MorningSignal
Research Platform
2026-09-28 13:40
type: earnings-brief session: AM date: 2026-07-23 daily_note: "[[Daily/2026-07-23]]" tags: [sellside, earnings, morning]
Back to [[Daily/2026-07-23]].
The morning is a positive industrial-demand tape and a negative duration/cash-quality tape. [[RTX]] and [[Lockheed Martin|LMT]] converted record backlogs into double-digit growth, raised guidance, and gained roughly 6% in early trading. [[Thermo Fisher Scientific|TMO]] also beat and accelerated, while [[T-Mobile US|TMUS]] raised free-cash-flow guidance despite slower subscriber growth. The common positive is real volume plus better forward estimates; the common caveat is that several cash-flow prints contain timing or merger effects and none of today's major calls had a fully reviewed transcript plus Q&A at the 08:00 ET decision cutoff.
The prior-evening catch-up changes three judgments. [[Alphabet|GOOG/GOOGL]] is now FINAL — POST CALL: Search and Cloud demand strengthened, but management gave no numerical return hurdle for a $195-$205 billion capex plan. [[Tesla|TSLA]] is now FINAL — POST CALL: volume recovered, but the transcript confirms that margin, free cash flow and autonomy economics remain the constraint. [[ServiceNow|NOW]] is now FINAL — POST CALL: AI monetization is measurable and cRPO held, while roughly half the quarterly beat was federal timing. [[Texas Instruments|TXN]] and [[IBM]] remain PROVISIONAL because a complete current Q&A record and prior-call comparison were not acquired.
Action changes
The deterministic AM collector generated its evidence bundle at 08:03 ET. It identified 49 qualifying US-listed BMO reporters above $2 billion, 30 with current-result evidence. No qualifying name matched an open position or call in the TIF Analytical Ledger. Tiering therefore follows market cap, causal read-through, reaction and primary-source sufficiency rather than presumed ownership.
| Company | Session obligation | Tier / status | Why |
|---|---|---|---|
| [[Alphabet | GOOG/GOOGL]] | prior AMC | T1 — FINAL POST CALL |
| [[Tesla | TSLA]] | prior AMC | T1 — FINAL POST CALL |
| [[RTX]] | current BMO | T1 — PROVISIONAL | $261B issuer; commercial aerospace and defense read-through; call record incomplete at cutoff |
| [[Lockheed Martin | LMT]] | current BMO | T1 — PROVISIONAL |
| [[ServiceNow | NOW]] | prior AMC | T2 — FINAL POST CALL |
| [[T-Mobile US | TMUS]] | current BMO | T2 — PROVISIONAL |
| [[Thermo Fisher Scientific | TMO]] | current BMO | T2 — PROVISIONAL |
| [[Texas Instruments | TXN]] | prior AMC | T2 — PROVISIONAL |
| [[IBM]] | prior AMC | T2 — PROVISIONAL | enterprise AI/software and guide-cut read-through; complete Q&A unavailable |
| Remaining current BMO names | current BMO | T3 — COVERAGE LEDGER | verified discovery status and exact missing evidence below |
| Remaining prior AMC names and older BMO queues | catch-up | T3 — ROLLED | still lack complete primary expectations/call/model gates |
Pre-print stack. The prior guide was $180-$190 billion of 2026 capex and a capacity-constrained Cloud business. Visible Alpha's dated pre-print stack carried approximately $117.2 billion of revenue, $3.32 of GAAP EPS and 30.8% Cloud operating margin; a separate FactSet public point was $117.06 billion of revenue and $2.88 of adjusted EPS. No verified buy-side whisper was available, so none is imputed. The valuation-implied bar was not merely a revenue beat: a roughly $4.2 trillion market value required the company to prove that AI infrastructure could accelerate Cloud without destroying incremental returns. There was no open TIF Ledger threshold.
Print and variance. Revenue was $119.80 billion, +24% year over year, about $2.6 billion above the Visible Alpha point. Search and other revenue rose 17% to $63.27 billion; YouTube advertising rose 13% to $11.06 billion. Cloud revenue accelerated 82% to $24.77 billion, and Cloud operating margin reached 35.6%, roughly 480 basis points above the pre-print point. Headline GAAP EPS of $9.11 was not operating earnings: approximately $6.26 per share came from an unrealized equity gain, leaving an operating-equivalent result near $2.85. Quarterly operating cash flow was $39.07 billion, but $44.92 billion of capex produced approximately negative $5.86 billion of free cash flow. Management raised 2026 capex to $195-$205 billion.
Call read and Q&A forensics. The complete current transcript was reviewed, including the full Q&A, and compared with Alphabet's official Q1 transcript. Management disclosed that model APIs were processing about 22 billion tokens per minute versus 16 billion in Q1, AI Mode had approximately one billion monthly users, and roughly nine million developers were building with Gemini. Cloud backlog was cited around $514 billion, and management again described supply as constrained. The crucial change from Q1 is not demand—it is the explicit willingness to bridge near-term shortages with third-party capacity and carry substantially more capex into 2027.
Three exchanges mattered:
The most important prior-call delta is that Q1 framed Cloud as demand-constrained and capital allocation as capacity expansion. Q2 adds an explicit near-term third-party capacity bridge, higher FY2026 spending, significant FY2027 investment and modest margin pressure. Demand proof strengthened; return visibility weakened.
Operating-quality reconciliation. The revenue beat deserves more weight than headline EPS because it came from Search and Cloud rather than a low-quality tax item, while the EPS beat deserves materially less weight because the equity mark is not repeatable. The internal economic tension is visible in the cash statement: a business generating $39.1 billion of quarterly operating cash still consumed cash after investment. That does not make the investment wrong, but it changes the valuation anchor from near-term earnings to an eventual return on installed compute. Depreciation will lag the cash outlay and can pressure future reported margin even if demand remains excellent. Third-party capacity can accelerate revenue before owned capacity comes online, but it is likely more expensive and may explain management's warning about modest pressure. A correct FY1 model therefore raises gross profit and depreciation together and does not flow the full revenue beat to free cash flow.
What the stock is discounting. At the pre-print equity value, a conventional “beat and raise” is insufficient. The stock needs Search growth to remain in the mid-teens, Cloud to retain hyperscaler-like growth with software-like incremental margin, and capex intensity to crest before Cloud growth normalizes. If Cloud revenue growth were to fall toward 35% while annual capex remained near $200 billion, the market would likely de-rate the consolidated multiple even if accounting EPS rises. Conversely, Cloud growth above 50%, margin above 32% and a visible reduction in capex/revenue would validate management's investment choice. That is why the next two quarters' utilization, backlog conversion and cash conversion matter more than the one-quarter EPS headline.
Information still owed. The call did not provide a capacity-utilization bridge, useful-life assumption sensitivity, external versus internal TPU allocation, or incremental Search AI serving cost. It also did not quantify how much of the $514 billion backlog is committed, cancellable or tied to third-party capacity. These are not cosmetic omissions: each determines the durability and capital efficiency of the Cloud acceleration. The debate should remain open even though the operating result was strong.
Debate and falsification.
FY1/FY2 bridge. FY1 revenue and Cloud estimates should rise because both Search and Cloud exceeded the pre-print bar and capacity additions support the second half. Operating EPS moves less because depreciation, third-party capacity and broader AI expense offset gross-profit upside. FY1 free cash flow must move down for the $15 billion midpoint capex increase. FY2 revenue can rise if backlog converts, but the value of that revenue depends on Cloud margin holding above 32% and capex growth decelerating. A sensitivity of 300 basis points on FY2027 Cloud margin and $15-$20 billion on annual capex is more important than the Q2 revenue beat for equity value.
Scenario map. In a bull case, Search stays at least +15%, Cloud remains above +50% through the next two quarters, Cloud margin holds 34%-36%, and 2027 capex growth slows; estimates and multiple can rise together. In the base case, Search settles at 12%-15%, Cloud decelerates toward 40%, margin holds near 32%-34% and capex remains elevated; EPS rises but valuation remains capped by cash intensity. In the bear case, third-party capacity raises cost before owned capacity is utilized, Cloud falls below 35%, Search AI costs dilute Services margin and capex stays near $200 billion; free-cash-flow yield becomes the binding valuation constraint. The $300 add level is meant to improve asymmetry against that base/bear range, not to predict a precise bottom.
Thesis delta. Demand/volume improved; pricing/mix improved; Cloud margin architecture improved; competitive position reinforced; capital allocation weakened near term; management credibility unchanged because operating delivery was excellent but the return framework remained qualitative. Old narrative: AI capex was necessary to defend Search and catch hyperscalers. New narrative: Search and Cloud demand are demonstrably strong, but the stock now debates how much of the growth converts to free cash flow.
Decision: HOLD; add only at $300 or below with Cloud margin at least 32% and a stabilizing FY2027 free-cash-flow path. Business delta: positive. Estimate delta: revenue/EPS positive, FCF negative. Stock delta: mixed.
Pre-print stack. Tesla's company-compiled July 17 consensus expected $27.584 billion of revenue, $20.048 billion automotive revenue, $3.773 billion energy revenue, 19.5% GAAP gross margin, 5.4% operating margin, $0.55 non-GAAP EPS, $3.445 billion operating cash flow and negative $3.254 billion free cash flow. No independent buy-side hurdle was verified. The valuation-implied bar for a roughly $1.4 trillion company was commercial autonomy proof plus evidence that more than $25 billion of annual capex would not permanently suppress free cash flow. There was no open TIF Ledger threshold.
Print and variance. Revenue of $28.236 billion beat company consensus by $652 million, and automotive revenue of $20.516 billion beat by $468 million. Deliveries rose 25% year over year. Energy revenue of $3.139 billion missed by $634 million despite 13.5 GWh deployed. Services and other revenue beat by $818 million. The economic miss was material: GAAP gross margin was 16.8%, 270 basis points below consensus; operating margin was 1.4%, 400 basis points below; non-GAAP EPS was $0.33 versus $0.55. Operating cash flow beat the point, but $5.789 billion of capex left free cash flow at negative $1.092 billion. The print was therefore volume-positive, price/mix-negative and cash-negative in absolute terms.
Call read and Q&A forensics. The complete current transcript and Q&A were reviewed and compared with a complete Q1 transcript source. Automotive gross margin excluding credits was 16.3% versus 19.2% in Q1. Management said Q1 benefited by about $230 million from warranty and tariff items that did not repeat, implying a less severe normalized sequential deterioration, but the absolute margin remains below the thesis threshold. Energy margin fell to 20.4% from 39.5%; management cited a roughly $240 million warranty true-up this quarter and more than $200 million of nonrecurring tariff benefit in Q1, then framed a normalized longer-term margin in the low-to-mid 20s. Services margin rose to a record 14.1% from 9.2%.
FSD had approximately 1.5 million paid customers, split roughly 55% upfront and 45% subscription, and more than 55% of North American deliveries included FSD at delivery. Robotaxi vehicles had accumulated about 380,000 unsupervised miles across six cities and two states with no notable incident according to management. These are operating milestones, not commercial economics: the company did not disclose rides, revenue, ARPU, intervention frequency or contribution margin. Capex more than doubled sequentially, and management explicitly said it would accept lower capital efficiency to move faster. Net income also contained an approximately $1 billion SpaceX mark-to-market benefit, partly offset by foreign exchange and Bitcoin losses, so headline earnings overstate operating quality.
Five exchanges define the call:
The prior-call delta is subtle but negative for the stock debate. Q1 already warned that more than $25 billion of 2026 capex would drive negative free cash flow. Q2 confirms the burden, while auto and energy margins normalize sharply lower and the autonomy discussion still lacks commercial-unit economics. The volume recovery improves factory absorption; it does not yet finance the AI build.
Operating-quality reconciliation. The quarter contains three separate earnings streams that should not be valued alike. Core automotive volume recovered, but price/mix and cost held ex-credit margin below the threshold required to fund the rest of the company. Energy deployment was exceptional, but accounting margin swung because both Q1 and Q2 contained large, opposite one-time items; a low-to-mid-20s normalized margin is a more defensible base than either quarterly endpoint. Services is the cleanest current improvement because revenue and margin advanced together, but it remains too small to offset the capex program. Finally, the SpaceX mark-to-market gain belongs below operating value and should not be capitalized as Tesla earning power.
Autonomy evidence hierarchy. Unsupervised miles and geographic expansion are leading operational indicators. Paid rides, utilization, disengagement/intervention rates, insurance losses and contribution profit are the commercial proof. Management disclosed the first layer and not the second. The correct analytical treatment is therefore to raise the probability of technical deployment without raising near-term robotaxi profit materially. The “march of nines” answer reinforces that reliability improves nonlinearly, but without a current baseline it cannot be translated into a launch date or reserve requirement. The company may be ahead technically and still miss the equity market's implied earnings schedule.
Capital-allocation test. Management's willingness to accept lower efficiency for speed can be rational where platform leadership has winner-take-most properties. It also removes a traditional downside guardrail. Investors need project-level milestones: factory utilization, Cybercab production cost, paid fleet utilization, AI5 yield and inference cost per mile. Until those arrive, the cash outlay should be modeled as an option premium rather than a high-confidence return project. A second consecutive negative-FCF quarter would not itself falsify the long-run thesis—management already guided to it—but it would raise financing and dilution sensitivity if core margins remain below 5% operating.
Debate and falsification.
FY1/FY2 bridge. FY1 revenue can move modestly higher on deliveries and services, but gross-profit and EPS estimates should fall because auto and energy margins missed substantially. FY1 free cash flow remains negative. FY2 is a convex autonomy case: without disclosed robotaxi unit economics, a base case should not capitalize substantial service profit. At an 18% ex-credit auto margin and 5% operating margin, the core business starts funding a larger share of the build; below those levels, value depends increasingly on distant autonomy outcomes.
Scenario map. A bull case requires deliveries to hold their recovery, ex-credit auto margin to recover above 18%, energy margin to normalize above 22%, and paid robotaxi economics to appear before capex peaks. A base case carries volume growth but 16%-18% auto margin, low-20s energy margin and negative FY2026 FCF, giving little support to the present valuation from conventional earnings. A bear case combines renewed price cuts, auto margin below 16%, continued negative cash generation and delayed commercial autonomy. The $300 price gate does not make the conventional valuation cheap; it creates more room for milestone slippage while retaining upside to autonomy success.
Thesis delta. Demand/volume improved; pricing/mix weakened; margin/cost weakened; autonomy execution improved but remains financially unresolved; capital allocation weakened; management credibility unchanged because milestones advanced but economic disclosure did not. Old narrative: a Q1 order recovery plus autonomy launch could restore growth while capex ramps. New narrative: volume is back, but the company is deliberately exchanging current return efficiency for speed and has not yet proved autonomy monetization.
Decision: WAIT; no new money until price is at or below $300 and ex-credit automotive gross margin is at least 18%, operating margin at least 5%, and quarterly free cash flow positive. Business delta: mixed. Estimate delta: negative. Stock delta: negative.
Pre-print stack. The prior Q1 framework was $92.5-$93.5 billion of adjusted sales, $6.70-$6.90 of adjusted EPS and $8.25-$8.75 billion of free cash flow. A dated public consensus point was $22.88 billion of Q2 revenue and $1.66 of adjusted EPS; no verified range or buy-side hurdle was available. The valuation-implied bar required backlog conversion, commercial aftermarket durability and credible defense capacity—not just bookings. No open TIF Ledger threshold was found.
Actuals. Sales were $24.708 billion, +14% reported and +16% organic, about $1.83 billion above the public point. Adjusted EPS was $1.89, $0.23 or 13.9% above; GAAP EPS was $1.57 after $0.27 of acquisition-accounting adjustments and $0.05 of restructuring/other significant items. Operating cash flow was $3.547 billion and free cash flow $2.878 billion. Backlog reached $289 billion, +22% year over year, split $170 billion commercial and $119 billion defense. Versus Q1, sales rose approximately 11.8%, total backlog 6.6%, commercial backlog 4.9% and defense backlog 9.2%.
The segment evidence is unusually broad. Collins sales rose 8% reported and 13% organic to $8.210 billion; adjusted segment profit rose 10% to $1.370 billion and margin expanded 30 basis points to 16.7%. Commercial OE grew 26%, aftermarket 10% and defense 7%. Pratt sales rose 16% to $8.889 billion; adjusted profit rose 22% to $740 million and margin increased to 8.3%, with commercial aftermarket +25%, military +23% and commercial OE -8%. Raytheon sales rose 18% to $8.269 billion; profit rose 29% to $1.042 billion and margin reached 12.6%, driven by Patriot, Standard Missile and AMRAAM volume.
Guide and estimate bridge. FY adjusted sales rose to $95-$96 billion from $92.5-$93.5 billion; organic growth to 8%-9% from 5%-6%; adjusted EPS to $7.10-$7.25 from $6.70-$6.90; and free cash flow to $8.50-$8.75 billion. The midpoint sales raise is $2.5 billion and the EPS raise $0.375. FY1 estimates should move close to the new midpoint because three segments participated and backlog expanded. FY2 depends on aftermarket longevity, engine shop-visit throughput, Pratt warranty exposure and missile capacity. The quality of the FCF raise is less strong than the sales/EPS raise because the low end moved while the high end was unchanged.
Release-only debate. The bull case says commercial aerospace scarcity extends the aftermarket cycle while defense replenishment creates multi-year visibility. The release supports both. The bear case says supply chain and Pratt execution can prevent backlog conversion; commercial OE at Pratt fell 8%, so this remains the main internal warning. A 5.8%-6.0% premarket gain is directionally deserved, but it capitalizes part of the guide reset before management has answered the capacity and cash-conversion questions.
Quality and causality. This was not a single-program or price-only beat. Organic sales grew 16%, all three businesses grew, Collins and Raytheon expanded margin, and Pratt profit grew despite lower commercial OE. The causal chain is commercial fleet scarcity → more flying and older aircraft → shop visits and parts demand, plus geopolitical replenishment → missile orders → production volume. Backlog growth strengthens duration but does not guarantee margin: labor, castings, propulsion components and supplier qualification can turn booked demand into late delivery or cost overruns. The 8%-9% organic guide implies management expects the production system—not only pricing—to improve in the second half.
Model sensitivity. At the new guide midpoint, each 50 basis points of consolidated margin is economically material, but the release does not provide enough information to distinguish price/cost from productivity. FY1 should capture the sales and EPS midpoint increases while retaining a working-capital reserve. FY2 deserves upside only if Pratt shop-visit throughput rises without additional remediation expense and defense capacity additions earn acceptable margins. If backlog continues to rise while inventory and contract assets grow faster than sales, the apparent demand benefit will not convert to equity cash. That is the principal falsification monitor.
Required call questions: What portion of the $2.5 billion sales-guide midpoint raise is price, volume, FX and acquisition? What shop-visit throughput and powder-metal remediation assumptions underpin Pratt? Which bottlenecks constrain Patriot/AMRAAM/Standard Missile deliveries? How much of Q2 FCF was customer-advance or working-capital timing? What operating margin is embedded in the second-half guide?
Thesis delta: demand improved; backlog visibility improved; segment margins improved; cash conversion improved but not fully proven; capacity risk unchanged. Decision: HOLD; do not chase. Upgrade to ADD only after the complete call confirms capacity-supported conversion and durable FCF. Status remains PROVISIONAL because the 07:30 ET call transcript, full Q&A and prior-call wording comparison were not acquired and reviewed by cutoff.
Pre-print stack. Q1 sales were approximately $18.0 billion, EPS $6.44 and free cash flow negative $291 million; management had reaffirmed $77.5-$80.0 billion of sales, $29.35-$30.25 of EPS and $6.5-$6.8 billion of free cash flow. The dated Q2 consensus was about $19.37 billion of revenue and $7.23 of EPS. No verified buy-side hurdle or TIF Ledger threshold was available.
Actuals. Q2 sales were $20.063 billion, +11% year over year, about $693 million above consensus. EPS was $7.94, $0.71 above. The year-over-year EPS comparison is distorted because Q2 2025 included $1.6 billion of program losses plus $169 million of other charges. Operating cash flow was $3.235 billion and free cash flow $2.917 billion versus $201 million and negative $150 million last year; management explicitly attributed much of the cash improvement to customer-receipt timing and lower taxes. Backlog reached a record $230 billion after roughly $65 billion of orders, including a $35 billion multi-year THAAD award. Versus Q1, sales rose approximately 11.5% and FCF swung by $3.208 billion.
All segments grew. Aeronautics sales rose 9.3% to $8.112 billion; Missiles and Fire Control 19.5% to $4.101 billion; Rotary and Mission Systems 9.0% to $4.354 billion; Space 5.7% to $3.496 billion. MFC profit rose 24% to $594 million. Aeronautics and RMS profit comparisons are not clean because the prior-year period carried reach-forward losses.
Guide and estimate bridge. FY sales rose to $79.75-$81.75 billion from $77.5-$80.0 billion; segment operating profit to $8.5-$8.7 billion from $8.425-$8.675 billion; EPS to $29.95-$30.65 from $29.35-$30.25; and FCF to $7.0-$7.2 billion from $6.5-$6.8 billion. Capital expenditure fell to $2.0-$2.4 billion from $2.5-$2.8 billion. FY1 revenue/EPS and FCF estimates should rise, but the FCF improvement includes lower capex and working-capital timing rather than solely higher structural earnings. FY2 depends on whether missile-system capacity can translate the backlog without cost growth and whether F-35 production remains on schedule.
Release-only debate. The bull case is no longer simply geopolitical demand: orders are contractual, backlog is at a record and MFC growth is nearly 20%. The bear case is execution—fixed-price program risk, supplier constraints and the possibility that Q2 cash merely borrows from the second half. The 6.1% early gain recognizes a legitimate guide raise, but a clean call must separate throughput from favorable timing.
Quality and causality. The order intake validates demand, while MFC revenue and profit validate some physical conversion. The aeronautics and RMS profit rebound is less informative because last year's comparison carried large charges. The cleanest earnings evidence is therefore MFC, Space and the consolidated guide—not the 437% headline growth in net earnings. Backlog of $230 billion provides duration, yet the $35 billion THAAD award also concentrates execution risk in a long-cycle, supplier-intensive program. A funded multi-year award lowers demand uncertainty and raises schedule/cost importance.
Cash and model sensitivity. The FCF guide rose more than the segment-profit guide while capex guidance fell $450 million at the midpoint. That tells us a meaningful portion of the cash improvement is capital timing or lower investment rather than a proportional increase in operating earnings. FY1 estimates can adopt the new range; FY2 should not extrapolate the Q2 cash conversion rate. The critical bridge is customer advances plus milestone receipts, inventory growth and payable timing. If second-half operating cash fails to offset Q2 timing, the annual guide can still be met but the sustainable conversion ratio will be lower.
The premarket gain is therefore a rational estimate reset, not yet proof that LMT has escaped program-risk cyclicality or supplier constraints today.
Required call questions: What physical bottleneck limits THAAD, PAC-3 and precision-strike output? How much of $65 billion of orders contains advance funding? What portion of the FCF raise comes from capex reduction versus operating cash? Are prior reach-forward-loss programs on schedule and at what margin? What second-half F-35 delivery cadence is embedded in guidance?
Thesis delta: demand/backlog improved; munitions volume improved; cash improved with timing caveat; program execution not yet cleared; capital intensity improved. Decision: HOLD; do not chase. Upgrade only if the call supports sustainable production throughput and at least $7.0 billion of FY FCF. Status remains PROVISIONAL because the scheduled 08:30 ET call occurred after the 08:00 automation cutoff.
Pre-print stack. The prior framework called for approximately 20% constant-currency cRPO growth and continued operating-margin expansion. Public consensus was about $3.93 billion of revenue and $0.86 of adjusted EPS. The stock had closed at $95.46 after falling 6.5% before the print, so the implied bar was maintenance of 20% cRPO and proof that AI products were incremental rather than bundle rhetoric. No verified buy-side hurdle or Ledger threshold was available.
Actuals. Revenue was approximately $3.99 billion and adjusted EPS $0.90. Subscription revenue was $3.877 billion, +24.5% reported and +23% constant currency. cRPO reached $13.2 billion, about +21% constant currency; total RPO was approximately $29 billion. Renewal rate was 98%, 658 customers had more than $5 million of ACV, operating margin reached 29.5% and free-cash-flow margin 16%. The FY subscription-revenue midpoint increased about $15 million to $15.755-$15.770 billion, with operating margin 31.5% and FCF margin 35%.
Call and Q&A. The full current transcript and Q&A were reviewed. Management disclosed more than $1 billion of AI ACV, net-new AI ACV up more than 40% sequentially, a ninefold increase in agentic-AI customers in production over nine months, Pro Plus uplift above 30% and AI-native SKU uplift of 20%-30%. More than 40 Level 1 support customers were resolving 80%-85% of tickets without a human, cutting resolution time from roughly two days to about 20 minutes. These metrics move AI monetization from anecdote toward causality.
The caveat was equally explicit: roughly half the Q2 beat came from federal on-premise business pulled forward from Q3. Management said all net-new ACV outperformance was included in the FY guide and retained prudence for the back half. The Q3 guide is $3.975-$3.980 billion of subscription revenue, about 20% cRPO constant currency and 31% operating margin. Hyperscaler and AI expense pressure subscription gross margin near term, while management argues model choice and scale lower costs later.
Q&A grades: Goldman Sachs on Level 1 AI resolution received quantified customer, resolution and time-to-value evidence—A-. Wells Fargo on federal timing received a direct answer and pipeline context—A-, though no exact contract list. Citi on license/usage architecture received specific uplift ranges and a defensible hybrid-pricing explanation—A-. Deutsche Bank on AI gross margin received the mechanism but no magnitude/timeline—B-.
Compared with Q1, when delayed Middle East on-premise deals created a roughly 75-basis-point headwind, Q2 demonstrates that timing can reverse quickly. The business is not decelerating as sharply as the pre-print tape feared, but quarterly beat quality must be normalized for deal timing.
Operating-quality reconciliation. The 98% renewal rate, 20%-plus cRPO and customer-count expansion indicate durable platform demand. AI ACV and uplift evidence address the key debate more directly than generic product bookings because they link deployment to contracted value. However, the federal pull-forward means Q2 revenue/cRPO quality is not as strong as the headline, and near-term gross-margin pressure is real. The company is using operating-expense leverage to protect consolidated margin while compute cost rises. That can work during strong growth, but it becomes less forgiving if cRPO slips below 20%.
What changes the FY2 case. If production AI customers continue to multiply, uplift remains above 20% and renewal holds near 98%, FY2 estimates can incorporate a larger Pro Plus/AI mix and operating leverage. If customers treat agents as usage experiments and expansion slows after initial bundles, gross-margin cost will arrive before durable ACV. The disclosed Level 1 outcomes are persuasive but concentrated; the next proof is broader vertical adoption, net retention and a quantified cost-to-serve curve.
FY1/FY2 bridge and debate. FY1 subscription revenue and margin estimates move modestly higher, not by the full beat, because half was timing. FY2 can rise if AI ACV converts at current uplift rates without gross-margin dilution. The bull claim that AI monetization is incremental is strengthened by ACV, uplift and production-use metrics. The bear claim that hyperscaler cost will erode margin remains unresolved; the 31.5% operating-margin guide suggests opex leverage offsets gross-margin pressure for now.
Decision: HOLD; add below $90 if cRPO stays at least 20% constant currency and operating margin at least 31%. Business delta: positive. Estimate delta: modestly positive. Stock delta: positive but timing-adjusted.
Q2 service revenue was $19.0 billion, +9%; postpaid service revenue $15.9 billion, +13%; Core adjusted EBITDA $9.5 billion, +12%; operating cash flow $7.5 billion, +7%; and adjusted free cash flow $4.8 billion, +4%. Diluted EPS was $2.99, +5%, including $0.14 of UScellular merger costs. Postpaid ARPA rose 2% to $152.91. Postpaid net account additions were 277,000, down 13% year over year, with postpaid account churn at 0.99%.
Versus Q1, account additions rose about 28%, ARPA 0.6%, service revenue 1.1%, postpaid service revenue 1.9%, EBITDA 3.3% and adjusted FCF 4.3%. The year-over-year rate of change is less strong: service growth slowed from 11% in Q1 to 9%, postpaid service from 15% to 13%, and account additions moved from +6% to -13%. The business is generating more dollars sequentially, but merger-assisted scale and slowing organic account growth complicate the quality read.
FY postpaid net-account guidance remained 950,000-1.05 million and Core adjusted EBITDA $37.1-$37.5 billion. Operating cash flow remains $28.1-$28.7 billion and capex approximately $10 billion. Adjusted FCF rose to $18.4-$18.8 billion from $18.1-$18.7 billion. FY1 cash estimates should rise by roughly the midpoint change; revenue/EPS estimates should not be moved aggressively until the call separates UScellular contribution, price and underlying churn. FY2 depends on whether ARPA growth can offset subscriber moderation without promotional intensity.
Quality and scenario bridge. ARPA plus churn is more important than gross additions alone. A 2% ARPA gain with 0.99% account churn suggests pricing remains acceptable, but the slowdown in account adds and service growth warns that merger contribution may be masking a softer organic rate. In a bull case, network advantage holds churn below 1%, ARPA grows 2%-3%, synergies lift EBITDA and annual FCF exceeds $18.8 billion. A base case holds the guide with account growth near the low end. A bear case sees promotions rise, churn exceed 1.05% and merger integration consume the FCF raise.
Capital allocation. Stockholder returns of $3.3 billion are support only if the underlying FCF is recurring. Q2 adjusted FCF grew more slowly than EBITDA, while cash capex rose 13%. The complete call must show whether working capital, merger costs or timing explains the difference. Until then, the FCF guide raise is a positive estimate delta but not proof of better structural conversion.
Debate: the bull case is network-led ARPA and merger synergy with visible cash returns. The bear case is that account growth is decelerating as integration costs and capex rise. The release supports cash durability but not a clean organic acceleration. Decision: HOLD. Call questions: organic postpaid growth excluding UScellular, merger-synergy cadence, price versus mix in ARPA, churn by legacy cohort, and the working-capital contribution to the FCF raise. Status PROVISIONAL: the complete 07:30 ET transcript/Q&A and prior-call comparison were not reviewed.
Revenue rose 10% to $11.99 billion versus $11.01 billion in Q1 and $10.85 billion a year ago. Adjusted EPS rose 13% to $6.03 versus a dated $5.71-$5.72 public consensus, a 5.4%-5.6% beat; GAAP EPS was $4.68. The stock was indicated up about 5.2% premarket. The rate of change improved materially from Q1 revenue growth of 6% and adjusted EPS growth of 6%.
The release-level result supports improving life-science tools and services demand, but the primary current release page was not indexed and ingested before cutoff, so segment organic growth, bookings, bioproduction demand, China exposure, PPD conversion, margin and the full-year guide bridge are not asserted. FY1 estimates should move higher only by the verified beat until the guide is reconciled. FY2 depends on organic growth rather than acquisition/currency contribution and on whether adjusted operating margin expands from Q1's approximately $2.40 billion of adjusted operating income on $11.01 billion of revenue.
Expectations and quality. The $6.03 adjusted EPS result exceeded the dated point by about $0.31-$0.32 and revenue accelerated roughly four points from Q1. That combination usually implies some mix of organic acceleration and margin leverage, but without the release bridge it is not safe to assign the improvement to bioproduction, analytical instruments, diagnostics or PPD. Currency and acquisition effects can lift reported growth without changing the core demand trajectory. The 5.2% premarket response indicates the market interpreted the result as a meaningful positive revision, which raises the burden for the call to confirm broad organic strength.
Scenario bridge. A bull case requires at least mid-single-digit organic growth, improving biopharma orders, stable China and margin expansion; that would support higher FY1 and FY2 EPS. A base case assumes the verified beat but waits for segment composition. A bear case would reveal reported growth dominated by acquisition/FX, weak instrument orders or a guide that embeds second-half deceleration. Falsification is explicit: organic growth below 4%, declining bioproduction orders or a lower full-year organic guide would reject the recovery interpretation.
The required decision discipline is to let the primary segment bridge—not the price reaction—determine whether this is a cyclical turn or a reported-growth artifact.
Debate: the bull case is a real tools-cycle recovery with broad end-market acceleration; the bear case is that currency/acquisition and easy comparisons are overstating organic demand. The current facts favor the bull direction but do not resolve composition. Decision: WATCH; no fresh action before call review. Required call questions: organic growth by segment, biopharma versus academic/government demand, China order trends, PPD book-to-bill, pricing, margin bridge and the exact FY guide raise. Status PROVISIONAL: the 08:30 ET call was after the scheduler cutoff and the complete current primary release was not available in the evidence bundle.
The official release reported $5.463 billion of revenue, +23% year over year, $1.98 billion of net income and $2.14 of EPS, including a $0.05 benefit not in prior guidance. Q3 guidance is $5.65-$6.15 billion of revenue and $2.23-$2.57 of EPS. The result confirms a broad analog-cycle rebound and strong data-center demand, but the earnings quality must be separated between utilization, inventory replenishment, end demand and the $0.05 benefit.
Pre-print and variance logic. The deterministic bundle did not contain a reliable current revenue consensus, so no revenue surprise is invented. The $0.05 item means normalized EPS was approximately $2.09, still a strong result but below the headline. The Q3 revenue midpoint of $5.90 billion implies about 8% sequential growth from Q2; the EPS midpoint of $2.40 implies roughly 12% sequential growth. That operating leverage is consistent with better factory utilization, yet utilization-led margin recovery is cyclical rather than proof of pricing power.
FY1 revenue/EPS estimates rise because the Q3 midpoint is $5.90 billion and quarterly revenue is accelerating. FY2 is more sensitive to automotive/industrial sell-through, pricing and depreciation from the capacity build. A restocking-led rebound supports one to two quarters; a true end-demand recovery supports FY2. The absence of complete Q&A prevents a confident split.
Debate and model tests. The bull case says analog inventories have normalized and broad industrial/automotive demand is entering a multi-quarter upcycle. The bear case says data center plus channel restocking are temporarily lifting utilization while core industrial demand remains uneven. Three indicators resolve it: customer inventory days, cancellations/lead times and sequential industrial/automotive revenue. Gross margin should also expand despite depreciation if utilization is the dominant driver. If revenue reaches the Q3 midpoint but channel inventory rises and free cash flow does not improve, the quality is lower than the EPS suggests.
Stock-versus-business delta. The business delta is positive because revenue growth and the Q3 guide show higher utilization. The estimate delta is positive for FY1 and only tentatively positive for FY2. The stock delta depends on whether investors already discount a full analog upcycle after the recent rerating; without a verified valuation stack, the brief does not turn a cyclical acceleration into an unconditional BUY. The decisive PM transcript task is management's language on industrial breadth versus customer restocking, compared line by line with Q1.
Decision: HOLD/WATCH. Upgrade only if industrial and automotive orders remain positive for two quarters, internal inventory does not reaccelerate and gross margin expands despite depreciation. Falsify: Q4 guide implies sequential revenue decline, cancellations rise, or channel inventory grows faster than sell-through. Status PROVISIONAL: complete Q2 Q&A with analyst anchors and a prior-quarter transcript comparison were not acquired by cutoff.
The official Q2 release reaffirmed the July 14 preannouncement. Revenue was approximately $17.2 billion and operating EPS $2.93, with quarterly free cash flow about $2.5 billion. The full-year constant-currency revenue framework moved to 4%-5% from more than 5%. The core negative is lower top-line conversion in a market that continues to reward platform and infrastructure spend; the offsets are productivity, margin and retained free-cash-flow guidance.
Expectations and operating quality. Because the guide reduction was preannounced, the relevant bar was whether Software, Red Hat and free cash flow deteriorated further. The release did not compound the warning, but it also did not restore the prior growth trajectory. Operating EPS and cash provide downside support; they do not erase the revenue-rate reset. Acquisitions make the mix harder to read because reported growth can improve while organic conversion and integration costs lag. The current release alone cannot separate recurring software demand from transaction contribution with enough precision for a final call.
FY1 revenue estimates should move to the 4%-5% framework. EPS moves down less because margin productivity offsets part of the revenue miss, while free cash flow remains the main valuation support. FY2 depends on converting Red Hat, HashiCorp and Confluent demand without further integration drag and on Consulting returning to growth. The quarter contrasts with ServiceNow: enterprises are spending on control layers and data/AI infrastructure, but transformational-services demand remains timing-sensitive.
Scenario and falsification. A bull case requires Red Hat above 12%, Software above 8%, Consulting back to positive growth and retained FCF, allowing mix and productivity to offset the lower consolidated rate. A base case accepts 4%-5% revenue and stable cash, producing earnings growth without re-rating. A bear case has Software below 5%, Consulting negative and another guide reduction, which would show the issue is demand/positioning rather than timing. The next complete Q&A must explain pipeline conversion, acquisition contribution, generative-AI bookings and the difference between bookings rhetoric and recognized revenue.
Stock-versus-business delta. The business delta is negative on growth and neutral-to-positive on productivity. The FY1 estimate delta is negative for revenue but less negative for EPS and cash. The stock delta is mixed because the July 14 preannouncement already reset expectations; relief after the release would represent confirmation that the warning did not worsen, not a return to the old thesis. A final rating requires the transcript to show whether the cut came from discrete deal timing or a broader loss of execution velocity.
Decision: WAIT. Upgrade only if Q3 constant-currency revenue is at least 5%, Red Hat above 12%, Software above 8% and quarterly FCF grows. Falsify: Software below 5%, Consulting negative or an FY FCF cut. Status PROVISIONAL: no verified complete current Q&A and prior-call wording comparison were acquired by cutoff.
Every qualifying current reporter is included below. “Evidence present” means the deterministic collector surfaced a current result record; it does not mean that the primary release, dated consensus, complete call and FY1/FY2 model gates were satisfied.
| Ticker | Current evidence | Decisive missing input / PM catch-up |
|---|---|---|
| [[TotalEnergies | TTE]] | result evidence present |
| [[Union Pacific | UNP]] | result evidence present |
| [[Blackstone | BX]] | result evidence present |
| [[Freeport-McMoRan | FCX]] | result evidence present |
| [[Comcast | CMCSA]] | result evidence present |
| [[Norfolk Southern | NSC]] | result evidence present |
| [[Honeywell | HON]] | result evidence present |
| [[Nokia | NOK]] | result evidence present |
| [[STMicroelectronics | STM]] | release: revenue $3.49B, gross margin 34.8%, EPS $0.24 |
| [[argenx | ARGX]] | result evidence present |
| [[Nasdaq | NDAQ]] | result evidence present |
| [[Ameriprise Financial | AMP]] | result evidence present |
| [[PG&E | PCG]] | result evidence present |
| [[Infosys | INFY]] | result evidence present |
| [[Huntington Bancshares | HBAN]] | result evidence present |
| [[Roper Technologies | ROP]] | result evidence present |
| [[Shinhan Financial | SHG]] | result evidence present |
| [[Dover | DOV]] | result evidence present |
| [[Teck Resources | TECK]] | result evidence present |
| [[West Pharmaceutical Services | WST]] | result evidence present |
| [[First Citizens BancShares | FCNCA]] | result evidence present |
| [[Quest Diagnostics | DGX]] | result evidence present |
| [[Dow | DOW]] | result evidence present |
| [[Snap-on | SNA]] | result evidence present |
| [[Cemex | CX]] | result evidence present |
| [[Tractor Supply | TSCO]] | result evidence present |
| [[Allegion | ALLE]] | collector runtime limit |
| [[Popular | BPOP]] | collector runtime limit |
| [[Ryder System | R]] | collector runtime limit |
| [[Old Republic | ORI]] | collector runtime limit |
| [[American Airlines | AAL]] | collector runtime limit |
| [[FirstCash | FCFS]] | collector runtime limit |
| [[Valley National Bancorp | VLY]] | collector runtime limit |
| [[Mobileye | MBLY]] | collector runtime limit |
| [[Albertsons | ACI]] | collector runtime limit |
| [[Pool Corp | POOL]] | collector runtime limit |
| [[FirstService | FSV]] | collector runtime limit |
| [[Cleveland-Cliffs | CLF]] | collector runtime limit |
| [[Lazard | LAZ]] | collector runtime limit |
| [[Vita Coco | COCO]] | collector runtime limit |
| [[Bread Financial | BFH]] | collector runtime limit |
| [[Harley-Davidson | HOG]] | collector runtime limit |
| [[Ardagh Metal Packaging | AMBP]] | collector runtime limit |
| [[Visteon | VC]] | collector runtime limit |
| [[IMAX]] | collector runtime limit | primary release, box office, installations, backlog and call |
The 46 lower-priority prior-evening AMC issuers remain explicit obligations: [[CSX]], [[KMI]], [[URI]], [[WCN]], [[CCI]], [[RJF]], [[LVS]], [[AVB]], [[EQR]], [[LUV]], [[ROL]], [[PKG]], [[RS]], [[MEDP]], [[PNFP]], [[GL]], [[RNR]], [[ELS]], [[KNX]], [[GGG]], [[EGP]], [[MOH]], [[SEIC]], [[FR]], [[FAF]], [[EPRT]], [[WH]], [[WEX]], [[RLI]], [[SON]], [[FULT]], [[TCBI]], [[OII]], [[CATY]], [[LBRT]], [[CVBF]], [[QS]], [[SLG]], [[VTMX]], [[FRME]], [[KALU]], [[EFSC]], [[BANR]], [[GTY]], [[STC]] and [[NTST]]. Their exact gate remains the same: current primary release/filing, dated consensus and range, decisive KPI bridge, complete call where available, and FY1/FY2 estimate translation. No new BUY/SELL verdict is assigned.
The older BMO queue—[[GE Vernova|GEV]], [[TE Connectivity|TEL]], [[Rogers Communications|RCI]], [[RPM International|RPM]], [[AT&T|T]], [[Philip Morris|PM]], [[PulteGroup|PHM]], [[Wabtec|WAB]], [[CME Group|CME]], plus [[Equinor|EQNR]], [[Moody's|MCO]], [[Northern Trust|NTRS]], [[Teledyne|TDY]], [[Otis|OTIS]], [[Stifel|SF]], [[Old National Bancorp|ONB]], [[Iridium|IRDM]], [[Travel + Leisure|TNL]], [[Badger Meter|BMI]], [[First BanCorp|FBP]], [[Cal-Maine Foods|CALM]], [[BankUnited|BKU]] and [[First Bancorp|FBNC]]—still lacks full Q&A forensics and prior-call language comparison. Full-session price action was reconciled in [[EarningsBrief/EarningsBrief_2026-07-22_PM]], but price alone is not analytical closure.
| Priority | Ticker/group | Exact required input | Deadline |
|---|---|---|---|
| 1 | RTX | complete 07:30 transcript/Q&A; capacity, Pratt remediation, FCF timing; prior-call delta | 2026-07-23 PM |
| 2 | LMT / TMO | complete 08:30 transcripts/Q&A; guide, throughput/segment, cash and prior-call deltas | 2026-07-23 PM |
| 3 | TMUS | complete call; organic account/ARPA, merger synergy, churn and FCF bridge | 2026-07-23 PM |
| 4 | TXN / IBM | complete Q2 Q&A and Q1 wording comparison | 2026-07-23 PM |
| 5 | CMCSA / UNP / TTE / BX / FCX / NSC / HON | primary release, consensus stack, decisive KPI/model bridge and call | 2026-07-23 PM |
| 6 | Remaining current BMO Tier 3 | exact row-level missing primary/call/model inputs above | next scheduled brief, market-cap/reaction order |
| 7 | Prior AMC and older BMO ledgers | complete primary expectations/call/model gates | rolling; retain until closed |
| Company | Tier | Status | Causal KPIs | Q&A exchanges | Prior-call source | FY1/FY2 bridge | Failed/deferred gate |
|---|---|---|---|---|---|---|---|
| GOOG/GOOGL | 1 | FINAL POST CALL | 7 | 3 graded | official Q1 transcript | complete direction/sensitivity | buy-side hurdle and numerical ROIC unavailable |
| TSLA | 1 | FINAL POST CALL | 8 | 5 graded | complete Q1 transcript source | complete direction/sensitivity | robotaxi unit economics undisclosed |
| RTX | 1 | PROVISIONAL | 9 | 0 | Q1 release | complete direction/sensitivity | complete current transcript/Q&A |
| LMT | 1 | PROVISIONAL | 8 | 0 | Q1 release | complete direction/sensitivity | 08:30 call after cutoff |
| NOW | 2 | FINAL POST CALL | 8 | 4 graded | Q1 release/event | complete direction/sensitivity | independent buy-side hurdle unavailable |
| TMUS | 2 | PROVISIONAL | 8 | 0 | Q1 release | complete direction | complete current transcript/Q&A |
| TMO | 2 | PROVISIONAL | 4 | 0 | Q1 release | preliminary only | complete primary release and 08:30 call |
| TXN | 2 | PROVISIONAL | 4 | 0 | Q1 release | complete direction | complete current Q&A |
| IBM | 2 | PROVISIONAL | 5 | 0 | Q1 release | complete direction | complete current Q&A |
| 45 additional current BMO rows | 3 | LEDGER | discovery only | 0 | none | deferred | exact row-level inputs listed |
| 46 prior AMC + older BMO queue | 3 | ROLLED | prior report | 0 | incomplete | deferred | primary/call/model gates |
Evidence collection limited to top market-cap reporters for runtime./Users/max/Documents/TIF/Daily/2026-07-23.md already existed before backlink work. No skeleton was created and no existing content was overwritten./Users/max/morningsignal-research/state/earnings/earnings_context_2026-07-23_AM.json, generated 2026-07-23 08:03 ET; 49 qualifying BMO companies, 30 broad current-result flags. Discovery flags are not final evidence.(source: EarningsBrief-AM, 2026-07-23) #sellside