UCTT · Ultra Clean Holdings, Inc. · Stock research

Current Thesis

The July note framed this as a failed breakout waiting on a binary print. The print came, it was the strongest quarter in the company's history, and the tape did not care. On 2026-08-03 after close Ultra Clean reported record Q2 revenue of $644.9M against $587.7M consensus, non-GAAP EPS of $0.70 against $0.53, and guided Q3 to $700–750M revenue and $0.83–$1.03 non-GAAP EPS against $664.7M and $0.76. Three brokers raised targets the next morning — Oppenheimer to $150 from $115, TD Cowen to $140, Needham to $125. The stock closed the 2026-08-04 session 5.2% lower at $84.54 (Simply Wall St, 2026-08-05), and the last completed close on 2026-08-14 was $84.93. Seven sessions, a 9%-above-consensus guide, three target raises and a record quarter from the largest customer produced roughly half a percent of net price movement.

That is the whole read. The narrative leg an investor would be buying here is AI-driven wafer-fab-equipment capacity expansion flowing through a subsystems supplier with operating leverage — management now says 2027 WFE spend "could exceed $200 billion" and is building toward a $3.5B annualized capacity by end-2026, a $4B run-rate in early 2027 and $5B by late 2028. The market's stated objection is cash: cash and equivalents fell to $255.9M from $323.5M at Q1, operating cash flow was reported at negative $41.1M, and GAAP net income was $8.7M ($0.19 basic) against the $32.3M non-GAAP figure. A supplier funding a capacity build with a convert-heavy balance sheet and negative operating cash flow gets marked on liquidity, not on the order book.

The narrative is saturated. The dating is specific — mainstream sell-side clustering on a single day (2026-08-04, three raises to $125–150), a record print met with a 5.2% down session, and Applied Materials' own record FQ3 on 2026-08-13 ($9.115B revenue, +25% YoY, non-GAAP EPS $3.50, raised outlook) producing no follow-through into the 2026-08-14 close. Good news arriving without a bid is what a thin marginal buyer looks like. The alternative reading — that this is a maturing narrative being re-rated on cash conversion rather than an exhausted one — is live, and a daily close back over $105 on expanding volume would argue for it.

Bullish and bearish views on Ultra Clean Holdings, Inc.

The model's bull view on Ultra Clean Holdings, Inc. (UCTT), in brief: The guide is the largest sequential step-up in the company's disclosed history. The bear view: Cash conversion is the open wound. Operating cash flow reported at negative $41.1M with cash down $67.6M sequentially to $255.9M, driven by inventory built ahead of demand. Revenue growth financed by working capital in a business with $601.9M of long-term debt at the Q1 FY26… Both cases follow in full.

Bull Case

• The guide is the largest sequential step-up in the company's disclosed history. Q3 midpoint $725M against Q2's $644.9M is ~+12% sequential, on top of Q2's ~+21% sequential from $533.7M. Guided 2026-08-03.
• Margin ladder is moving in the right direction, quarter by quarter. Non-GAAP gross margin 16.7% in Q2 vs 16.5% in Q1; Products 15.1% vs 14.6%; operating margin 7.0% vs 5.1%. Management targets ~17% through the remainder of 2026 as utilization rises, and 20% at a $4B run-rate during 2027 (Q2 call, 2026-08-03).
• Concentration risk keeps falling. Top-two customer revenue moved to the high-50s percent from 64% (Q2 call, 2026-08-03), continuing the Q1 move to 59% from 67% for the OEM group.
• Published targets now sit 47–77% above the tape. Needham $125, TD Cowen $140, Oppenheimer $150, all dated 2026-08-04, against an $84.93 close on 2026-08-14. In late June the stock traded through every published number; that condition has fully reversed.
• Customer demand signal confirmed independently eight sessions later. Applied Materials' 2026-08-13 FQ3 was a record with a raised outlook, which is the order-flow source for this supplier's 2027 capacity plan.

Bear Case

• Cash conversion is the open wound. Operating cash flow reported at negative $41.1M with cash down $67.6M sequentially to $255.9M, driven by inventory built ahead of demand. Revenue growth financed by working capital in a business with $601.9M of long-term debt at the Q1 FY26 balance-sheet date (largely convertibles) is a liquidity question, and the 2026-08-04 session answered it with a 5.2% decline.
• GAAP and non-GAAP are $23.6M apart in a single quarter. GAAP net income $8.7M / $0.19 basic vs $32.3M / $0.70 non-GAAP. The Q3 guide carries the same spread: GAAP $0.67–$0.87 vs adjusted $0.83–$1.03.
• The June breakout has not been repaired. Price is ~$10 under the ~$95 shelf the June move launched from and 40.4% below the 2026-06-30 all-time close of $142.59. RSI(14) at 48.7 on 2026-08-14 is the middle of the range — no oversold spring, no momentum.
• Three-month return is -1.2% through 2026-08-14, across a window containing a record quarter, a record customer quarter and three target raises. That is the measurement; the inference is that incremental positive information is not clearing at these prices.
• Beta asymmetry is documented. On 2026-06-05 the stock fell 12.4% to $82.07 while Lam fell 9.85%, KLA 9.47%, Applied 9.70%. The July drawdown repeated it with no company news. The next sector risk-off day is unlikely to be gentler.

Setup & Price Structure

Last completed daily close $84.93 (2026-08-14). The structure has three reference points. Below: the $80.55 June low and the ~$82 shelf that the July note flagged — losing that on a weekly basis converts a two-month sideways attempt into a new leg down. Above: the ~$95 breakout shelf, now overhead supply, and $105, the level that last acted as the dead-cat-bounce high on 2026-07-09. Between $85 and $95 there is no structure of consequence in either direction.

Positioning and crowding observables, stated without a verdict: three price-target raises clustered on one calendar day (2026-08-04); a 47–77% gap between the published target range and the last close; no imminent earnings inside 30 days (Q3 is roughly early November); no insider transactions in the filing record reviewed; and, notably, no "extended above a rising moving average" condition — price sits 40.4% under its 52-week high, so the usual momentum-crowding markers are absent. What is present instead is analyst enthusiasm without price confirmation, which is a different and less commonly measured form of one-sided positioning.

The base has not formed. A stock that holds $84–86 for several weeks and then reclaims $95 on volume has repaired something; one that drifts is still distributing.

Catalyst Calendar (next 30 days)

• ~2026-09 (est.) — semiconductor investor-conference season. No confirmed Ultra Clean appearance was found in the record reviewed; treat as unscheduled rather than announced.
• ~2026-10-22 (est.) — Lam Research quarterly print. Unconfirmed date. Lam's December-quarter guide is the read-through on whether the $700–750M Q3 guide extends into 2027 orders.
• ~2026-11-02 (est.) — Q3 FY26 print. Unconfirmed date. Tests the guide, the ~17% gross-margin path, and whether operating cash flow turns positive at $700M+ revenue.

Elapsed catalysts

• 2026-08-25 — CFO transition effective. Michael Keogh succeeds Sheri Savage (named 2026-07-08). The 2026-08-03 call was the outgoing CFO's. The incoming CFO owns the cash-conversion narrative that just cost the stock 5.2%. (passed 1d ago)

What Would Change Our Mind

The structure that matters is the $80.55 June low. It has held through the entire post-print period, and it is what separates "range while the market waits for cash flow" from "the WFE re-rating is unwinding." A weekly close below $82 breaks it and the July–August repair attempt becomes a downtrend; that is the gradeable condition.

Two non-price conditions carry equal weight. First, a Q3 print landing at or below the $700M guide floor after a 9%-above-consensus guide would say management mis-read its own order book, which is disqualifying for a supplier whose entire case is order visibility. Second, a third consecutive quarter of negative operating cash flow with cash below the $255.9M reported at Q2 — or a convert or equity issuance to fund the Malaysia/Singapore/Czech capacity build — would confirm that the growth is being financed rather than generated.

On the other side: a daily close over $105 on expanding volume, or a Q3 print showing non-GAAP gross margin above 16.7% with operating cash flow positive, would argue the 2026-08-04 reaction was a liquidity scare rather than a re-rating, and would move the life-cycle label off saturated.

Correlation Notes

• ICHR — near-1:1 correlation, same OEM customers, same catalyst window. Treating both as separate ideas double-counts a single exposure.
• LRCX / AMAT — the demand source and the lead indicator. Applied's 2026-08-13 record print with a raised outlook did not lift this name into the 2026-08-14 close, which is the first hard observation that the usual 2–4 week trailing relationship is not currently transmitting.
• FORM / ONTO / ACMR / AMKR — moved -10.8%, -10.6%, -8.1% and -10.0% alongside this name in the July sector drawdown with no company-specific news in any of them. Sector beta dominates single-name news in downdrafts.
• MU and the HBM complex — memory-pure names have historically front-run this supplier by 4–8 weeks; their trend is the earlier signal on whether the 2027 ">$200B WFE" framing gets underwritten.