Q1 adjusted free cash flow was $28 million against $411 million of adjusted EBITDA, and the full-year FCF guide still went up 10%.
Thesis: Qnity monetizes node transitions through consumables, not equipment: every additional layer and CMP step multiplies content per wafer with POR lock-in, so the margin engine is the angstrom-era schedule, not the cycle. The bear's utilization-ceiling point caps volume upside in semi, but ASP and mix carry it from here, and the $28M Q1 FCF is variable-comp timing against a $500-600M full-year guide, not a cash break.
Verdict: LONG — Conviction: MEDIUM
Catalyst: 2-nanometer volume scaling in 2H26 plus HBM4 ramp, with Taiwan site online early 2027 removing capacity constraint ahead of demand; MSI guided to mid-to-high single digits with 'room to do better' in 2H.
Key Risk: $20M mature-node restocking rolls off while semi guides roughly flat sequentially, and semi EBITDA margin is already down year-overyer on 12% organic growth. If pricing doesn't cover the $20M raw material/logistics headwind, the EBITDA guide (up 4%) trails revenue (up 5%) and keeps trailing.
The Tell: Kemp, unprompted after the 22% ICS beat: 'I don't think we're at the point where I want to update guidance on the ICS segment, but we're excited by the continued momentum.' He declined to raise segment guidance on a quarter he admitted beat internal expectations. Classic sandbagging, and the CEO confirmed it by refusing the raise.
Friction Level: MODERATE_FRICTION — Both sides agree on the numbers. Disagreement is whether $28M Q1 FCF and $20M restocking are timing artifacts (bull) or evidence of adjusted-metric earnings quality (bear), and whether high-80s DRAM utilization is demand strength or the volume lever running out.
2026Q1 · May 12, 2026
3 of 5 readings agreed3 LONG · 2 HOLDRead the full call
ICS Up 50%, They Still Won't Raise Segment Guide
Qnity monetizes node transitions through consumables, not equipment: every additional layer and CMP step multiplies content per wafer with POR lock-in, so the margin engine is the angstrom-era schedule, not the cycle. The bear's utilization-ceiling point caps volume upside in semi, but ASP and mix carry it from here, and the $28M Q1 FCF is variable-comp timing against a $500-600M full-year guide, not a cash break.
Kemp, unprompted after the 22% ICS beat: 'I don't think we're at the point where I want to update guidance on the ICS segment, but we're excited by the continued momentum.' He declined to raise segment guidance on a quarter he admitted beat internal expectations. Classic sandbagging, and the CEO confirmed it by refusing the raise.
8th consecutive quarter of strong profitable organic growth. Organic sales +17%, adj EBITDA +22%, EPS +33%. Full-year guide raised across every line: revenue +5%, EBITDA +4%, EPS +6%, FCF +10% at midpoints, plus MSI raised from mid-single to mid-to-high single digits.
Adj EBITDA margin +125bps YoY to 31.3%; ICS margin +280bps sequentially to 28.5% on mix toward Advanced Packaging, thermal, AI PCB. Each additional stack layer multiplies CMP steps and material content per wafer.
Management took a $20M geopolitical raw material/logistics headwind assumption into the guide and committed to targeted pricing actions to pass it through, on a local-for-local model with diversified suppliers.
POR positions across both segments with multi-year visibility. NVIDIA materials collaboration and Apple American manufacturing program inclusion put OEMs into materials selection, raising switching costs above the fab level.
$61.5M Taiwan site plus Delaware 385,000 sq ft expansion, capex at 9% of sales vs 6% long-term framework, deployed after POR wins booked. $100M EBITDA run-rate transformation targeted by end-2028.
DRAM high-80s heading toward 90%, mainstream logic low-80s, NAND climbing to low-80s. Volume lever exhausts at these levels; next leg depends on wafer pricing and node mix while semi EBITDA margin is already down YoY despite 12% organic growth.
ICS Advanced Packaging, thermal, and AI PCB grew >50% off what management calls shorter-cycle wins scaling from last year. Shorter-cycle revenue has weaker commitment than multi-year POR positions, per management's own description.
+75
CONFIDENT"Kemp gave node-level utilization data (DRAM mid-80s to high-80s, NAND mid-70s to low-80s), named POR wins by node (2nm, 16/14/10 angstrom), stated capacity timing by site, and committed to pricing pass-through on the $20M cost headwind. No deflection on any question."
Segment-level revenue, margins, utilization by memory type, inventory days, capex ratio, leverage, and full-year guide by line item. Both open roles disclosed rather than hidden. Only gap is undisclosed revenue attached to NVIDIA/Apple collaborations.
2-nanometer volume scaling in 2H26 plus HBM4 ramp, with Taiwan site online early 2027 removing capacity constraint ahead of demand; MSI guided to mid-to-high single digits with 'room to do better' in 2H.
$20M mature-node restocking rolls off while semi guides roughly flat sequentially, and semi EBITDA margin is already down year-overyer on 12% organic growth. If pricing doesn't cover the $20M raw material/logistics headwind, the EBITDA guide (up 4%) trails revenue (up 5%) and keeps trailing.
“We had an excellent start to the year with first quarter net sales of $1.3 billion, up 18% year-over-year, and 11% sequentially.”
Changed less than 72% of calls
Wording compared with the Feb 26, 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.
Eight straight beat-and-raise quarters with organic +17%, EBITDA +22%, EPS +33%, and every full-year line raised. ICS grew 22% organic with the three AI-exposed lines above 50%; semi 12% on advanced logic and HBM. Content-per-wafer is structural: 3nm scaling, 2nm volume in 2H, HBM4 ramp, angstrom nodes already in POR, each multiplying CMP and litho steps. POR across both segments plus OEM-level involvement via NVIDIA and Apple locks share for years. ICS margin +280bps sequential with stated trajectory from mid-20s to mid-to-high 20s. Pricing actions committed against the $20M cost headwind. Capex at 9% of sales booked after wins, Taiwan online early 2027. Order books strong, mainstream logic recovering mid-70s to low-80s utilization, an unpriced early-cycle leg.
$20M of semi's $722M came from mature-node restocking, and semi guides roughly flat sequentially, so underlying semi growth is single digits. Utilization is near ceiling: DRAM high-80s heading to 90%, logic low-80s. Volume lever is done; semi margin is already down YoY on 12% organic growth. Adjusted EBITDA of $411M converted to $28M of adjusted FCF, funded alongside $4B debt at 2.2x leverage, a dividend, and buybacks that only offset dilution. Capex at 9% of sales vs 6% framework. EBITDA guide raise (4%) trails revenue raise (5%), embedding margin dilution from the $20M cost assumption offset only 'with some timing variability.' Interim CFO still seated, Head of Semiconductor unfilled two quarters post-spin. Growth derivatives off TSMC, Samsung, and SK hynix capex calendars, and ICS's >50% growth comes off short-cycle wins management itself distinguishes from POR. NVIDIA/Apple announcements carry no disclosed revenue.
High Utilization as Ceiling
DRAM high-80s to above 90% and logic low-80s are presented as demand strength. At these levels the volume lever ends, yet semi EBITDA margin is already down YoY despite 12% organic growth.
Backlog Quality
The >50% ICS growth drivers are, per Kemp, shorter-cycle wins from last year scaling up. Weaker commitment structure than POR positions, so the fastest-growing segment has the least durable backlog.
Inventory Divergence
$20M of mature-node customer restocking in Q1, similar to Q1 2025, while the company guides semi roughly flat sequentially. Strip it and semi growth falls to single digits in line with MSI.
LONG at medium conviction. The asymmetry: the bear's cash argument is a Q1 timing artifact (variable comp, one-third of Taiwan capex in the quarter, FCF guide raised 10% to $500-600M), while the bull's argument compounds through the node schedule. Content-per-wafer with POR lock-in and demonstrated pricing power is the durable edge; utilization ceilings cap volume but mix and ASP carry it. Own it, size to the restocking rollback in Q2 semi as the near-term test, and let the ICS sequential raise confirm.