - 4 claims
- 2026-10-21editedre-struck at the 20 October close before the September-quarter results
- 2026-10-23editedre-struck at the 22 October close, the first session after the September results: the shares fell 33.7%
How we read the multiple
A buyer now pays roughly for the old normal.
1. Over the last four quarters NVE earned $3.69 a share. 2. The shares cost about 18.5 times that, against a usual 21 over six years. 3. Run September's quarter for a year and the price is about 26 times.
So the price is fair if sales settle near the $8.4M a quarter management expects, and cheap if June's level returns. If sales stay near September's $5.8M, 18.5 times is not cheap.
Against its own history. Buyers now pay a little less than NVE's usual price for its earnings, where before the results they paid about a third more.
The evidence under this - 4 claims
- On the six months' earnings of $1.97 a share run for a year the price is about 17 times, and the dividend yield is near 5.9%, so the price is cheap if September was only a timing change.
- Costs are small and mostly fixed, so a recovery in revenue lifts profit faster.
- 2026-10-21editedre-struck at the 20 October close and the trailing earnings to September 2026
- 2026-10-23editedre-struck at the 22 October close: 18.5 times trailing earnings, from 27.7 at $102.05, after the shares fell 33.7%
Priced against its rivals
None of the listed rivals above sells the same thing to the same buyers as a large part of what it does. A price comparison needs at least two, so the price is judged against its own history instead.
Where we disagree with the market
September was a pause, but the shares have been marked down hard because June's level did not repeat.
Part of June was a defense rebound that has faded before, so the old normal is the fair base, and at 18.5 times earnings the price now pays for that.
Twice before, in 2022 and 2023, a big quarter lifted the shares and sales fell back; in September it happened again within one quarter. Defense fell 67% in fiscal 2026 before rebounding, and the retiring chief executive sold.
Against us: NVE blames timing, and on the call said sales should recover to about $8.4M a quarter in the second half; six-month sales are still up 36%. The 33% fall has taken most of the premium out of the price.
What analysts say, and the evidence
The evidence under this - 8 claims
No filing gives the defense split in dollars; the 10-K never mentions defense sales at all, so neither the fall nor the recovery can be sized from the filings.
The 10-Q says only that both direct and distributor sales rose; it names no customer and gives no product split.
- 2026-10-21editedthe October 10-Q gave no customer or product split, so the July claim is rolled to the next print
- The 10-Q gives a different cause, a customer's temporary change in inventory needs, and the transcript's '95% decrease in defense product sales' appears in no filing, so the size of each cause is still unstated (S20, S23).
- With no backlog tracked, the second-half forecast cannot be checked against orders on hand (S23).
What could go wrong, and how we would know
The risks
- The spike has faded Sales fell to $5.8M in September; if December is also low, the price has little to rest on.
- Defense swings Defense sales fell 67% in fiscal 2026 and are driven by government buying cycles NVE does not control.
- One big customer Abbott is about 37% of sales, and its supply agreement runs out at the end of 2027.
- Bigger rivals In factory sensors and data couplers, much larger chip makers sell similar parts, and NVE says price pressure is rising.
- Dividend above profit NVE paid out $9.68M in six months against $9.57M of profit; if sales stay low, the cash pile shrinks or the dividend is cut.
The evidence under this - 6 claims
No filing gives the defense split in dollars; the 10-K never mentions defense sales at all, so neither the fall nor the recovery can be sized from the filings.
- Gross margin recovered to 81.3% in Q1 fiscal 2027 while volume rose 82%, so the pricing pressure did not show then.
- Fiscal 2026 product sales rose partly on price increases, which a business with no pricing power could not make.
What we still do not know
- Which customer rescheduled its orders in September, how much it was, and when the orders return; no filing says.
- How much of June's rise was defense and how much new products; the 10-Q gives no split.
- Whether Abbott will sign again when its agreement ends in December 2027, and on what terms.
The evidence under this - 3 claims
Six-month revenue is still up 35.5%, and the 10-Q calls the fall a timing change at one customer, which would let sales recover.
- 2026-10-21editedthe October print answered the open question: the jump did not repeat
- 2026-10-22editedthe call gave a second cause, a pull-forward of orders into June, and a second-half outlook
The 10-Q says only that both direct and distributor sales rose; it names no customer and gives no product split.
- 2026-10-21editedthe October 10-Q gave no customer or product split, so the July claim is rolled to the next print
The concentration note calls the customer 'Customer A' and never names it, and the risk factor names 'certain distributors' beside Abbott, so a distributor could in principle be the 37%.
What would change our view
- Neutral→AttractiveA cheaper price on today's earnings
It needs:
- ○The shares fall to about 15 times the last four quarters' earnings, from about 18.5 today.about $55.43 today, 52% below the last close
- Neutral→AttractiveThe step-up proves a new level
It needs all of:
- ○Quarterly sales stay at or above $10.0M in both the December 2026 and March 2027 quarters.checked at the next results, due by 20 Jan
- ○Management credits the growth mainly to non-defense products and new customers, not a defense order.checked at the next results, due by 20 Jan
- Neutral→UnattractiveThe fade is confirmed, or the price returns to paying for the jump
It needs any of:
- ○The shares rise to about 28 times the last four quarters' earnings, from about 18.5 today.about $103.46 today, -11% above the last close
- ○Quarterly sales are below $8.0M in the December 2026 quarter.checked at the next results, due by 20 Jan
We would also be wrong if
- Gross margin stays below 75% for two quarters while sales grow.
- Dividends exceed profit again for the full fiscal year 2028.
What is coming
- 20 January 2027 Q3 fiscal 2027 results: Whether revenue recovers to $8.0M or more after September's $5.83M, as management said it would when it forecast a second half like the first, about $8.4M a quarter.
Every dated event ahead
- 20 January 2027 Q3 fiscal 2027 results: Whether revenue recovers to $8.0M or more after September's $5.83M, as management said it would when it forecast a second half like the first, about $8.4M a quarter.
- 6 May 2027 Fiscal 2027 10-K: Customer A's share, the year's revenue and whether the Abbott agreement has been renewed.
- 31 December 2027 Abbott Supplier Partnering Agreement expiry: Whether the largest customer signs again, and on what terms.
Since we wrote this
- 23 OctPrice moveShares fell 33% after results; view back to NeutralAt 18.5 times trailing earnings, near NVE's usual level, the price no longer counts on June returning; December's sales are the next test.
- 22 OctThesis reviewView moved to UnattractiveThe owner adopted the recommendation after September's sales fell below the $8.0M line.
- 22 OctEarningsThe call: second half should match the firstManagement forecast about $8.4M a quarter, inside our band, so the view held.
- 21 OctEarningsSales fell to $5.8M; June's jump did not holdThe September quarter missed the $9.0M test and the $8.0M line. The view stays Neutral while a move to Unattractive waits for the owner.
- 7 OctThesis writtenNeutral, medium convictionWritten two months after a 55% one-day jump on Q1 fiscal 2027, two weeks before the next print.
Sources 24
Sources - 24
Every claim above cites one of these, or states the reasoning that derived it. The quote is frozen at the date it was read: a link can rot, and an agent fetch returns a reading of a page rather than the page. Backs counts the whole claim ledger, so it can name a claim this page does not surface.
- S01primarycompany filing2026-05-06products, fab, distribution, competition classes, Customer A at 37%, the Abbott agreement and its exhibits, risk factors, statements, dividends, headcount and the absence of trading plansbacksC01C02C03C04C10C12C16C18C19C26C33C38C39C40
- S02primarycompany release2026-05-06Q4 fiscal 2026 revenue of $7.65M and EPS of $1.02, and the year's resultsbacksC02C23
- S03primarycompany release2026-07-22revenue of $11.03M, net income of $6.39M, the income statement and June balance sheet, the dividendbacksC13C18C23C38C40
- S04primarycompany filing2026-07-22MD&A: growth in defense and non-defense sales through both channels, cost growth, cash flow, capital spendingbacksC13
- S05primarytranscript2026-05-06NVE Q4 fiscal 2026 earnings call transcriptdefense down 67% and non-defense up 21% for the year; defense expected to rise in fiscal 2027; the expansion; data centres and robotics; questions from three shareholders and analysts
a 21% increase in nondefense product sales, partially offset by a 67% decrease in defense sales, which can be volatile because of defense procurement cycles
backsC14C16C37C41 - S06primarytranscript2026-07-22NVE Q1 fiscal 2027 earnings call transcriptnew-product volume mostly to existing customers, the CEO transition, the MRI-safe sensors; a long-time holder's question whether the quarter is a new plateau
We'd say that's probably the strongest growing area and the most promising for our technology.
backsC15C16C25C37C41 - S07primarycompany filing2026-06-22ownership table, the succession, directors' biographies, pay formulas and amountsbacksC25C26C27C28
- S08primarycompany filing2026-09-09Form 4 filings by Daniel A. Baker, 27 July to 8 September 2026 Daniel A. Bakereleven open-market sales in 2026, none marked as under a Rule 10b5-1 plan, read by insider_fetch.pybacksC26
- S09primaryregulator2014-12-15FDA premarket approval database: Abbott supplements naming a GMR component, and the TactiFlex sensor supplement US Food and Drug AdministrationP030035 S061 and P880086 S180 (28 October 2009): change in the test coverage range for the GMR component in Abbott pacemakers and CRT-Ps; P030054 S279 and P910023 S349 (15 December 2014): alternate attach material for the GMR component in Abbott defibrillators; P220013 S018 (11 April 2025): a magnetic sensor process change 'to support more flexible sourcing'
ALTERNATE COMPONENT ATTACH MATERIAL FOR THE GMR COMPONENT ON HYBRID ASSEMBLIES FOR THE DEVICES.
backsC06 - S10primaryregulator2025-05-02FDA recall Z-1072-2025 and PMA supplement P970003 S247 (LivaNova VNS Therapy) US Food and Drug Administrationthe recall's reason (a mechanically activated component that may become stuck closed), 81,606 generators affected, and the fix approved on 2 May 2025
approval for replacement of the reed switch with giant magnetoresistor sensor for utilizing magnet mode within the implantable pulse generator (Model M1000 and M1000-D) as a corrective action for recall Z-1072-2025
backsC08 - S11primaryregulator2026-09-0121 CFR 814.39, PMA supplements (eCFR) US Food and Drug Administrationthe rule that a component change affecting safety or effectiveness needs a PMA supplement approved before it is made
Changes in the performance or design specifications, circuits, components, ingredients, principle of operation, or physical layout of the device.
backsC07 - S12primaryindustry data2026-09-24International Federation of Robotics, World Robotics 2026 press release International Federation of Roboticsrobot installations and the operating stock for 2025
The global operational stock of industrial robots surged 9% to a record 5 million units in 2025. This was driven by an 11% jump in annual installations: Factories worldwide installed more than 600,000 new units over the year.
backsC35 - S13primaryother2011-06-27Mond and Proclemer, The 11th world survey of cardiac pacing and ICDs, calendar year 2009 (PubMed 21707667) World Society of Arrhythmias survey, Pacing and Clinical Electrophysiologypacemakers and 328,027 defibrillators implanted in 2009 across 61 countries, more than 80% of the world's implants
The 2009 survey involved 1,002,664 pacemakers, with 737,840 new implants and 264,824 replacements.
backsC34 - S14primarycompany filing2026-05-21Allegro's Hall, GMR and TMR sensor portfolio and its markets, robotics among thembacksC29
- S15primarycompany filing2025-11-25iCoupler isolators sold against optocouplers, made on standard processesbacksC30
- S16primarycompany filing2026-02-19reed switch based magnetic sensing in its product listbacksC31
- S17secondaryvendor data2026-10-07Research pack, frozen 2026-10-07 (FMP, FINRA, Form 4 via FMP)price, 52-week range, the 252-day move split, daily closes and volumes since 2022, short interest, insider rowsbacksC21C24
- S18primarycompany filing2026-10-07quarterly revenue from fiscal 2011 to June 2026, annual statements and diluted share counts as filedbacksC11C12C19C21C23C40
- S19primaryindustry data2025-09-25International Federation of Robotics, World Robotics 2025 press release International Federation of Robotics542,000 robots installed in 2024, 2% below 2022
The new World Robotics statistics show 2024 the second highest annual installation count of industrial robots in history - only 2% lower than the all-time-high two years ago
backsC36 - S20primarycompany filing2026-10-21revenue of $5.83M, the order-rescheduling and defense explanation, the income statement, September balance sheet, dividendsbacksC17C22C42C43
- S21primarycompany release2026-10-21revenue, net income per share, the $1.00 dividend payable 30 November 2026backsC43
- S22secondaryvendor data2026-10-20Daily closes to 20 October 2026 (FMP end-of-day prices)close of $102.05 on 20 October 2026, 12.4% below the $116.53 of 6 OctoberbacksC20
- S23primarytranscript2026-10-21Q2 fiscal 2027 earnings call transcript (FMP)data/cache/transcripts/NVEC-2027Q2.json (FMP earning-call-transcript, label 2027Q2, call held 2026-10-21)outlook for the second half, the cause of the order variability, distributor sales growth, robotics and power-conversion answers, ASP per sensor
we expect revenues in the second half of the fiscal year to be consistent with the first half
backsC42C44 - S24secondaryvendor data2026-10-22Daily closes to 22 October 2026 (FMP end-of-day prices)close of $68.36 on 22 October 2026, the first session after the results, 33.7% below the $103.07 of 21 OctoberbacksC20C22