Amazon Trades $232.16 on 12.56M Shares Ahead of July 30 Earnings — AWS Must Clear 33% and Free Cash Flow Must Turn From $1.2B for $247.20 to Break

Amazon (AMZN) traded between $231.53 and $235.99 on Monday and finished the morning at $232.16, sitting 0.3% off the session low and 1.6% below the session high. Volume ran at 12.56 million shares against a 37.56 million daily average — light participation into the biggest earnings print of the company's year, three sessions away.

The intraday shape is the same one every risk asset traced Monday. The S&P 500 opened up 0.85% on futures and closed the morning flat at 7,411. Gold tagged $4,106 in Asia and sat back down near $4,075. Bitcoin touched $65,359 and sank toward $64,580. EUR/USD ran to 1.1420 and slipped under 1.14. Amazon opened near $236 and closed the morning at the bottom of its range. A collapse in Brent crude lifted everything at the open, and nothing held it.

The damage on the longer chart is what matters here. Amazon's 52-week high is $278.56, set in May 2026. At $232.16 the stock trades 16.7% below that peak — well past the 10% threshold that defines a correction and closing in on 20%. The 52-week low is $196.00. Market capitalisation stands at $2.50 trillion with an enterprise value of $2.56 trillion.

Last week did most of the damage. Amazon fell 7.15% across five sessions, including a 4.6% drop on Thursday during which the stock traded down to $234.81 intraday while the Nasdaq 100 fell nearly 2%. That Thursday decline had nothing to do with Amazon's own numbers — the company reported nothing. It was collateral damage from Alphabet lifting its 2026 capital expenditure guidance, compounded by a fresh Senate inquiry into Amazon's marketplace and news of additional layoffs.

The setup into Thursday's print is therefore unusually loaded. A company with accelerating cloud growth, record segment margins and a $70 billion advertising business trades at a 17% discount to its own high because the market has stopped rewarding the spending required to produce those numbers. Trailing twelve-month revenue is $742.78 billion. Trailing net income is $90.8 billion. Diluted trailing EPS is $8.36.

Three sessions from now, one report resolves which of those two readings the market has wrong.

Consensus for Thursday: $1.82 a Share on Roughly $199 Billion

Amazon reports second-quarter results after the close on Thursday, July 30. Consensus earnings sit at $1.82 per share, up from $1.68 in the year-ago quarter — modest growth for a company whose revenue is compounding in the mid-to-high teens, which is itself a statement about where the operating leverage is going.

Management guided second-quarter net sales to a range of $194 billion to $199 billion, representing 16% to 19% growth, against a pre-guidance consensus near $188.9 billion. Operating income was guided to $20 billion to $24 billion. Sell-side modelling has clustered toward the upper half of both bands, with one large US bank projecting $198.8 billion in revenue and $24.1 billion in operating profit.

The segment detail is where the report gets decided. North America total revenue consensus stands at $113.8 billion, a figure that has edged upward since the February release on resilience in the online retail business. AWS consensus has held steady at approximately $40.5 billion. Expectations for both North American and international online retail have remained stable quarter over quarter, which means the surprise capacity sits almost entirely in cloud and in margins rather than in top-line retail.

Prediction markets currently price a 95% probability that Amazon beats second-quarter estimates. That number should be read carefully rather than as reassurance. Amazon beats the headline number with high regularity — the question every quarter for the last four has been whether the beat matters, and for the last two the answer has been no. Alphabet delivered record revenue, 82% cloud growth and $9.11 in earnings per share last week, and its stock fell from $374 to $319.

The setup for Thursday is therefore not about whether the number clears. It is about which line the market chooses to trade. Four candidates: the AWS growth rate, the AWS operating margin, the third-quarter revenue guide, and the full-year capital expenditure figure. Each can surprise independently, and they do not all point the same way. A quarter that shows AWS accelerating while capex rises to fund it is simultaneously the bull case and the bear case depending on which number the desk anchors on.

That ambiguity is precisely why the stock has bled 7% into the print rather than rallying on a 95% beat probability.

The Three Margin Lines That Will Decide Thursday's Reaction

Beneath the headline numbers sit three margin consensus figures that carry more information than either revenue or EPS, and the width of the estimate ranges around them tells you how little the sell side agrees.

AWS operating margin consensus for the second quarter stands at 33.8%. That is down from the 37.7% analysts expected as recently as last October, and up from the 32.9% modelled in February 2026. The range of estimates has narrowed by 200 basis points to a 30.9% to 38.2% band. That 730-basis-point spread on the single most profitable segment in the company reflects genuine disagreement about whether AI infrastructure depreciation is compressing cloud margins faster than revenue growth can offset it.

North American retail operating margin consensus sits at 7.5%, with estimates ranging from 3.1% to 8.9% — a 580-basis-point spread. That figure represents a decline of roughly 100 basis points since February, and it is worth remembering that this segment was posting outright losses a few years ago. The recovery to mid-single-digit margins on a $113.8 billion quarterly revenue base is one of the more underappreciated achievements in the business, and the market currently assigns it no premium.

International operating margin consensus is 4.2% against a range of 1.7% to 13.4%. An 1,170-basis-point spread on a segment estimate is not a forecast; it is an admission that nobody has visibility.

The reason these matter more than EPS is arithmetic. Amazon's earnings are dominated by AWS profit dollars, and AWS profit dollars are the product of revenue growth and margin. If AWS grows 28% and holds a 33.8% margin, the print supports every bullish thesis on the stock. If AWS grows 28% and the margin comes in at 31%, the incremental return on a $200 billion capital programme starts looking materially worse, and the free cash flow trajectory gets pushed further out.

Management flagged higher transportation costs tied to fuel inflation on the last call, partially offset by a logistics surcharge. With Brent having run above $100 during the quarter before Monday's collapse, that headwind was live through most of the reporting period. Retail margin is where it shows up.

Q1 Was the Best Quarter in Years and the Stock Fell Anyway

The clearest guide to Thursday is what happened in April, because the pattern has already run once.

Amazon reported first-quarter revenue of $181.5 billion on April 29, up 17% year over year and comfortably above the $177.2 billion consensus and above its own guidance. Operating income rose 30% to $23.9 billion for a 13.1% margin. AWS grew 28% to $37.59 billion — its fastest pace in fifteen quarters, against the 26% analysts expected — on a $150 billion annualised revenue run rate, and the segment's operating margin hit a record. Earnings came in at $2.78 per diluted share against a $1.62 estimate, with net income of $30.3 billion.

That headline EPS carried a large asterisk: $16.8 billion of it was a non-cash gain from revaluing Amazon's stake in Anthropic. Stripping that out leaves operating performance that was still excellent but nothing like the reported figure implies. Year-ago comparison was $17.1 billion in net income and $1.56 per share.

The guidance was also strong. Second-quarter net sales of $194 billion to $199 billion against an $188.9 billion consensus, with operating income of $20 billion to $24 billion.

The stock fell roughly 3% after hours.

The reason was capital expenditure. Cash capex hit $43.2 billion in the quarter, primarily for AWS and generative AI, against a full-year commitment of approximately $200 billion. Trailing twelve-month free cash flow had compressed to $1.2 billion, a 95% decline from $25.9 billion in the same period a year earlier. The market read a record quarter and priced the cash flow instead.

On that call, the chief executive made the argument he will make again Thursday: the company is not gambling on AI but responding to demand, has high confidence the capital will be monetised because customer commitments already cover a substantial portion of it, and expects compelling operating margins and returns on invested capital. He added that AWS must invest ahead of demand because cash outlays for land, power, buildings and hardware occur six to twenty-four months before customers are billed.

That argument was not enough in April. It is being tested again from a base 17% lower.

The Capex Number Is Going Up, and Everyone Knows It

The single most consequential line item Thursday is full-year 2026 capital expenditure, currently guided at approximately $200 billion. The consensus expectation on the street is that it rises.

The trajectory is extraordinary. Consensus estimates have expanded roughly fourfold from $52.7 billion in fiscal 2023 to above $200 billion for fiscal 2026. Amazon spent roughly $131 billion in 2025 against an original $100 billion forecast, and the $200 billion figure disclosed with fourth-quarter results represented a 56% year-over-year increase against analyst estimates of $146.6 billion. First-quarter spending ran $43.2 billion to $44.2 billion, annualising above $175 billion before any acceleration.

The case for a raise is circumstantial but strong. Alphabet lifted its 2026 guidance from $180 billion to $190 billion up to $195 billion to $205 billion last week, explicitly citing acceleration in capacity delivery to meet demand — which puts it in line with or above Amazon's current plan. Alphabet had also raised in the first quarter while Amazon held steady, meaning Amazon is now the one with the stale number. Meta raised its own range to $125 billion to $145 billion from $115 billion to $135 billion. Tesla told investors its capital spending will exceed $25 billion this year. Combined 2026 AI infrastructure spending across Amazon, Microsoft, Alphabet and Meta now runs near $725 billion, up from roughly $410 billion to $440 billion in 2025 — a 70%-plus increase in twelve months.

Management has all but pre-announced it. On the first-quarter call the chief executive said the company would not hesitate to increase near-term capex if AWS continues demonstrating rapid growth. AWS grew 28%. The conditional has been satisfied.

The physical footprint backs the number. Amazon operates more than 900 data centre facilities according to a leaked internal count, with recently announced projects including a $3 billion campus in Vicksburg, Mississippi. Power access, advanced packaging and grid interconnection timelines are now the binding constraints across all four hyperscalers rather than capital availability.

The uncomfortable part for shareholders is what a raise signals. Hyperscaler capex commitments have become ratchet-up only — lowering guidance would require a structural break in AI demand. That makes the spending a fixed cost the equity absorbs regardless of what the return profile turns out to be, and it removes optionality from the story.

Free Cash Flow Fell 95% and That Is the Whole Bear Case

Amazon's trailing twelve-month free cash flow stands at $1.2 billion, down from $25.9 billion in the same period a year earlier — a 95% collapse. For a company generating $742.78 billion in trailing revenue and $90.8 billion in trailing net income, that number is arresting.

Analyst modelling expects the situation to deteriorate before it improves. Projections for fiscal 2026 point to negative free cash flow of roughly $8 billion to $11 billion as the $200 billion capital programme runs against operating cash generation. A megacap technology company with a 12.22% net margin and a 24.29% return on equity is on track to burn cash for a full year.

The bull framing is that this is a timing artefact rather than an economic problem. Capital expenditure is recognised when cash leaves; revenue from that capacity arrives six to twenty-four months later. High capex growth outpacing revenue growth mechanically compresses free cash flow during a build phase, and management has been explicit that the current situation mirrors the early AWS investment cycle — heavy upfront spending that produced one of the most profitable businesses in corporate history.

The bear framing is that the analogy is unproven and the scale is different. The early AWS build was measured in single-digit billions against a market with no comparable competitor. This build is $200 billion annually against three rivals spending similar sums into the same power-constrained markets with the same silicon supply chain. Depreciation from this capital starts hitting the income statement in earnest through 2027 and 2028 regardless of whether the revenue shows up on schedule.

There is also a balance-sheet dimension. A heavier debt load leaves less room for error if AI demand cools, and free cash flow is the metric that services it. Enterprise value stands at $2.56 trillion against a $2.50 trillion market capitalisation, implying roughly $60 billion of net debt — modest today, but the direction of travel is one way while free cash flow is negative.

What makes Thursday genuinely binary is that both readings are consistent with the same data. A quarter showing AWS at 28%-plus growth with a stable 33.8% margin supports the timing-artefact case. The same quarter with AWS margin at 31% supports the returns-are-degrading case. The report contains the answer; the question is which line the market reads first.

AWS Grew 28% and Still Looks Slow Next to Its Two Rivals

The competitive comparison is where Amazon's story gets uncomfortable, and it is the reason the stock does not get credit for a fifteen-quarter growth high.

In the first quarter, AWS grew 28% to $37.59 billion on a $150 billion annualised base. Google Cloud grew 63% with a backlog above $460 billion, nearly double the prior quarter. Microsoft Azure grew 40%. On a pure growth-rate comparison, AWS finished third among the three major hyperscalers despite posting its own best result in nearly four years.

The scale argument is real and frequently understated. AWS is growing 28% on a $150 billion base, which adds roughly $42 billion of annualised revenue. Google Cloud's 63% comes off a substantially smaller denominator. Adding forty-plus billion dollars to a business that size is a different achievement from adding the same percentage to a business a third as large, and the absolute dollar contribution is what flows to operating income.

The more useful framing is that the cloud market has stopped converging on a single growth rate and started fragmenting by workload type, silicon stack and integration depth. Google Cloud's 63% reflects AI-native workload capture. Azure's 40% reflects enterprise AI adoption riding Microsoft's distribution into existing seats. AWS's 28% reflects scale-driven execution at industrial cadence across the broadest installed base. Those are three different businesses being compared on one metric.

Amazon's own forward-looking disclosures support the case. AWS backlog stood at $244 billion at the fourth quarter, up 40% year over year. AWS AI revenue is running above a $15 billion annualised rate. Bedrock token processing volumes have grown sharply. Sell-side AWS growth forecasts for 2026 now cluster between 31% and 33%, with one house modelling nearly 35% for 2027, citing stronger customer demand and recent price increases.

If AWS prints above 28% Thursday — accelerating for a second consecutive quarter — the narrative that Amazon is losing the cloud race becomes difficult to sustain. That is the single most powerful thing this report can deliver.

If it prints at 26% or 27%, the deceleration lands on a stock already carrying a $200 billion capital commitment and negative free cash flow, and the reaction will be severe.

Custom Silicon at a $20 Billion Run Rate Is the Underpriced Asset

The part of Amazon's AI position that receives the least attention in the capex debate is the one that most directly addresses it. The company's in-house chip business — Trainium, Graviton and Nitro — has scaled to a $20 billion annual revenue run rate and is growing at triple-digit rates year over year.

That matters for two reasons. First, it converts a portion of the $200 billion capital programme from a payment to a third-party supplier into internal cost, which changes the return math on every dollar of AWS capacity. Second, it establishes custom silicon as a standalone strategic asset rather than merely a cost lever — a distinction the market has been slow to price. Management has framed the combination of broad functionality, proximity of inference to data, strong security and a unique collection of proprietary chips as the structural basis for AWS growth.

The external validation arrived through customer commitments. Amazon holds a large stake in Anthropic, revalued upward by $16.8 billion in the first quarter, and has struck Trainium arrangements that extend beyond its own inference workloads. Separate reporting has pointed to additional frontier-lab silicon relationships. Those deals matter more as proof of competitiveness than as revenue: a leading model developer choosing Trainium is a statement about the chip stack that no internal benchmark can replicate.

The Anthropic stake also introduces a reporting complication worth flagging ahead of Thursday. Revaluation gains flow through net income and can distort headline EPS materially — as they did in the first quarter, when $16.8 billion of a $30.3 billion net income figure came from that line. Any Thursday EPS number should be checked against operating income before it is compared to the $1.82 consensus.

Beyond silicon, capital is flowing into adjacencies. The Globalstar merger brings scarce global spectrum required for direct-to-device connectivity and satellite expertise, with a strategic side benefit: Apple, which holds a 20% stake in Globalstar, will use Amazon's satellite connectivity for some products. That deal extends AWS connectivity beyond terrestrial fibre into underserved markets and builds a commercial relationship with a company Amazon has historically had limited overlap with.

None of this is in the $1.82 consensus. All of it is in the $200 billion.

Advertising Crossed $70 Billion and Retail Margins Held

The businesses that actually fund the AI build get discussed least, and they are performing.

Amazon's advertising operation has crossed $70 billion in trailing revenue and continues to compound. That is a business roughly the size of a standalone S&P 100 constituent, carrying software-like incremental margins, embedded inside a company the market values on retail multiples. It benefits directly from the same AI investment being criticised — targeting, creative generation and campaign automation all improve with model capability, and the company has been rolling out infrastructure to let AI agents execute ad campaigns programmatically.

Retail is holding up better than the consumer-spending narrative implies. Prime Day sales rose 9.5% year over year, which one large US bank cited explicitly when raising its price target, arguing that consumer demand remains healthy despite the macro anxiety weighing on the stock. North American retail operating margin has climbed from outright losses a few years ago to a consensus 7.5% for the second quarter.

Two headwinds sit against that. Transportation costs are rising on fuel inflation, partially offset by a logistics surcharge — a live issue given Brent traded above $100 during the quarter. And the company recently introduced a policy targeting AI-generated people in product advertisements, a governance move that reflects the marketplace-integrity pressure now arriving from Washington.

The regulatory overlay is the newest risk. A Senate inquiry into Amazon's marketplace opened last week and contributed to Thursday's decline. Separately, tensions between Washington and Brussels over European fines on US technology companies escalated, with the administration warning of consequences — a dispute in which Amazon is a direct target. Fresh layoffs were also disclosed, which markets typically read as cost discipline but which arrive awkwardly alongside a $200 billion spending plan.

The composite picture is a company whose non-cloud earnings power is stronger than its share price suggests. Advertising above $70 billion, retail margins at 7.5% and international turning positive would, in isolation, support a materially higher multiple. They are being obscured entirely by a capital expenditure line that has quadrupled in three years.

Alphabet Reported a Perfect Quarter Last Week and Lost 15%

The most important precedent for Thursday landed on July 22 and requires no interpretation.

Alphabet reported record revenue of $119.8 billion, cloud growth of 82%, earnings of $9.11 per share and advertising growth above 14% driven by model integration. On any historical standard, that is an unimpeachable print from the company widely considered the best-positioned AI franchise among the megacaps — consumer model distribution, homegrown data-centre silicon, an accelerating cloud business.

The stock fell 7% Thursday and has traded from $374 down to $319. The problems were $44.9 billion of quarterly capital expenditure, full-year guidance lifted to $195 billion to $205 billion, and a credit-rating agency warning that followed. Alphabet did everything right operationally and was punished for the cost of doing it.

The contagion was immediate. Meta fell 4% on the same session. Amazon fell 4.6% without reporting anything. The index tracking the seven largest technology companies dropped 4.8% Thursday, its worst single session since the April 2025 tariff announcement. Tesla, which reported the same week, fell 14%.

What this establishes for Thursday is that the market's reaction function has inverted. Strong cloud growth is no longer read as evidence that the capital is working; it is read as the reason the capital number will go higher. Alphabet's 82% cloud growth did not protect it — arguably it caused the selloff, because acceleration is precisely what forces a capex raise.

Amazon walks into that reaction function with the largest capital programme of the four and the weakest free cash flow. It also walks in 17% below its high, having already absorbed a 7.15% weekly decline, which is not true of the companies that reported before it. Alphabet fell from a position of strength. Amazon would be falling from one of considerable weakness.

The counter-scenario deserves equal weight. Positioning into Thursday is light — Monday's 12.56 million shares against a 37.56 million average says nobody is adding. Discretionary investor positioning across the market has fallen to the seventeenth percentile, back to April lows. A stock that has pre-traded the bad news can rally on a print that merely fails to worsen it.

At 27x Earnings and 13.6x EBITDA, the Multiple Has Already Compressed

Amazon's valuation has done a great deal of work in the last three months, and the numbers no longer describe an expensive stock by megacap standards.

Trailing price-to-earnings sits at 27.76 against diluted trailing EPS of $8.36. Forward P/E is 27.40 — essentially no multiple expansion priced into next year's earnings. The PEG ratio is 1.25. Price-to-sales is 3.39 on $742.78 billion of trailing revenue. Price-to-book is 5.65. Enterprise value to revenue is 3.45 and enterprise value to EBITDA is 13.60. Profit margin runs 12.22%, return on assets 6.84% and return on equity 24.29%.

The trailing earnings figure carries the Anthropic revaluation distortion, so the 27.76 multiple flatters the stock somewhat. The EBITDA multiple is cleaner, and at 13.60 times trailing — roughly 12 times forward on sell-side modelling — Amazon trades below a wide range of retail and technology comparables. One house has built its entire bullish case on that observation, arguing the market is pricing Amazon like a mature retailer while assigning minimal value to AWS and the AI franchise.

The analyst distribution reflects genuine disagreement about which of those two readings is right. The consensus target sits between $309.55 and $313.02 depending on the panel, implying roughly 33% to 35% upside from $232.16. The high target is $370. The low is $207 — below the current price, and only 6% above the 52-week low. Ratings cluster at Strong Buy to Moderate Buy.

Recent individual moves have all been upward revisions or reiterations into weakness. One firm reiterated Buy at $335, raised from $325, citing healthy Prime Day demand and AWS growth forecasts near 33% for 2026 and 35% for 2027. Another reiterated Buy at $320 on the EBITDA argument. A third maintained Buy at $310, modelling $198.8 billion of second-quarter revenue. A fourth reiterated Outperform at $293 on July 23.

That every major house has held its rating while the stock fell 17% from its high is either conviction or anchoring. Thursday determines which.

The Technical Setup: Below Every Major Average With a ±7.2% Move Priced

Amazon trades beneath every meaningful moving average, and has for weeks.

The 20-day exponential moving average sits at $244.98 and the 50-day at $245.85 — both roughly 5.5% above spot and both sloping down. A separate 50-day simple calculation puts resistance at $251.16. The 200-day EMA at $236.48 was the last dynamic support in reach and has already been breached; it now functions as the first meaningful overhead level, sitting just above Monday's $235.99 high. That rejection is not coincidental.

Momentum readings show a stock stretched rather than stabilising. Daily RSI printed 38.73 last week, signalling persistent selling pressure without reaching classic oversold territory, while hourly RSI sank to 26.59 — deeply oversold on the short timeframe. MACD carries a sell signal on the daily with momentum accelerating downward. A pivot-top sell signal fired on July 15, and the stock has fallen roughly 9% since.

Support is layered close together. Monday's low at $231.53 is the immediate line. The daily Bollinger lower band sits at $230.75, and a break below it would mark a structural rather than a momentum event. Accumulated-volume support sits at $227.01, with the one-month range floor at $225.55. Below that, the chart is open toward the $215 area.

Resistance runs $235.99, then the 200-day EMA at $236.48, then $242.46 and $243.61, then the 20- and 50-day averages at $244.98 and $245.85, then the one-month high at $251.03. A significant longer-term resistance zone spans $272.69 to $275.00.

The options market has the earnings move priced at approximately ±7.2%, against a ten-year average of ±7.4% and an identical ±7.2% median. From $232.16 that implies a post-print range of roughly $215.40 to $248.90. The 95th-percentile historical outcome is ±13.7%, which would put the tails at $200.36 and $263.96. Four in five reports have stayed under ±10.1%, with a standard deviation near ±3.9%.

One statistic favours option buyers: across the last fifteen reports, Amazon's actual move exceeded its implied move roughly 60% of the time. The April 2026 reaction was +4.1%.

Forecast: Range-Bound Into Thursday, Then a Binary on the Capex Line

Into Thursday's close, the base case is continued compression between $230.75 and $236.48 on light volume. Assign roughly 55% weight. Positioning is thin, the stock has already absorbed a 7.15% weekly decline, and nobody adds size into a ±7.2% event with a Fed decision in between. Wednesday's FOMC statement and the Microsoft and Meta reports that evening will move Amazon in sympathy before Amazon reports anything of its own — a hawkish Fed or a poorly received Azure number pushes AMZN toward $227.01 ahead of its own print.

Post-earnings, the bullish path requires a specific combination: AWS growth at or above 28% with the operating margin holding near or above 33.8%, retail margin at 7.5% or better, third-quarter guidance above consensus, and — critically — a capital expenditure number the company can frame as demand-driven rather than defensive. That reclaims $236.48 and $244.98 quickly and opens $251.03 with the implied-move ceiling near $248.90. Assign 25%, targeting $250 — roughly 7.7% above spot, right at the implied move.

The bearish path is AWS decelerating below 28%, or the AWS margin printing near the 30.9% low end, or a capex raise toward $220 billion-plus without accompanying backlog disclosure to justify it. Any of the three breaks $230.75, and the 200-day EMA rejection at $236.48 already confirms sellers are defending it. That opens $227.01, then $225.55, then the $215 area at the lower implied bound. Assign 20%, targeting $218 — 6% lower. The Alphabet precedent is the roadmap: a strong quarter is not protection.

The trigger checklist for Thursday, in order: AWS revenue growth rate versus the 28% prior and roughly $40.5 billion consensus; AWS operating margin against 33.8%; full-year capex guidance against $200 billion; third-quarter revenue guidance; and free cash flow against the $1.2 billion trailing figure. Check any reported EPS against operating income first — Anthropic revaluation gains distorted the last print by $16.8 billion.

Calendar: FOMC Wednesday at 2 p.m. Eastern, Microsoft and Meta after Wednesday's close, Amazon and Apple after Thursday's close, with US second-quarter GDP and June PCE at 8:30 a.m. Thursday.

That's TradingNEWS