Gross margin guides up 100bps while the CEO admits the automotive market has been slow and the high-voltage GaN opportunity is a couple years out.
Thesis: Operational execution is real: beat, raise, GM guide up 100bps, inventory down 21 days to 292, FCF of $18M on $2M capex. But revenue growth of 3% YoY with data center producing zero P&L impact and auto described as slow means the market is paying for a narrative that management itself cannot size. The $3M OpEx reclassification further muddies the margin expansion story. Clean quarter, sub-scale growth, unpriced optionality that converts over years.
Verdict: HOLD — Conviction: MEDIUM
Catalyst: Q2 print landing at the high end ($120M) with GM at 55% would validate operating leverage. Data center design wins converting to booked orders with dollar figures attached would break the narrative gap.
Key Risk: 292 days on hand requires ~31% reduction to hit management's own 200-day target. If Q2 revenue lands at the low end, inventory days reverse direction and the efficiency narrative collapses.
The Tell: CEO states 'we're on track to double our automotive revenue this year' then three questions later says 'that's not huge revenue... it's been slow, the market's been slow, we've talked about a pushout.' Same CEO, same call, contradictory framing on the same segment.
Friction Level: MODERATE_FRICTION — Both sides agree operations improved. Disagreement centers on whether clean execution on 3% growth is worth a data center premium when zero dollars have converted from the $1B SAM narrative.
2026Q1 · May 7, 2026
3 of 5 readings agreed3 HOLD · 2 LONGRead the full call
Beat And Raise On 3% Growth, Data Center Revenue Still Fiction
Operational execution is real: beat, raise, GM guide up 100bps, inventory down 21 days to 292, FCF of $18M on $2M capex. But revenue growth of 3% YoY with data center producing zero P&L impact and auto described as slow means the market is paying for a narrative that management itself cannot size. The $3M OpEx reclassification further muddies the margin expansion story. Clean quarter, sub-scale growth, unpriced optionality that converts over years.
CEO states 'we're on track to double our automotive revenue this year' then three questions later says 'that's not huge revenue... it's been slow, the market's been slow, we've talked about a pushout.' Same CEO, same call, contradictory framing on the same segment.
Q1 beat midpoint with EPS at top of range and opex below. Q2 guide of $115-120M at 8.5% sequential with GM up 100bps and OM guide of 13.5-15.5%. Consistent pattern of conservative guidance met and exceeded.
$1B data center SAM by 2030, NVIDIA collaboration on 800V DC, 1250V and 1700V GaN sockets, all with zero quantified revenue contribution. CEO admits high-voltage GaN is 'a couple years out.' Two Taiwan design wins unquantified.
292 days on hand against 200-day target implies ~$50M further reduction on $108M quarterly revenue. Channel at 8.9 weeks against 8-week target. CFO frames inventory cleanup as 'cash' being recovered, suggesting prior build was excessive.
CFO Nancy Erba directly addressed channel question: 'not doing anything unnatural in the channel, demand is good, across the board.' Specific exit target of 8 weeks or below. No deflection on inventory either.
-18
CAUTIOUS"'Visibility is somewhat hampered by ongoing macro uncertainty.' Auto pushout language sits directly beside doubling claims. Q2 guidance qualified with 'it will depend upon the demand environment at the time.'"
Specific margin, inventory, and guide numbers are provided. Channel answer is clean. Data center discussion lacks dollar figures entirely. Auto pushout unquantified.
Q2 print landing at the high end ($120M) with GM at 55% would validate operating leverage. Data center design wins converting to booked orders with dollar figures attached would break the narrative gap.
292 days on hand requires ~31% reduction to hit management's own 200-day target. If Q2 revenue lands at the low end, inventory days reverse direction and the efficiency narrative collapses.
“We expect first quarter revenue to be between $104 million and $109 million.”
“Revenue was $108.3 million, up 3% from a year ago and 5% versus Q4 of last year.”
Companies tend to guide low: the median company we cover reports +1.7% against its guidance midpoint, so a small beat is the norm. A transcript can drop part of a guided range, so read both quotes.
“We expect revenue to be between $115 million-$120 million, which would be up 8.5% sequentially at the midpoint.”
Changed less than 56% of calls
Wording compared with the Feb 5, 2026 call, ranked among 97 calls reporting on Q1 2026. A measure of how much the story moved, not a signal about the stock.
Figures are extracted from the transcripts and checked in code: every quote appears word for word. Language change is computed from word frequencies, with no model involved. Descriptive only: none of this feeds the verdict.
Beat-and-raise pattern intact. GM guides up 100bps to 54-55% on volume and manufacturing efficiency, not FX. OM expanding 200bps sequentially to 11.7% with 13.5-15.5% Q2 guide. Channel inventory declining to 8.9 weeks on real demand per CFO. Balance sheet improving: 292 days on hand, down 21, with a credible path below 200. FCF of $18M on just $2M capex. Industrial up 23% YoY on structural electrification. Two new aux power design wins at Taiwan customers. NVIDIA collaboration on 800V DC architectures active with additional sockets identified. Opex growth capped at half of revenue growth going forward. New sales leadership from onsemi, Infineon, Cypress. TinySwitch-5 and TOPSwitchGaN ramping in 2H.
Revenue grew 3% YoY while the company wraps itself in an AI data center narrative. Zero dollars attributed to the $1B SAM. CEO admits high-voltage GaN is 'a couple years out' and auto is 'slow' with 'pushout' language while claiming to double. Compute and comms called 'sub-subscale' by the analyst and confirmed as non-growing by the CEO. $3M of the 200bps operating margin expansion comes from reclassifying engineers from SG&A to R&D, a line-item shuffle presented as operational efficiency. 292-day inventory requires 31% reduction to hit own target. Fading yen tailwind. Q2 guide comes off a weak base.
Inventory Divergence
292 days on hand needs ~31% cut to reach the company's own 200-day target, roughly $50M against $108M quarterly revenue. If Q2 revenue lands low, inventory days reverse.
AI Label Without Substance
$1B data center SAM, NVIDIA collaboration, 800V DC architecture narrative produce zero quantified revenue. CEO says high-voltage GaN is 'a couple years out.' SAM is marketing, not backlog.
Say/Do Gap
$3M reclassification from SG&A to R&D directly contributed to the 200bps OM expansion management presents as efficiency. Moving cost centers is not operating leverage.
HOLD. Operational numbers are clean: beat, raise, GM up, inventory down, FCF positive. But 3% YoY growth with data center revenue zero and auto pushing out does not give you a structural edge. The asymmetry is in the optionality, and you cannot size what management cannot quantify. Wait for data center bookings with dollar figures or a Q2 print at the high end proving the 8.5% sequential guide is sustainable, not a bounce off a weak base. No edge until the narrative converts to P&L.