What the first reaction saw: revenue beat the pre-print mean by 9.7%, adjusted EPS beat by 31.9%, and the Q3 guide midpoint topped the pre-print means by 7.9% on revenue and 20.3% on EPS. S1S5
What the regular session saw: the first analyst said Q3 was about where expected and could have been better against the largest customer's growth; meanwhile Q2 adjusted gross margin improved only 20 basis points, inventory rose $148M sequentially, and Q2 free cash flow was negative $57M. S4S1S2
Investment read: the operating thesis strengthened, but the stock thesis is mixed at roughly 17.5x current 2027 EPS. Do not chase the opening squeeze; wait for proof that Q3 converts the revenue ramp into gross margin near 17% and positive cash flow. J1
Why the stock reversed on a very green day.
Tape causality is inferred, not observable. The ranking below distinguishes direct facts from the most plausible interpretation and states what would disprove each explanation.
The whisper bar was above published consensus
The Q3 midpoint beat the Aug. 3 Fiscal.ai means, but Tim Arcuri framed it as “right where I thought it would be” and said it “could have been even better” against roughly 30% sequential systems growth at UCTT's largest customer. Management answered that integration and fiscal timing require a two-quarter comparison. That is the cleanest explanation for the opening buyers failing to hold the gap. S4S5
Falsifier: confirmed order timing or subsequent estimates showing Q4 acceleration enough to match/exceed the customer over the two-quarter window.
Revenue growth did not convert cleanly into gross margin and cash
Revenue rose 20.8% QoQ and non-GAAP operating margin expanded 190 basis points—real leverage—but non-GAAP gross margin improved only 20 basis points. Higher-margin Services grew just 6.2%, Services mix fell to 11.2%, inventory rose 30.7%, and Q2 FCF was -$57.3M. The market can reward the demand while discounting the funding burden. S1S2
Falsifier: Q3 gross margin near 17%, positive OCF, and inventory growth below sequential revenue growth.
Positioning turned a quality debate into profit-taking
UCTT entered the print up 7.0% on Aug. 3 and about 244% YTD from the Dec. 31 close, then opened near $96 and traded as high as $98.675. That created a large pool of gains to monetize when the call did not clear the whisper bar. The shares were still roughly 39% below their June peak, so positioning amplifies the move but does not fully explain it. S6S10J3
Falsifier: similar richly positioned semi-cap names holding large gains while UCTT alone continues lower on new company-specific information.
A subsystem-supplier basket was being sold
The sector itself was not the problem: SOXX gained 6.58%, SMH 5.22%, Lam 6.77% and Applied 5.41%. But Ichor—another lower-margin subsystem supplier that also reported—fell 10.53%. The cross-section suggests investors separated high-margin WFE leaders from working-capital-heavy subsystem assemblers. S11S12J4
Falsifier: a broader sample of comparable subsystem vendors moving with the green semi index rather than UCTT/ICHR.
Short-covering likely amplified the squeeze and reversal
Latest reported short interest was 3.56M shares, about 7.9% of the 45.1M share count, with only 2.16 days to cover. That is enough to add volatility but not enough to establish a squeeze as the primary cause. S9S5J5
Falsifier: borrow data and trade-level evidence showing no meaningful short covering during the premarket rise.
Reaction path: euphoria, doubt, opening fade
Fallback: $89.17 close → $97.11 after-hours high → $90.99 after-hours last → $96.50 premarket last → $98.675 day high → $87.285 at 12:58 ET.
Reference is the Aug. 3 regular close. After-hours and premarket extrema/last from Nasdaq; regular-session data from Nasdaq/Yahoo. Thin extended-hours prints can be noisy. S6S7S8
A genuine operating beat, with a noisy bottom line.
The revenue and operating-income evidence corroborate the beat. The deductions sit below gross profit—in tax treatment, recurring compensation exclusions and cash conversion.
| Metric | Q2 2026 | Q1 2026 | Q2 2025 | Read-through | Source |
|---|---|---|---|---|---|
| Revenue | $644.9M | $533.7M | $518.8M | +20.8% QoQ / +24.3% YoY | S1 |
| Products revenue | $572.7M | $465.7M | $454.9M | +23.0% QoQ; 88.8% of total | S1 |
| Services revenue | $72.2M | $68.0M | $63.9M | +6.2% QoQ; mix fell to 11.2% | S1 |
| GAAP gross margin | 16.1% | 15.8% | 15.3% | +30 bp QoQ / +80 bp YoY | S1 |
| Non-GAAP gross margin | 16.7% | 16.5% | 16.3% | Only +20 bp on 20.8% revenue growth | S1 |
| Non-GAAP operating income / margin | $45.1M / 7.0% | $27.1M / 5.1% | $28.5M / 5.5% | Strong fixed-cost leverage; opex +1.6% QoQ | S1 |
| GAAP diluted EPS | $0.19 | -$0.40 | -$3.58 | $0.01 below the prior company guide low | S1S3 |
| Non-GAAP diluted EPS | $0.70 | $0.31 | $0.27 | $0.10 above prior guide high | S1S15 |
Six-quarter revenue and GAAP margin path
Fallback: Q2 revenue rose to $644.9M, GAAP gross margin reached 16.1% and GAAP operating margin reached 4.6%. Q2 2025 operating margin was -27.3% because of a $151.1M goodwill impairment.
Fiscal-quarter revenue, GAAP gross margin and GAAP operating margin. Q2 2025 includes the goodwill impairment. Official releases control; Fiscal.ai supplies normalized series checks. S1S5S15
Five-quarter adjusted EPS versus consensus
Fallback: Q2 2026 adjusted EPS was $0.70 versus the $0.53076 pre-print mean, a 31.9% surprise.
Company-adjusted EPS matched to provider estimates on the same adjusted basis. Fiscal.ai does not expose contributor identities; historical timestamps are used where available. S1S5
Accessible chart data
| Quarter | Revenue | GAAP GM | GAAP OM | Adjusted EPS | Consensus mean |
|---|---|---|---|---|---|
| Q1 2025 | $518.6M | 16.2% | 2.5% | n/a | n/a |
| Q2 2025 | $518.8M | 15.3% | -27.3% | $0.27 | $0.2700 |
| Q3 2025 | $510.0M | 16.1% | 2.1% | $0.28 | $0.2175 |
| Q4 2025 | $506.7M | 15.3% | 2.2% | $0.22 | $0.2250 |
| Q1 2026 | $533.7M | 15.8% | 2.1% | $0.31 | $0.2625 |
| Q2 2026 | $644.9M | 16.1% | 4.6% | $0.70 | $0.53076 |
Published numbers moved up sharply. The debate is the customer-relative cadence.
Pre-print means use Fiscal.ai's Aug. 3 revision snapshot; Aug. 4 means are a partial post-print set. The provider does not expose analyst identities, so these are directionally useful rather than audit-grade consensus.
| Metric | Company Q3 range | Midpoint | Aug. 3 pre-print mean | Midpoint delta | Aug. 4 partial mean | Read-through |
|---|---|---|---|---|---|---|
| Revenue | $700M–$750M | $725.0M | $671.72M | +$53.28M / +7.9% | $702.67M | Published bar cleared; post-print mean moved up $31.0M. |
| Adjusted EPS | $0.83–$1.03 | $0.93 | $0.7733 | +$0.1567 / +20.3% | $0.8808 | Published bar cleared; post-print mean moved up $0.1075. |
| GAAP EPS | $0.67–$0.87 | $0.77 | Not matched | n/a | Not matched | The deck shows adjusted EPS only; filed release contains GAAP. |
Q3 sequential math
Revenue midpoint implies +12.4% from Q2. The low end still implies +8.5%. Management did not guide gross margin, cash flow, capex, backlog or book-to-bill. S1J6
Two-quarter defense
Management argued that Lam integration timing and different fiscal calendars distort a one-quarter comparison; UCTT expects two-quarter growth to match or exceed its largest customer. That claim is testable after Q4. S4
The operating beat survives scrutiny. The $0.70 headline does not survive fully burdened economics unchanged.
These bridges are screens, not alternate GAAP measures. Each adjustment is kept distinct so investors can choose what they consider recurring.
Filed EPS bridge
| Step | $M | EPS |
|---|---|---|
| GAAP net income attributable to UCT | 8.7 | $0.19 |
| Amortization | +6.8 | +$0.15 |
| PR-defined stock compensation | +8.1 | +$0.18 |
| Restructuring | +0.7 | +$0.01 |
| Debt refinancing | +0.7 | +$0.01 |
| Remove unrealized FX gain | -1.9 | -$0.04 |
| Tax effect of adjustments | -2.9 | -$0.06 |
| Valuation-allowance tax adjustment | +12.1 | +$0.26 |
| Company adjusted result | 32.3 | $0.70 |
Source: filed reconciliation. The $0.26 valuation-allowance adjustment is 37% of adjusted EPS. S1S2
Recurring-economics screens
The screens are non-additive unless their tax interactions are modeled. They show sensitivity, not a claim that $0.44 or $0.56 is “true EPS.”
Cash moved into inventory ahead of the ramp
Fallback: inventory rose from $390.9M at year-end to $481.9M in Q1 and $629.9M in Q2; cash fell from $323.5M in Q1 to $255.9M in Q2; Q2 FCF was -$57.3M.
Q2 cash flow is derived from H1 less Q1 official tables; the deck directly confirms Q2 OCF of -$41.1M. S1S2J2
Two-year price: large gains, large drawdown
Fallback: UCTT closed 2025 at $25.33, reached $144.22 on June 30, closed Aug. 3 at $89.17 and traded at $87.285 at 12:58 ET on Aug. 4.
Daily closes through the partial Aug. 4 session. YTD return from the Dec. 31 close is approximately +244%; current price remains approximately 39% below the June high. S10J3
The call reinforced demand—but exposed cadence, margin and funding questions.
The available transcript is machine-generated. Official filed numbers and the deck control wherever ASR conflicts; timestamps below point to the call audio/transcript.
1. Q3 guide versus the largest customer
2. Outsourcing upside from OEM capacity constraints
3. Capacity built ahead of demand
4. Gross-margin path
5. Diversification versus concentration
6. Supply constraints and pushouts
7. 2027 WFE scenario
8. Services growth timing
At $87.285, the base recovery is no longer free.
Valuation uses the live price cut-off and Fiscal.ai current estimate means. Scenario ranges are illustrative—not price targets—and deliberately use the wide contributor range to show estimate risk.
Current earnings valuation
Estimate counts are small: four for 2026/2027 and only two for 2028. The 2028 EPS range is too dispersed for a hero metric. S5J8
What the price requires
- Q3 revenue lands within the new range without another inventory step-up.
- Gross margin approaches 17% and Services mix stops diluting.
- Q3/Q4 complete the promised customer-relative growth catch-up.
- 2027 EPS near $5 becomes repeatable cash earnings, not just adjusted operating leverage.
$3.49 2027 EPS × 14–17x. Margin stalls, cash burn persists or WFE/customer timing slips.
$4.97 2027 EPS × 17–20x. Revenue guide converts into ~17% GM and cash normalizes.
$6.59 2027 EPS × 18–22x. Outsourcing wins, cycle upside and 20% GM path become visible.
EPS inputs are the current low/median/high Fiscal.ai contributor estimates. Multiples are analyst-selected scenario assumptions, not consensus or company guidance. S5J8
What matters next.
The best next signal is not another demand anecdote. It is conversion: gross margin, operating cash flow, inventory and customer-relative cadence.
| Window | Proof point | Bull signal | Bear / falsifier | Source basis |
|---|---|---|---|---|
| Q2 10-Q | Inventory composition; SBC scope; cash-flow detail | Build is raw material/WIP against firm ramps; PR/GAAP adjustment scope reconciles | Finished-goods buildup, cancellations, or broader recurring adjustment scope | Not yet filed as of 12:56 ET; latest is Q1 10-Q. S13 |
| Q3 2026 | Revenue $700M–$750M; adjusted EPS $0.83–$1.03 | GM near 17%, positive OCF, inventory grows less than revenue | Volume lands but GM remains near 16.5% or cash burn continues | S1S4 |
| Q4 2026 | Two-quarter growth versus largest customer | UCTT matches/exceeds the customer as promised | The fiscal-timing explanation does not reconcile | S4 |
| 2027 | Services acceleration and $4B capacity utilization | Other OEM/Services outgrow top two; margin approaches 20% | Top-two concentration rises while utilization/cash conversion lag | S2S4 |
Thesis killers
- Q3 revenue within range but gross margin fails to expand.
- Inventory rises another 20%+ without matching revenue or cash proof.
- Top-two customer growth slows before diversification is visible.
- Recurring adjusted items keep GAAP and cash earnings structurally far below adjusted EPS.
- WFE expectations roll over while UCTT continues adding capacity.
Risk-aware action framework
Existing holder: the print alone does not break the thesis; size the position for execution and cash volatility, and require Q3 conversion.
New capital: the day-one fade improves entry price but does not yet provide the proof needed to underwrite the base case with high confidence.
Do not infer: this report is research, not a recommendation tailored to any portfolio, liquidity need, tax situation or risk limit.
Sources, definitions and analytical judgments.
Primary filings and company materials control. Market/provider data are time-stamped. Transcript claims are treated as lower-confidence where ASR conflicts with filed numbers.
Reported results, segments, outlook, balance sheet, cash flow and GAAP-to-non-GAAP reconciliation. Filed Aug. 3, 2026.
Customer/end-market mix, segment margins, Q2 OCF, outlook and long-term vision.
Controlling prior Q2 guidance and Q1 comparison.
Management narrative and Q&A; timestamps in debate map. ASR errors are disclosed and official numbers control.
Identifier NasdaqGS-UCTT; estimate means/revisions, valuation snapshot and historical statements, retrieved Aug. 4 around 12:50 ET through the local wrapper. Contributor identities are not exposed.
$87.285 / -2.11% at 12:58 ET; day range $82.855–$98.675.
Aug. 3 high $97.11, low $84.83, last $90.99; after-hours volume 120,810.
Aug. 4 last $96.50, high $96.99, low $90.00; premarket volume 107,440.
3,555,892 shares short at July 15, 2.16 days to cover; intraday volume ~1.99M versus 1.416M general average volume.
Two-year daily history and Aug. 4 partial session, retrieved Aug. 4. Used for returns and price chart; Nasdaq controls the live quote tile.
Subsystem-peer print, guide, margins, financing and cash-flow context.
Prior-quarter GAAP detail and SBC comparison. SEC submissions showed no Q2 10-Q as of Aug. 4, 12:56 ET.
Secondary records dated Aug. 4; full research notes were not available.
Prior-year margins, EPS and $151.1M goodwill impairment context.
Analyst-derived items
Data gaps and methodology limits
- Q2 10-Q was not filed by the report cut-off; inventory composition and the PR/GAAP SBC scope difference remain unresolved.
- Fiscal.ai's segment endpoint returned HTTP 500 and adjusted series were null; official issuer materials control those metrics.
- Fiscal.ai estimate means may contain partial post-print revisions; the report labels Aug. 3 and Aug. 4 snapshots separately.
- Extended-hours prints can be thin; Nasdaq high/low/last are shown rather than a synthetic continuous return series.
- No full sell-side research notes were available; price-target changes are secondary-feed records only.