Aehr’s miss is backward-looking; the tradable debate is whether $60 million to $80 million of second-half bookings converts before margins do

Aehr Test Systems reported the kind of quarter that screens badly in factor models and still may matter more as an inflection setup than as a miss. What was priced in was a loss-making, lumpy test-equipment supplier whose silicon carbide digestion and customer timing issues had already pushed revenue down from $18.3 million in Q3 FY2025 to $14.1 million in Q4 FY2025 and $11.0 million in Q1 FY2026. What actually surprised was split cleanly: revenue of $9.9 million came in below the $11.6 million Street estimate by -14.5%, while EPS of -$0.04 beat the -$0.08 estimate by +50.0%. The variant perception is that investors should not overread the EPS beat as quality, because the reported history still shows diluted EPS of -$0.11 and gross margin of 25.7% in Q2 FY2026, but should pay for evidence that the order book is finally reloading: second-quarter bookings were $6.2 million, quarter-end backlog was $11.8 million, and first six weeks of Q3 bookings added $6.5 million. That does not yet prove the fiscal 2027 story, but it shifts the question from “is there demand?” to “can booked demand arrive with contactor mix and margin recovery?”

That distinction matters because the revenue miss was not a rounding error, and the sequential trend still argues for caution on timing. The Street comparison basis was $9.9 million of revenue against $11.6 million expected, and the company’s own reported history shows Q2 FY2026 revenue of $9.9 million after $11.0 million in Q1 FY2026, a -9.9% sequential decline and -26.5% year-over-year decline. This is not a company exiting the downcycle in reported revenue yet. The gross margin history also shows why the EPS beat alone should not be mistaken for fundamental repair: gross margin fell to 25.7% in Q2 FY2026 from 33.9% in Q1 FY2026 and 40.1% in Q2 FY2025. The cleanest read is that consensus captured the loss profile better than it captured shipment timing, because EPS beat by +50.0% despite a -14.5% revenue miss. That creates an unusual setup where the P&L trough is visible, but the quality of the trough is weak.

The capacity story explains the margin guide, because Aehr’s model needs both systems and consumables to move together, and Q2 had the wrong mix. Chris Siu disclosed that contactor revenues, including WaferPaks, BIMs and BIBs, were $3.4 million, or 35% of total revenue, compared with $8.6 million, or 64% of revenue, in the prior-year quarter. That is the number behind the margin compression: non-GAAP gross margin was 29.8% compared with 45.3% a year ago, while the reported historical gross margin was 25.7% in Q2 FY2026. The miss therefore looks less like pure pricing pressure and more like a utilization and mix problem, because the highest-frequency consumable stream was down from $8.6 million to $3.4 million as a share fell from 64% to 35%. The Street should not capitalize a trough gross margin if WaferPak shipments normalize, but it also should not assume that revenue alone fixes the model unless contactor share rebounds toward last year’s 64% rather than staying near 35%.

The strongest part of management’s case is that the revenue trough is already being connected to explicit second-half numbers, not left as a thematic recovery narrative. Gayn Erickson’s key commitment was unusually specific: “Based on customer forecasts recently provided to Aehr, we believe our bookings in the second half of this fiscal year will be between $60 million and $80 million, which would set the stage for a very strong fiscal '27 that begins on May 30.” The wording matters because it puts a customer-forecast basis behind the range and ties the bookings, not just pipeline language, to fiscal 2027. Against a quarter with $6.2 million of bookings and $11.8 million of backlog, that $60 million to $80 million second-half target is a major escalation. It also means Q3 commentary will be judged less on revenue recognition and more on whether bookings quickly move from the $6.5 million received in the first six weeks of Q3 toward the implied second-half range management has put in front of investors.

The revenue guide then becomes the bridge between backlog reality and bookings ambition. For the second half of fiscal 2026, management expects revenue between $25 million and $30 million for the period that began November 29, 2025 and ends May 29, 2026. With Q2 revenue at $9.9 million and first six weeks of Q3 bookings at $6.5 million, the company is effectively asking investors to underwrite conversion from effective backlog of $18.3 million into that $25 million to $30 million revenue window, while also accepting that a larger $60 million to $80 million bookings wave is more about fiscal 2027 than about Q2 repair. Chris Siu’s phrasing on backlog was not casual: “Including these recent bookings, our effective backlog has now grown to $18.3 million, providing increased visibility as we move through the remainder of fiscal 2026.” The commitment embedded in that sentence is visibility, not upside; the upside case requires the next layer of bookings to land before investors have to wait for May 29.

That is why the WaferPak timing detail is more important than the headline miss suggests. Erickson said the company delayed approximately $2 million in WaferPak shipments from last quarter into this quarter, along with some system enhancements. A $2 million deferral is material against $9.9 million of Q2 revenue and against $3.4 million of total contactor revenues, but the implication is not simply that Q3 gets a one-time lift. The more important read is that customer transitions and system changes can move high-margin consumables between quarters, which helps explain the low 25.7% gross margin but also raises the bar for judging any Q3 rebound. If those WaferPak shipments show up without a contactor mix recovery beyond $3.4 million and 35% of total revenue, then the margin thesis breaks. If they show up alongside a better system mix, the Q2 margin trough should look less structural.

The customer read-through is narrow but actionable because Aehr’s demand signal is tied to wafer-level burn-in, not general semiconductor capital equipment. For STMicroelectronics and Infineon, both identified as customers for wafer-level burn-in and test systems, Aehr’s call implies ongoing capacity and process-transition work rather than a pause in qualification activity. The most concrete technical signal is the lead customer transition from 150 millimeters to 200-millimeter wafers, which Erickson said nearly doubles output without adding new FOX-XP systems, using WaferPaks that contact 100% of the die in a single touchdown. That matters for STMicroelectronics and Infineon because a customer can increase die output without a matching FOX-XP unit count, which can depress near-term system demand even as WaferPak and consumable opportunity persists. The second-order risk for Aehr is therefore embedded in its own value proposition: wafer-size transitions can raise customer output while reducing the urgency to buy $4 million, $5 million machines in the same quarter.

The competitive comparison reinforces why this has to be treated as a company-specific conversion story rather than a broad test-equipment beta trade. In the peer set, recent revenue YoY numbers include +43.8% for ATEYY, +12.3% for DSCSY, +13.7% for 7729.T, +48.3% for 6871.T, +22.6% for 6315.T, -3.7% for 6125.T, +43.0% for 6941.T, and +11.2% for 6140.T. Aehr’s Q2 FY2026 revenue YoY was -26.5%, which sits far away from that peer cadence. Its gross margin of 25.7% also clusters nearer the lower end of peers such as 6125.T at 25.4% and 6140.T at 28.3%, not the higher-margin peer prints of 67.4% at ATEYY and 70.8% at DSCSY. That does not make Aehr uninvestable; it clarifies the bet. Investors are not buying a synchronized test recovery in Aehr. They are buying a step-function order conversion in wafer-level and packaged-part burn-in where the company’s small revenue base makes bookings volatility dominate current margins.

The packaged-part burn-in data point adds a second growth vector, but it should be sized properly rather than treated as a new thesis by itself. During fiscal third quarter to date, Aehr received more than $5.5 million of orders from multiple customers for Sonoma ultra-high-power packaged-part burn-in systems, including initial orders from a premier Silicon Valley test lab. That compares with $9.9 million of Q2 revenue and $11.8 million of quarter-end backlog, so it is meaningful for near-term visibility. It is not yet enough to displace the wafer-level burn-in debate. The more interesting implication is mix diversification: if Sonoma orders become repeat business rather than initial lab placements, they can reduce the volatility caused by one customer delaying WaferPaks or moving from 150 millimeters to 200-millimeter wafers. Until the company reports a larger packaged-part revenue contribution, though, the second-half guide still rests primarily on converting the identified backlog and customer forecasts.

The call delivery supports the idea that management is pushing a bookings inflection, but the tone data also warns that the message was more scripted than de-risked. The tone history shows Q2 FY2026 sentiment at 0.24, up from 0.21 in Q1 FY2026, and guidance_tone at 0.54, up from 0.44. Prepared_sentiment jumped to 0.49 from 0.06, while qa_sentiment fell to 0.04 from 0.19, tone_confidence dropped to 0.21 from 0.31, and uncertainty rose to 49.0 from 34.2. That combination is important: management’s prepared script was much more positive, but the Q&A did not confirm the same intensity, and confidence fell while uncertainty rose. The one mitigating data point is qa_evasiveness, which dropped to 4.8 from 34.3, meaning the weaker Q&A sentiment was not accompanied by a more evasive pattern. Investors should hear the $60 million to $80 million bookings target as a real marker, but not as a fully de-risked conversion path.

The balance sheet and dilution details explain why management can pursue the second-half opportunity, but also why equity holders will remain sensitive to the cadence of orders. Aehr ended the quarter with $31 million in cash, cash equivalents and restricted cash, up from $24.7 million at the end of Q1. During Q2 FY2026, the company raised $10 million in gross proceeds through the sale of about 384,000 shares under an ATM framework that followed a $100 million S-3 shelf-registration and an ATM offering of up to $40 million. That capital raise reduces near-term liquidity pressure while the P&L is negative, but it also means investors are being asked to accept dilution ahead of proof that the $60 million to $80 million bookings range converts. Non-GAAP operating expenses were $5.7 million, down 4% from $5.9 million in Q2 last year, so the immediate issue is not expense escalation. It is whether revenue and contactor mix can rise fast enough for operating leverage to reappear before the market prices in more equity funding risk.

The thesis therefore comes down to a simple asymmetry: the reported quarter was worse on revenue and gross margin than investors wanted, but the forward booking language is too specific to ignore. The bear case is that Q2’s $9.9 million revenue, 25.7% gross margin, $3.4 million contactor revenue, and 35% contactor mix reveal a lower-quality revenue base, while the $60 million to $80 million bookings target remains customer-forecast language rather than purchase orders. The bull case is that $6.2 million of Q2 bookings, $11.8 million of quarter-end backlog, $6.5 million of first-six-weeks Q3 bookings, and $18.3 million effective backlog are the early stages of a second-half reload, with $25 million to $30 million of guided second-half revenue acting as the first conversion test. I would side with the latter only as a bookings-led setup, not as a margin-recovery victory lap, because the gross margin series has to move from 25.7% back toward the levels implied by better contactor mix before EPS quality improves.

What to watch next is concrete. By the next update, the first confirmation is whether cumulative second-half bookings are tracking toward Gayn Erickson’s $60 million to $80 million range rather than merely adding to the $6.5 million already received in the first six weeks of Q3. The second is whether effective backlog moves materially above $18.3 million while second-half revenue remains within the $25 million to $30 million guide for the period ending May 29, 2026. The third is mix: contactor revenue needs to recover from $3.4 million and 35% of total revenue, especially with approximately $2 million in WaferPak shipments delayed into this quarter. The fourth is margin: reported gross margin at 25.7% and non-GAAP gross margin at 29.8% need to move toward the Q1 FY2026 gross margin of 33.9% as an initial sign that Q2 was timing and mix, not pricing or utilization damage. The next scheduled investor touchpoints are the 28th Annual Needham Growth Conference on Tuesday, January 13, and the 15th Annual Susquehanna Technology Conference on Thursday, February 26; any walk-back from $60 million to $80 million of second-half bookings, $25 million to $30 million of second-half revenue, or $18.3 million effective backlog would break the variant thesis.