MorningSignal
Research Platform
2026-09-28 13:40
type: earnings-brief session: PM date: 2026-07-28 daily_note: "[[Daily/2026-07-28]]" tags: [earnings, sellside]
← [[Daily/2026-07-28|Back to the daily note]]
This was a high-volume AMC tape with six releases worth immediate underwriting and a much larger tail that cannot responsibly be promoted from discovery evidence into analysis. The best fundamental revisions were [[STX]], [[BE]], and [[TER]]: Seagate’s next-quarter revenue/EPS guide moved the nearline-storage earnings slope materially higher; Bloom converted the data-center power bottleneck into a $1.065 billion quarter and a second large FY26 guide raise; Teradyne’s memory-test cycle produced a second consecutive record-revenue quarter. All three securities entered the print with demanding valuation and/or positioning, so the actions are HOLD / WAIT, not momentum-chasing BUYs.
[[F]] delivered the most attractive cash-flow valuation but also the lowest GAAP quality: adjusted EBIT and free-cash-flow guidance rose, while $4.1 billion of charges tied mainly to the BlueOval SK exit and EV-program cancellations produced a GAAP loss. The business delta is positive, the estimate delta is positive, but the capital-allocation error is real. HOLD; add only at or below $14 with FY26 adjusted FCF still at least $6 billion.
[[V]] showed no consumer break—payments volume, processed transactions, and cross-border volume all grew double digits—but client incentives and operating expense grew faster than net revenue. At roughly 27.6x annualized adjusted EPS before the print, a routine beat is not a sufficient re-rating catalyst. HOLD; new money at or below $330 or after incentive growth recedes below net-revenue growth.
[[CSGP]] produced a genuine profitability inflection but cut the revenue slope: Q2 revenue grew 18% and net income rose to $55 million, yet the $935–$945 million Q3 revenue guide and $3.72–$3.76 billion FY26 range sit below the prior trajectory. WAIT at or below $25 or until annualized net new bookings reaccelerate.
No AMC company clears the full-call gate tonight. “Earnings call highlights” pages are not complete transcripts; therefore all six underwrites remain PROVISIONAL — RELEASE ONLY, conviction is capped, and sentiment is unscored. [[KLAC]] is the most important explicit deferment: the collector mixed stale March-quarter financials into the June-quarter record, while the current issuer release/shareholder letter and complete Q&A were not retrievable by cutoff. It is queued first for 2026-07-29 08:00 ET.
| Ticker(s) | Report-date evidence | Market cap / reaction relevance | Transcript status | Tier | Reason |
|---|---|---|---|---|---|
| V | 7/28 result validated; public result detail | $686.8B; global consumer/payments read-through | Complete current/prior Q&A not verified | TIER 1 — FULL UNDERWRITE | Mega-cap read-through and material expense/incentive debate |
| STX | 7/28 result validated; prior guide and current result detail | $183.2B; strong positive initial reaction after a -16.2% five-day drawdown | Highlights only | TIER 1 — FULL UNDERWRITE | AI storage estimate reset and large guide delta |
| F | 7/28 result validated; current result detail | $58.5B; positive initial reaction | Complete current/prior Q&A not verified | TIER 1 — FULL UNDERWRITE | Auto-cycle, cash-flow, EV-charge and capital-allocation read-through |
| BE | 7/28 result validated; current result detail | $53.5B; roughly +8% to +12% initial reaction after a -26.3% five-day drawdown | Complete current/prior Q&A not verified | TIER 1 — FULL UNDERWRITE | AI data-center power bottleneck and guide reset |
| TER | 7/28 issuer-distributed release | $52.4B; positive initial reaction after a -14.3% five-day drawdown | Complete current/prior Q&A not verified | TIER 1 — FULL UNDERWRITE | Memory/AI semiconductor-test cycle and record revenue |
| CSGP | 7/28 result validated; issuer-distributed detail | $11.9B; approximately -13% initial reaction reported | Highlights only | TIER 1 — FULL UNDERWRITE | Guidance cut versus profitability inflection |
| KLAC | Date/call validated, but current financial package failed | $265.6B; reported roughly -10% initial move | Highlights only | TIER 3 — DEFERRED / ESCALATED | Collector financials were stale; current issuer release, shareholder letter, guide, and full Q&A absent |
| WM, MDLZ, NXPI, ACGL, EXR, FE, PPG, OMC, VLTO, EXE, WPC, LOGI, UNM, CR, BXP, UMBF, SWKS, MANH, AXS, ZWS, QRVO, STAG, THG | Result/headline evidence detected | $7B–$95B; lower portfolio priority than completed slate | None or highlights only | TIER 3 — DEFERRED | Current issuer package, dated expectations range, and complete current/prior Q&A not all verified |
| NOV, CHE, LSTR, CZR, RUSHA, RUSHB, CAR, VRNS, ENPH, EXLS, CAKE, MIR, MEOH, RNST, HIW, GEF, GEF.B, CGAU, APAM, SBCF, TTAM, ASH, AKR, ACHC, TRMK, BUSE, WERN, SLDE, ROG, MBIN, FCF | Calendar entry only; collector stopped evidence collection below the top market-cap cohort | $2B–$7B | Not verified | TIER 3 — DEFERRED | Actual release, primary financial package, expectations stack, and call record absent |
PM decision line — PROVISIONAL — RELEASE ONLY. The pre-print bar was not merely the collector’s $3.23 adjusted-EPS consensus; at $366.59, Visa traded at about 27.6x annualized Q3 adjusted EPS and therefore needed durable double-digit volume growth without a faster incentive/expense burden. Fiscal Q3 net revenue was $11.6 billion, up 14%, and adjusted EPS was $3.32, up 11% and 2.8% above the collector’s dated mean. The operating engine remains healthy, but the cost of acquiring volume is rising faster than revenue. Action: HOLD; medium confidence; add only at or below $330, or after client-incentive growth falls below net-revenue growth while cross-border ex-Europe remains at least 10%.
| Metric | Prior guide / known bar | Dated reported consensus | Hurdle | Actual | Rate of change / classification |
|---|---|---|---|---|---|
| Adjusted EPS | No verified quarterly guide | $3.23, 11 estimates | Buy-side hurdle not verifiable | $3.32 | +11% YoY; 2.8% beat; STRUCTURAL POSITIVE |
| Net revenue | No verified range in evidence set | Reliable range not available | At least low-double-digit growth implied by valuation | $11.6B | +14% YoY; STRUCTURAL POSITIVE |
| Payments volume, constant currency | No numeric guide verified | n/a | Preserve high-single/low-double-digit growth | +10% | Demand intact; STRUCTURAL POSITIVE |
| Cross-border ex-Europe | No numeric guide verified | n/a | Double digit | +12% | Travel/commerce resilient; STRUCTURAL POSITIVE |
| Client incentives | n/a | n/a | Grow no faster than revenue over time | $4.7B, +18% | 4 points faster than revenue; STRUCTURAL NEGATIVE |
The collector’s $11.23 billion revenue “actual” was rejected because the current public result detail states $11.6 billion. The EPS quality read is mixed but not low-quality by the brief’s rule: GAAP EPS was $2.97 versus $3.32 adjusted, with a reported $563 million workforce-reduction/restructuring burden affecting GAAP expense. That item does not create the adjusted beat; rather, it explains much of the GAAP-to-adjusted gap. A complete reconciliation and tax/share-count bridge remain required before assigning a final quality grade.
Payments volume grew 10% in constant currency, processed transactions reached 71.7 billion and grew 10%, and cross-border volume excluding intra-Europe grew 12%. Those three measures jointly say the network’s consumer and travel throughput is not breaking: volume feeds service revenue, transactions feed data-processing revenue, and cross-border mix feeds the higher-yield international transaction line. Data-processing revenue rose 17% to $6.0 billion and service revenue 14% to $4.9 billion, while international transaction revenue grew only 6% to $3.9 billion. The divergence suggests mix/FX and pricing mechanics matter more than the headline cross-border volume rate.
The buried signal is client incentives: $4.7 billion, up 18%, outgrew net revenue by four points. Incentives are the contractual price Visa pays issuers, merchants, and partners to route volume over its rails. If that spread persists, each incremental payment-volume point transmits less efficiently to net revenue and operating leverage. This is not a demand break, but it is the exact variable that separates a high-quality volume beat from an earnings-multiple expansion.
No reliable current numeric FY guide or post-print analyst revision was available by cutoff. The transparent algebra is: payments volume × yield + processed transactions × data-processing yield + cross-border mix − client incentives − operating expense. Holding 10% volume and transaction growth, 12% cross-border ex-Europe growth, and mid-teens value-added/data-processing growth supports low-double-digit FY1 revenue growth. If incentives continue four points faster than revenue and expense remains elevated, FY1 EPS grows closer to high single/low double digits; if the incentive gap closes, EPS can again outgrow revenue by several points. FY2 is therefore more sensitive to incentive normalization and restructuring savings than to a heroic consumer rebound.
At $366.59, annualizing $3.32 produces $13.28 and a 27.6x run-rate P/E before seasonality. A $330 entry equals roughly 24.8x that run rate; a 30x multiple on $14.50 normalized forward EPS supports about $435, while 23x on $13.50 supports about $311. The current price discounts continued double-digit growth and leaves limited room for an incentive-led margin miss.
| Live claim | Evidence required | Evidence received | Verdict / next proof |
|---|---|---|---|
| Consumer spending remains resilient | Volume and transaction growth at least high single digit | Both +10% | Strengthened; check October quarter |
| Cross-border remains the premium growth leg | Ex-Europe growth ≥10% and international revenue conversion | Volume +12%, revenue +6% | Partial; yield/mix bridge unresolved |
| Visa can preserve operating leverage | Incentives and expense no faster than revenue | Incentives +18%, expense +19%, revenue +14% | Weakened |
Thesis matrix: demand REINFORCED; pricing/yield UNRESOLVED; margin/cost architecture WEAKENED; competitive position UNCHANGED; capital allocation UNRESOLVED; management credibility PENDING — CALL; catalyst timing UNCHANGED. Entering the print, Visa was a premium compounder needing a clean double-digit algorithm. After the release, the story is “demand intact, but conversion costs are rising.” The initial after-hours slip is consistent with that gap, but it is not a settled reaction and is not used as causal proof.
Business delta: positive demand evidence, negative conversion-cost evidence. Estimate delta: modestly positive revenue/EPS versus the dated mean, but below what the volume headline alone implies. Stock delta: fairly valued to slightly rich before the print. Three call questions: What portion of the incentive growth is mix/timing versus structural repricing? What is the annualized savings/payback from the $563 million action? Why did international transaction revenue grow materially slower than cross-border ex-Europe volume? Confirmation: incentives grow no faster than revenue next quarter. Falsification: payments volume below 7% or incentive growth remains at least five points above revenue. Next proof: fiscal Q4, expected October 2026. 10-second PM line: Visa’s rails are healthy; the monetization spread, not the consumer, is tonight’s risk.
PM decision line — PROVISIONAL — RELEASE ONLY. The prior-quarter company guide was $3.45 billion revenue ±$100 million and $5.00 adjusted EPS ±$0.20; the market needed proof that nearline demand and HAMR-led areal-density economics could extend the March-quarter record rather than peak. Seagate delivered about $3.6 billion and $5.71, then guided the September quarter to $4.0–$4.2 billion revenue and $7.10–$7.50 adjusted EPS, both far above public pre-print estimates cited in the evidence set. This is a genuine FY1/FY2 estimate reset, but the stock’s pre-print close of $747.30 already represented a premium storage-cycle multiple. Action: HOLD; do not chase the positive initial reaction; new money at or below $650 or after a full transcript proves capacity, pricing, and free-cash-flow durability.
| Metric | Prior guide | Dated reported consensus | Hurdle | Actual / new guide | Classification |
|---|---|---|---|---|---|
| Q4 revenue | $3.35–$3.55B | Approx. $3.45B public bar | Above high end and evidence of continued nearline pull | About $3.6B | At least $50M above high end; STRUCTURAL POSITIVE |
| Q4 adjusted EPS | $4.80–$5.20 | Collector $5.09 / calendar $4.89 | Clear high-end beat | $5.71 | $0.51 above high end; STRUCTURAL POSITIVE |
| Q1 FY27 revenue guide | n/a | About $3.75B reported | Sustain sequential growth | $4.0–$4.2B | Midpoint roughly 9% above bar; STRUCTURAL POSITIVE |
| Q1 FY27 adjusted EPS guide | n/a | About $5.80 reported | Preserve gross-margin/earnings leverage | $7.10–$7.50 | Midpoint roughly 26% above bar; STRUCTURAL POSITIVE |
Revenue grew about 49% YoY and adjusted EPS about 120%. The EPS rate far above revenue signals powerful gross-margin and operating leverage, consistent with the March quarter’s 47.0% non-GAAP gross margin. The current detailed GAAP/non-GAAP bridge, tax rate, dilution from the 2028 exchangeable notes, and share-count effect were not retrievable by cutoff; therefore the EPS-quality gate is PARTIAL, not failed and not fully passed.
The causal engine is exabytes shipped × price per terabyte × areal density − component/capacity cost. Nearline cloud demand raises mass-capacity volume; HAMR and higher terabytes per drive raise the value delivered per chassis and reduce customer rack/power cost; disciplined industry supply supports price and margin. The quarter’s 49% revenue growth, 120% EPS growth, and next-quarter guide acceleration jointly indicate that the positive cycle is not merely inventory refill. In the prior quarter Seagate produced $953 million of free cash flow and retired $641 million of debt, proving that gross-margin expansion was already converting to balance-sheet repair.
The buried signal is the step from $5.71 actual EPS to a $7.30 next-quarter midpoint. That is a 28% sequential guide increase against only roughly 14% sequential revenue growth from $3.6 billion to $4.1 billion. The implied incremental margin is unusually high and therefore forces the debate onto product mix, pricing, and HAMR yields. If the bridge is primarily sustainable areal-density economics, FY27 numbers reset structurally; if it is allocation/timing or exceptionally favorable mix, the guide may mark the peak.
The minimum annualized FY27 revenue run rate from the September midpoint is $16.4 billion, versus $13.8 billion annualized from the prior $3.45 billion guide. The EPS run rate moves from $20.00 at the old Q4 midpoint to $29.20 at the new Q1 midpoint. This is not a formal forecast because storage is seasonal and cyclical, but it frames the magnitude: the new guide implies roughly $9.20 more annualized earnings capacity than the prior guide.
FY1 algebra is nearline exabytes × blended $/TB × gross margin − R&D/SG&A − interest/tax/dilution. A 100-basis-point gross-margin change on a $16.4 billion revenue run rate is about $164 million of annual operating income before tax. FY2 depends on HAMR mix, competitive capacity discipline, and whether hyperscaler demand commitments convert without price concessions. The prior $641 million debt retirement also matters: less debt reduces interest drag and improves the equity conversion of each EBITDA dollar.
At the $747.30 pre-print close, the new quarterly EPS midpoint annualizes to $29.20, or 25.6x; $650 equals 22.3x. A bull case of $34 FY27 EPS at 27x supports $918; a base of $29 at 23x supports $667; a bear of $22 at 18x supports $396. Weights of 25%/50%/25% yield roughly $662, below the pre-print close, showing that even a large estimate beat was partly priced.
| Live claim | Required evidence | Evidence received | Verdict |
|---|---|---|---|
| AI/cloud creates a structural nearline upcycle | Sequential revenue/guide acceleration | $4.1B next-quarter midpoint after ~$3.6B actual | Strengthened |
| Margin gains are product/yield, not temporary allocation | Gross-margin and HAMR yield bridge | EPS guide accelerates; detailed bridge absent | Unresolved |
| Cash converts and leverage falls | FCF and debt retirement | Prior quarter $953M FCF and $641M debt retired; current quarter missing | Partial |
Thesis matrix: demand IMPROVED; pricing/mix IMPROVED provisionally; margin IMPROVED provisionally; competition UNRESOLVED; balance sheet REINFORCED from prior-quarter cash evidence; management credibility PENDING — CALL; catalyst timing IMPROVED. The narrative moved from “late-cycle storage rerating vulnerable after a large run” to “earnings slope still steepening.” Incremental bears must now argue that the September guide is peak mix, not that nearline demand has already rolled.
Business delta: materially positive. Estimate delta: materially positive FY1 and probably FY2, subject to gross-margin proof. Stock delta: initial positive reaction is directionally justified, but the pre-print valuation limits chase upside. Three call questions: How much of the September EPS guide is gross margin versus opex/tax/share count? What share of nearline shipments is HAMR and what are current yields? Are customer commitments firm enough to protect volume and pricing into calendar 2027? Confirmation: September revenue at least $4.0 billion and non-GAAP gross margin not below the June level. Falsification: guide cut below $3.8 billion or gross margin falls more than 200 bps without a timing explanation. Next proof: fiscal Q1 FY27, expected October 2026. 10-second PM line: the guide resets earnings; the transcript must prove it is product economics rather than peak-cycle mix.
PM decision line — PROVISIONAL — RELEASE ONLY. The pre-print question was whether Ford could turn a normalizing production quarter into durable free cash flow while containing warranty, commodity, and EV restructuring costs. Revenue was $48.3 billion, adjusted EPS $0.42 versus roughly $0.35 consensus, and adjusted EBIT $2.5 billion; FY26 adjusted EBIT rose to $10–$11 billion and adjusted free cash flow to $6–$7 billion from $5–$6 billion. Yet Ford reported a GAAP loss because $4.1 billion of charges—$3.6 billion tied to exiting the BlueOval SK battery joint venture and about $0.5 billion tied to EV-program cancellations—overwhelmed operating profit. Action: HOLD; add only at or below $14 while adjusted FCF guidance remains at least $6 billion and warranty cost does not reaccelerate.
| Metric | Prior guide / Street | Actual / new guide | Classification | |---|---:|---:| | Revenue | About $47.2B public consensus | $48.3B | ~2% beat; STRUCTURAL POSITIVE / mix still pending | | Adjusted EPS | $0.33 collector; ~$0.35 public bar | $0.42 | 20%–27% beat; STRUCTURAL POSITIVE | | Adjusted EBIT | ~$2.15B public bar | $2.5B | ~16% beat; STRUCTURAL POSITIVE | | FY26 adjusted EBIT | $8.5–$10.5B | $10–$11B | Midpoint +$1.0B; STRUCTURAL POSITIVE | | FY26 adjusted FCF | $5–$6B | $6–$7B | Midpoint +$1.0B; STRUCTURAL POSITIVE |
The adjusted beat passes the operating test but fails a clean GAAP-quality test. The $4.1 billion charges are more than 30% of the GAAP/adjusted earnings variance by any reasonable measure; they are not a small accounting footnote. They represent the cost of prior capital allocation and portfolio decisions. Investors should value the forward cash engine on adjusted EBIT/FCF while explicitly subtracting cash restructuring needs and refusing to call the quarter “high quality” solely because adjusted EPS beat.
Ford’s engine is wholesales × net price/mix − material/warranty/labor cost, plus Ford Pro software/service contribution, Model e losses, and Ford Credit financing income. Ford Blue generated $1.14 billion of EBIT, above a roughly $793 million public estimate; Ford Pro generated $1.72 billion but declined 26% YoY; Model e lost $919 million, better than the roughly $1.09 billion loss expected. The compound is mixed: core combustion and commercial franchises support cash, Model e losses are narrowing, but Pro deceleration and restructuring charges show that improvement is not uniform.
The buried signal is the $1 billion midpoint increase in adjusted FCF guidance. At the pre-print $58.5 billion market capitalization, the new $6.5 billion midpoint is an 11.1% adjusted FCF yield before Ford Credit and balance-sheet complications. That cash yield can support dividend and reinvestment even without multiple expansion. The risk is that “adjusted” cash omits or delays the economic cost of restructuring and warranty; the call must reconcile the $4.1 billion accounting charge with actual 2026–27 cash outflows.
FY1 bridge: Q2 adjusted EBIT of $2.5 billion plus the raised $10.5 billion FY midpoint implies a similar $5.25 billion combined H2 requirement after $3.5 billion in Q1—roughly $2.6 billion per quarter. The new guide is therefore achievable without an implausible hockey stick, but it still assumes the Novelis recovery, product mix, warranty savings, and Model e loss reduction remain on plan. Each $1 billion change in annual adjusted EBIT is about $0.19 per share after a 25% tax rate and roughly 4.0 billion shares, before financing differences.
FY2 depends on Ford Pro margin stabilization, paid-software growth, warranty cost, and the cash burden from the battery/EV reset. If $6.5 billion adjusted FCF persists, a 10% equity FCF yield implies $65 billion of equity value, or roughly $16.25 per share. At an 8% yield it implies about $20.30; at a 13% yield about $12.50. The $14 add trigger corresponds to a roughly 11.6% yield on the midpoint and provides a modest cushion for restructuring leakage.
| Live claim | Required evidence | Evidence received | Verdict |
|---|---|---|---|
| Ford can earn through Model e losses | Blue/Pro/Credit profit exceeds Model e drag | Blue $1.14B, Pro $1.72B, Model e -$0.919B | Strengthened |
| FCF is improving structurally | Raised guide and cash conversion | Guide +$1B midpoint; cash bridge absent | Strengthened but incomplete |
| EV reset is behind the company | Limited new charges and declining cash drain | $4.1B new charges | Falsified for this quarter |
Thesis matrix: demand UNRESOLVED; pricing/mix IMPROVED; margins IMPROVED; competitive position UNCHANGED; capital allocation WEAKENED; management credibility PENDING — CALL; catalyst timing IMPROVED. The narrative moved from “cheap but trapped by EV/warranty cash leakage” to “core earnings can fund the reset, but the reset still costs real money.” The initial positive reaction reflects the guide raise; it should not erase the capital-allocation error.
Business delta: positive core profitability, mixed segment breadth. Estimate delta: FY1 EBIT/FCF rise materially; FY2 depends on charge-to-cash conversion. Stock delta: attractive cash yield, but not a clean BUY without the call bridge. Three call questions: What is the cash timing of the $4.1 billion charges? Why did Ford Pro EBIT fall 26%, and is the cause volume, mix, pricing, or cost? What warranty-cost rate is embedded in the $10–$11 billion guide? Confirmation: FY26 adjusted FCF ≥$6 billion and Pro margin stabilizes next quarter. Falsification: adjusted EBIT guide below $9.5 billion, FCF below $5 billion, or warranty costs rise YoY. Next proof: Q3, expected October 2026. 10-second PM line: Ford is cheap on forward cash; the JV exit proves that adjusted cash must be discounted for capital-allocation leakage.
PM decision line — PROVISIONAL — RELEASE ONLY. Bloom entered the print after a 26.3% five-day drawdown and two public short theses questioning backlog quality and the infrastructure/permitting path behind large data-center projects. The company reported $1.065 billion of revenue, up about 166% YoY, approximately $196 million of profit, $0.78 adjusted EPS versus roughly $0.41 consensus, and raised FY26 revenue to $3.9–$4.2 billion and adjusted EPS to $2.55–$2.85. The operating evidence is exceptional, but at the $166.84 pre-print close the stock already traded at about 61.8x the new EPS-guide midpoint before the positive initial reaction. Action: HOLD / WAIT; no chase; initiate only at or below $135 or after project-level backlog conversion and permit dependencies are quantified.
| Metric | Prior guide / Street | Actual / new guide | Classification | |---|---:|---:| | Q2 revenue | Approx. $759M public expectation | $1.065B | ~40% beat; STRUCTURAL POSITIVE | | Adjusted EPS | Collector $0.23; public bar about $0.41 | $0.78 | 90% above higher bar; STRUCTURAL POSITIVE | | FY26 revenue | $3.4–$3.8B after prior raise | $3.9–$4.2B | Midpoint +$450M; STRUCTURAL POSITIVE | | FY26 adjusted EPS | Prior detailed range not verified | $2.55–$2.85 | New earnings floor, quality bridge pending | | Non-GAAP gross margin | Prior framework >30% | About 34% reported | Operating leverage confirmed provisionally |
The EPS quality gate is PARTIAL because the complete GAAP/non-GAAP reconciliation, tax/share-count bridge, and customer concentration schedule were not retrievable. The swing to approximately $196 million of profit alongside 166% revenue growth argues that the beat is predominantly operating, but the brief will not infer the contribution of one-time, tax, or below-the-line items without the schedules.
Bloom monetizes deployed fuel-cell systems × system price + installation/service, less stack/material, manufacturing, and project execution cost. Its advantage in the AI-power bottleneck is deployment speed and on-site reliability versus multi-year grid interconnection or turbine queues. Revenue rising 42% sequentially and 166% YoY while non-GAAP margin reaches roughly 34% suggests manufacturing absorption and mix are working together; the result is not just more units at flat economics.
The buried signal is management’s claim that backlog is growing faster than revenue and customers are placing longer-duration orders. If verified contractually, that changes the debate from a one-project revenue spike to a multi-year demand floor. But it is also the highest-risk claim: backlog without deposits, site control, gas supply, permits, or cancellation penalties is not equivalent to financed revenue. The New Mexico pipeline/permitting issue cited before the print demonstrates why project-level dependencies must be separated from headline backlog.
The new FY26 revenue midpoint of $4.05 billion is $450 million above the prior $3.60 billion midpoint. At a 34% non-GAAP gross margin, the incremental guide contributes about $153 million of gross profit before operating expense. The $2.70 EPS midpoint × an assumed 300 million diluted shares implies about $810 million of adjusted net income; the exact share count must be verified. For FY2, the bridge is contracted MW × realized $/kW × installation cadence × gross margin − service/warranty cost − opex. A 100-basis-point margin change on $4.05 billion is $40.5 million of operating income.
At $166.84, the new midpoint is 61.8x EPS; $135 is 50x. A bull case of $4.00 FY27 EPS at 60x gives $240; base $3.25 at 45x gives $146; bear $2.00 at 30x gives $60. Weights of 25%/50%/25% yield about $148, below the pre-print close. That is why a spectacular print does not automatically make the stock a BUY.
The missing-information matrix is economically decisive. Contract quality requires backlog by customer, deposit status, cancellation penalties, and project-level notice-to-proceed. Delivery capacity requires factory utilization, stack/component availability, installation labor, and working-capital days. Project risk requires site control, gas/power interconnection, emissions permits, and the share of FY26/FY27 revenue dependent on each unresolved approval. Those data should be available first in the complete Q&A and then in the Q3 filing. Until they are disclosed, the analysis assumes neither that every backlog dollar converts nor that the short thesis is correct; it caps conviction and demands a price cushion.
| Live claim | Required evidence | Evidence received | Verdict |
|---|---|---|---|
| AI data centers create demand faster than grid supply | Revenue/backlog acceleration | Revenue +166%; backlog said to outgrow revenue | Strengthened |
| Backlog is high quality and diversified | Deposits, cancellation terms, project/permit detail | Management assertion only in available evidence | Unresolved |
| Scale drives sustainable margin | Sequential revenue and gross-margin leverage | Revenue +42% QoQ; ~34% non-GAAP GM | Strengthened provisionally |
Thesis matrix: demand IMPROVED; pricing/mix IMPROVED; margin IMPROVED; competition UNRESOLVED; balance sheet/capital needs UNRESOLVED; management credibility PENDING — CALL; catalyst timing IMPROVED. The narrative moved from “backlog and permitting could expose an H2 math problem” to “current revenue and margin make the guide credible, but contract quality still determines durability.” Bears must now attack conversion quality rather than deny current demand.
Business delta: materially positive. Estimate delta: large FY1 raise and positive FY2 carry. Stock delta: the initial rally is justified directionally, but valuation remains the gating risk. Three call questions: What percentage of backlog has deposits or cancellation penalties? What revenue and margin depend on the New Mexico project or any single customer? What capacity, stack supply, and working-capital investment is needed to deliver $4.05 billion? Confirmation: FY26 revenue ≥$3.9 billion, non-GAAP GM ≥32%, and no material permit-driven slippage. Falsification: guide below $3.6 billion, gross margin below 28%, or a top project delayed beyond two quarters without replacement. Next proof: Q3, expected October 2026. 10-second PM line: Bloom proved the power bottleneck is revenue; it has not yet proved every backlog dollar is financeable and permit-ready.
PM decision line — PROVISIONAL — RELEASE ONLY. The pre-print bar was elevated after a 268% one-year stock return and a first-quarter record; the company had guided Q2 revenue to $1.15–$1.25 billion and adjusted EPS to $1.86–$2.15. Teradyne delivered $1.329 billion and $2.47, its second consecutive record-revenue quarter, with memory test at a record as DRAM strength broadened into NAND final test. Q3 revenue guidance of $1.20–$1.30 billion also exceeded a roughly $1.04 billion public expectation. Action: HOLD; new money only at or below $280, or after complete Q&A establishes that memory-test demand and mix persist into 2027.
| Metric | Prior guide / Street | Actual / new guide | Classification | |---|---:|---:| | Q2 revenue | $1.15–$1.25B | $1.329B | $79M above high end; STRUCTURAL POSITIVE | | Q2 adjusted EPS | $1.86–$2.15 | $2.47 | $0.32 above high end; STRUCTURAL POSITIVE | | Q2 GAAP EPS | $1.83–$2.12 guide | $2.38 | High-end beat; STRUCTURAL POSITIVE | | Q3 revenue | Public expectation about $1.04B | $1.20–$1.30B | Midpoint ~20% above bar; STRUCTURAL POSITIVE |
Revenue rose 104% YoY and adjusted EPS more than 300%. The adjusted/GAAP difference was only $0.09 per share, principally acquired-intangible amortization, restructuring/other charges, and related tax effects. That small gap means the beat is high quality under the 30% rule: one-time/below-line exclusions do not explain the magnitude of the EPS variance.
Semiconductor Test produced $1.122 billion, Product Test $107 million, and Robotics $100 million. The causal engine is test-cell demand × tester content per device × utilization/upgrade rate, with mix and installed-base service determining margin. AI accelerators raise complexity and test time; HBM/DRAM and NAND final test expand memory tester demand; advanced packaging creates more insertion points. Revenue doubling while adjusted EPS quadruples indicates powerful fixed-cost absorption and favorable semiconductor-test mix.
The buried signal is not merely “AI.” It is NAND final-test resurgence alongside DRAM strength. A DRAM-only boom could be a narrow HBM cycle; NAND participation broadens the memory recovery and makes the test upcycle less dependent on one product. The risk is cyclicality: tester orders can be pulled forward, and a record quarter plus a sequentially lower Q3 midpoint could indicate normalization even while YoY comparisons remain strong.
The Q2 revenue beat versus the high end was $79 million. At an illustrative 55% incremental gross margin and limited variable opex, that can explain roughly $35–$45 million of incremental after-tax profit, or about $0.22–$0.28 per share on roughly 160 million shares—consistent with much of the $0.32 EPS beat. This is a sensitivity, not a company-provided bridge.
For FY1, the Q3 midpoint of $1.25 billion preserves a $5.0 billion annualized revenue run rate, far above the prior cycle trough. FY2 algebra is AI/compute test intensity + memory unit/capacity growth + advanced-packaging insertions + robotics recovery − customer capex cyclicality. A 10% change in Semiconductor Test revenue at the current $1.122 billion quarterly base equals about $449 million annualized before mix and margin.
At $320.65, annualized Q2 adjusted EPS of $9.88 implies 32.5x; $280 implies 28.3x. Bull/base/bear cases of $12×35, $10×28, and $7×22 yield $420, $280, and $154. At 25%/50%/25%, value is about $284, below the pre-print close. The earnings beat raises the floor, but the pre-print stock already capitalized a long duration of AI-test strength.
The missing-information matrix focuses on durability rather than the reported beat. First, customer concentration and tester acceptance timing are needed to distinguish installed demand from pull-forward; those should appear in full Q&A or the Q3 filing. Second, memory revenue must be separated into HBM/DRAM and NAND final test, with utilization and backlog, to judge whether breadth is real. Third, Robotics revenue of $100 million needs orders, book-to-bill, and margin because it remains a different cycle from semiconductor test. Fourth, gross-margin guidance and tester/service mix are needed to translate the Q3 revenue range into FY1 EPS without extrapolating Q2’s unusually strong operating leverage.
| Live claim | Required evidence | Evidence received | Verdict |
|---|---|---|---|
| AI complexity creates a structural test-content cycle | Record semi-test revenue and forward guide | $1.122B semi test; Q3 guide above Street | Strengthened |
| Memory breadth is improving | DRAM plus NAND demand | Record memory; NAND final test resurgence | Strengthened |
| Growth is not simply peak-cycle pull-forward | Backlog, utilization, customer/order timing | Not available without Q&A | Unresolved |
Thesis matrix: demand IMPROVED; pricing/mix IMPROVED; margin IMPROVED; competition UNRESOLVED; balance sheet UNCHANGED; management credibility PENDING — CALL; catalyst timing IMPROVED. The narrative moved from “priced-for-perfection AI tester” to “numbers are still outrunning the valuation debate.” The stock can be right on the business and still offer poor forward return if the multiple compresses.
Business delta: materially positive. Estimate delta: FY1 rises; FY2 rises but remains cyclical. Stock delta: the initial rally risks over-discounting the beat. Three call questions: How much Q2/Q3 semi-test revenue is HBM/DRAM versus NAND and compute? What backlog or customer commitments support 2027? Did any customer pull forward tester acceptance from H2? Confirmation: Q3 revenue ≥$1.20 billion and Semiconductor Test ≥$1.0 billion. Falsification: Q4 guide below $1.0 billion or memory revenue declines more than 25% sequentially without a timing backlog. Next proof: Q3, expected October 2026. 10-second PM line: record memory breadth validates the engine; the price still assumes the cycle lasts.
PM decision line — PROVISIONAL — RELEASE ONLY. The pre-print stack was Q2 revenue guidance of $922–$932 million, adjusted EBITDA of $160–$180 million, adjusted EPS of $0.27–$0.30, and FY26 revenue of $3.78–$3.82 billion. CoStar delivered about $925 million revenue, up 18%, net income of $55 million versus $6 million, and adjusted EPS of $0.32, demonstrating a real profitability inflection. But it guided Q3 revenue to $935–$945 million and cut FY26 revenue to $3.72–$3.76 billion, below both the prior company range and a roughly $967.7 million public Q3 consensus. Action: WAIT; initiate at or below $25 or after annualized net new bookings and residential-marketplace monetization reaccelerate.
| Metric | Prior guide / Street | Actual / new guide | Classification | |---|---:|---:| | Q2 revenue | $922–$932M | ~$925M | Near midpoint; NOISE / IMMATERIAL | | Q2 adjusted EPS | $0.27–$0.30 | $0.32 | $0.02 above high end; STRUCTURAL POSITIVE | | Q3 revenue | Public consensus ~$967.7M | $935–$945M | Midpoint ~$27.7M below; STRUCTURAL NEGATIVE | | FY26 revenue | $3.78–$3.82B | $3.72–$3.76B | Midpoint -$60M; STRUCTURAL NEGATIVE unless timing proven | | Net income | $6M prior-year quarter | $55M | +817%; STRUCTURAL POSITIVE |
The collector’s $897 million “revenue actual” was rejected as stale Q1 data. The current public result detail supports roughly $925 million. Adjusted EPS of $0.32 versus GAAP earnings around $0.13 per share indicates that adjustments remain material; although profitability is improving, the complete SBC/amortization/acquisition bridge is needed before the EPS-quality gate can pass.
CoStar’s engine combines high-retention subscription data/analytics with advertising/marketplace monetization. The causal formula is subscribers × price/retention + marketplace audience × advertiser conversion − residential acquisition/traffic investment. Commercial Real Estate generates the cash; Homes.com, Domain, and other residential assets consume investment while building audience and monetization. Q2’s 18% revenue growth and more-than-doubling adjusted EBITDA show the fixed-cost base can lever, but the Q3/FY revenue reset says top-line conversion is slower than the market expected.
The buried signal is the separation between profit and revenue guidance. Cost discipline or slower investment can lift EBITDA and EPS even while bookings/revenue slow. That improves near-term cash but does not automatically validate marketplace competitive strength. If Homes.com audience is not converting into durable agent/advertiser bookings, cost cuts can make FY1 look better while reducing FY2 strategic value.
The FY26 revenue midpoint fell $60 million, from $3.80 billion to $3.74 billion. At a 75%–80% incremental gross margin typical of subscription/data revenue, that removes roughly $45–$48 million of gross profit before any expense offset. Q3 adjusted EBITDA guidance of $190–$210 million suggests cost leverage can absorb much of the revenue shortfall, but the exact FY EBITDA/EPS revision was not verified.
FY1 algebra is net new bookings × conversion timing + renewal/price + acquired contribution − residential sales churn, then gross profit less traffic acquisition and sales/marketing. FY2 depends on whether Homes.com and Domain reduce loss through monetization rather than only spending restraint. A 1-point revenue-growth change on a $3.74 billion base is $37 million; at 75% incremental margin, about $28 million of EBITDA.
At $30.33, a $1.40–$1.60 normalized adjusted-EPS range implies 19–22x, but the stock should not be valued only on P/E while residential losses and acquisition integration distort earnings. A $25 trigger provides about 18x $1.40 and requires proof that the core subscription franchise is not decelerating. Bull/base/bear price cases of $40/$28/$18 weighted 20%/50%/30% yield $27.40.
The missing-information matrix has three branches. Core quality requires renewal rate, price realization, subscription seat growth, and annualized net new bookings. Residential monetization requires Homes.com and Domain audience, advertiser count, ARPU, sales productivity, and customer acquisition cost—not traffic alone. Profit quality requires the full reconciliation of GAAP earnings to adjusted EBITDA/EPS, segment losses, acquisition integration cost, and any spending delayed rather than removed. Full Q&A is the first source; Q3 bookings and segment disclosure are the first objective test. Without those data, the FY revenue reduction cannot be classified as a reversible timing item, so the brief treats it as structural until proved otherwise.
| Live claim | Required evidence | Evidence received | Verdict |
|---|---|---|---|
| Core subscriptions can fund residential expansion | Revenue growth plus EBITDA leverage | +18% revenue; EBITDA more than doubled | Strengthened |
| Homes.com/Domain monetization is on plan | Net new bookings, advertiser conversion, ARPU | Not available in release evidence | Unresolved |
| Profitability can rise without harming growth | Stable/raised revenue plus margin | Revenue guide cut while profit improved | Weakened |
Thesis matrix: demand WEAKENED; pricing/retention UNRESOLVED; margin IMPROVED; competitive position UNRESOLVED; balance sheet UNCHANGED; management credibility PENDING — CALL; catalyst timing DELAYED. The narrative moved from “profitability inflection plus durable high-teens growth” to “profitability inflection, but the revenue ramp slipped.” The approximately 13% initial decline is directionally consistent with a duration-stock guide cut, but it is not yet a settled price.
Business delta: mixed—profit engine better, revenue conversion worse. Estimate delta: FY1 revenue down, EPS/EBITDA potentially buffered; FY2 risk rises. Stock delta: reaction appears justified unless Q&A identifies a discrete timing reversal. Three call questions: Which product/geography explains the $60 million FY revenue cut? What are annualized net new bookings and Homes.com monetization conversion? Is the EBITDA improvement cost removal, delayed spend, or unit-economics improvement? Confirmation: Q3 revenue ≥$945 million and annualized net new bookings reaccelerate YoY. Falsification: FY revenue below $3.72 billion or residential losses improve only through audience-investment cuts. Next proof: Q3, expected October 2026. 10-second PM line: CoStar proved it can earn; it did not prove the residential revenue ramp is intact.
The deferred cohort receives no fresh BUY/SELL verdict. All remain outstanding obligations.
| Cohort | Known target-date fact | Exact blocked input | Status / deadline |
|---|---|---|---|
| KLAC | Reporter/call verified; reported large negative initial move | Current issuer earnings release, shareholder letter, correct split-adjusted actual/guide, complete current Q&A, prior full call | ESCALATED; 2026-07-29 08:00 ET |
| WM, MDLZ, NXPI, ACGL, EXR, FE, PPG, OMC, VLTO, EXE, WPC, LOGI, UNM, CR, BXP, UMBF, SWKS, MANH, AXS, ZWS, QRVO, STAG, THG | Collector detected result or headline evidence | Full issuer release/schedules, dated consensus range/hurdle, current full Q&A, prior-call comparison, reliable settled reaction | ROLLED; 2026-07-29 08:00 ET |
| NOV, CHE, LSTR, CZR, RUSHA, RUSHB, CAR, VRNS, ENPH, EXLS, CAKE, MIR, MEOH, RNST, HIW, GEF, GEF.B, CGAU, APAM, SBCF, TTAM, ASH, AKR, ACHC, TRMK, BUSE, WERN, SLDE, ROG, MBIN, FCF | Calendar qualification only | Actual issuer release, expectations stack, market reaction, current/prior call record | ROLLED; 2026-07-29 20:00 ET |
The morning brief’s [[GLW]] work already cleared the full-call gate and remains CONFIRMED: the operating beat and capacity-constrained optical demand were real, but Q3 guidance did not clear the valuation-implied acceleration hurdle. No new evidence changes the WAIT-below-$110 action.
For the seven morning provisional Tier 1 names, no complete current and prior Q&A pair was verified by the PM cutoff. The required judgment is therefore PARTIAL, not “confirmed”: the release conclusion has not been falsified, but the call-dependent claims are still open.
| Ticker | Morning view | PM judgment | Exact unresolved call input | Action unchanged |
|---|---|---|---|---|
| BA | Positive FCF milestone; wait ≤$200 | PARTIAL | Full current/prior Q&A; production cadence, Defense charge, FCF durability | WAIT ≤$200 |
| UPS | Guide raise; mix/margin quality unresolved | PARTIAL | Full current/prior Q&A; volume-price bridge, network savings, Amazon glide-down | HOLD / no new money >$100 |
| RCL | Demand beat and guide raise substantially priced | PARTIAL | Full current/prior Q&A; close-in pricing, cost/JV contribution, booking curve | HOLD / wait ≤$285 |
| SHW | Share gain and margin beat, weak end markets | PARTIAL | Full current/prior Q&A; volume/price split, housing sensitivity, FY bridge | HOLD / wait ≤$325 |
| CARR | Orders strong, margin conversion weak | PARTIAL | Full current/prior Q&A; data-center order quality, mix, input-cost/margin bridge | WAIT ≤$58 or margin stabilization |
| IQV | Book-to-bill supports CRO recovery | PARTIAL | Full current/prior Q&A; backlog cancellations/conversion, bookings durability | HOLD / wait ≤$220 |
| TRU | Broad growth but margin/consumer mix mixed | PARTIAL | Full current/prior Q&A; U.S. financial-services durability, consumer interactive, margin | HOLD / wait ≤$78 |
Analytical Ledger reconciliation: no open Ledger entry matched the six AMC underwrites or the eight BMO reconciliation names. No catalyst row was therefore eligible for mutation; the existing Ledger was left unchanged rather than creating or silently replacing a thesis from provisional evidence.
| Priority | Ticker(s) | Required source / question | Deadline |
|---|---|---|---|
| 1 | KLAC | Correct issuer financial package, split-adjusted guide, complete current/prior Q&A | 2026-07-29 08:00 ET |
| 2 | V, STX, F, BE, TER, CSGP | Complete current transcript with Q&A, prior full call, exact language/commitment deltas | 2026-07-29 08:00 ET |
| 3 | BA, UPS, RCL, SHW, CARR, IQV, TRU | Complete current/prior Q&A to close morning provisional status | 2026-07-29 08:00 ET |
| 4 | Result-detected Tier 3 cohort | Issuer package, dated expectations, settled price reaction, call record | 2026-07-29 08:00 ET |
| 5 | Calendar-only Tier 3 cohort | Validate actual report and >$2B eligibility, then acquire source stack | 2026-07-29 20:00 ET |
| Ticker | Tier | Status | Analytical words | Causal KPIs | Q&A | Sourced debate claims | Prior-call deltas | Omissions | FY1/FY2 bridge | Transcript provenance | Sentiment | Tone delta | Answer quality | Pressure delta | Tracker read-back | Failed/deferred gates |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| V | 1 | PROVISIONAL — RELEASE ONLY | >900 | 7 | 0 | 3 | 0 | 3 call questions | Complete sensitivity | Current/prior full Q&A absent | PENDING_TRANSCRIPT | n/a | n/a | n/a | Required | Call-only gates |
| STX | 1 | PROVISIONAL — RELEASE ONLY | >900 | 7 | 0 | 3 | 0 | 3 call questions | Complete sensitivity | Highlights only; not a transcript | PENDING_TRANSCRIPT | n/a | n/a | n/a | Required | Call and EPS-reconciliation gates |
| F | 1 | PROVISIONAL — RELEASE ONLY | >850 | 7 | 0 | 3 | 0 | 3 call questions | Complete sensitivity | Current/prior full Q&A absent | PENDING_TRANSCRIPT | n/a | n/a | n/a | Required | Call and charge-cash bridge |
| BE | 1 | PROVISIONAL — RELEASE ONLY | >700 | 7 | 0 | 3 | 0 | 3 call questions | Complete sensitivity | Current/prior full Q&A absent | PENDING_TRANSCRIPT | n/a | n/a | n/a | Required | Call, backlog-quality, GAAP bridge |
| TER | 1 | PROVISIONAL — RELEASE ONLY | >700 | 7 | 0 | 3 | 0 | 3 call questions | Complete sensitivity | Current/prior full Q&A absent | PENDING_TRANSCRIPT | n/a | n/a | n/a | Required | Call-only gates |
| CSGP | 1 | PROVISIONAL — RELEASE ONLY | >700 | 7 | 0 | 3 | 0 | 3 call questions | Complete sensitivity | Highlights only; not a transcript | PENDING_TRANSCRIPT | n/a | n/a | n/a | Required | Call, bookings, GAAP bridge |
| KLAC | 3 | DEFERRED / ESCALATED | 190 | 0 | 0 | 0 | 0 | n/a | Failed | Current package contaminated by stale data | DEFERRED | n/a | n/a | n/a | Required | Source, model, call |
| Other AMC reporters | 3 | DEFERRED | Coverage rows | 0 | 0 | 0 | 0 | n/a | Failed | Missing or highlights only | DEFERRED | n/a | n/a | n/a | Required | Source/expectations/call gates |
| GLW BMO catch-up | 1 | FINAL — POST CALL / CONFIRMED | See AM report | 7 | 6 | 4 | 3 | 2 | Complete | Full current/prior transcripts in AM | SCORED in AM | 0 | 62 | -25 | AM read-back reverified | None |
| BA, UPS, RCL, SHW, CARR, IQV, TRU | 1 | PROVISIONAL / PARTIAL | See AM report | 7 each | 0 | 3 each | 0 | 3 each | Complete sensitivity in AM | Full pairs absent | PENDING_TRANSCRIPT in AM | n/a | n/a | n/a | AM read-back reverified | Call-only gates |
/Users/max/morningsignal-research/state/earnings/earnings_context_2026-07-28_PM.json, generated 2026-07-28 20:03 ET.EarningsBrief/EarningsBrief_2026-07-28_AM.md.not verifiable.