Microsoft Stock (NASDAQ: MSFT) Forecast: Shares at $397 After 28% Drawdown From $555 — $625 Billion Backlog, $81.3 Billion Quarter, and 47% Operating Margins Make This the Most Mispriced Mega-Cap on the Planet

Microsoft Corporation (NASDAQ: MSFT) is trading near $397 and the stock looks like it has been abandoned by momentum. Down 28.5% from the July 2025 all-time high of $555.45, down 16% since Q2 earnings that beat on both revenue and EPS, and trading at 22-24 times forward earnings while generating $175.26 billion in trailing EBITDA — MSFT is now the cheapest large-cap AI infrastructure play among the Mag 7 on a forward earnings basis. The market is pricing Microsoft as though its $625 billion commercial remaining performance obligations are fiction, its $250 billion OpenAI commitment is worthless, and its 47% operating margins are about to collapse under the weight of $145 billion in annual capital expenditure. Every one of those assumptions is wrong.

The Q2 fiscal 2026 report delivered $81.3 billion in revenue — $1 billion above consensus — with earnings per share of $4.14 against a $3.91 estimate. Azure grew 39%, in-line to 1% above expectations. Full year EPS guidance rose to $13.08 from $13.00. These are not the numbers of a company in distress. They are the numbers of a company being punished for investing aggressively in infrastructure that its customers are desperate to consume. Demand exceeds supply across Azure and AI services. The backlog is exploding. The copilot seat count is up 160% year-over-year. And the stock has fallen every week for the past month.

The $37 Billion Quarter That Spooked Wall Street — Why MSFT's CapEx Is an Asset, Not a Liability

Microsoft spent $37 billion on capital expenditure last quarter. The Street expected $33 billion. That $4 billion gap — less than 5% of the company's Q2 revenue — was the primary catalyst for a 16% decline in the stock. On a run-rate basis, MSFT is on pace to spend approximately $145 billion in capex for fiscal year 2026, well above its own guidance of $110 billion. Roughly two-thirds of the spend is directed toward short-lived assets — Nvidia GPUs, Microsoft's own Maia accelerators, and networking hardware. The remainder goes to data center construction and modernization.

The reaction borders on irrational when measured against the profitability expansion that this capital deployment has generated. Since fiscal 2021, Microsoft has allocated $264.74 billion toward capex — a 302.94% increase on an annualized basis. Over that same window, EBITDA has grown by 116.86%, adding $94.44 billion in absolute terms. Trailing twelve-month EBITDA stands at $175.26 billion. Cash from operations has expanded by 109.16%, an $83.77 billion increase. The correlation between capex and profitability at Microsoft is as direct as it gets: every dollar poured into AI infrastructure has generated more than a dollar in incremental operating cash flow within a reasonable timeframe.

Compare this with Apple. Since fiscal 2021, Apple has spent $67.06 billion on capex — increasing its annual spend by just 9.59% ($1.06 billion). AAPL's EBITDA growth over that period: 27.17%, or $32.67 billion. Microsoft's EBITDA has gone from 67.22% of Apple's in 2021 to 114.62% today. MSFT now generates $175.26 billion in EBITDA versus Apple's $150.90 billion — despite Apple producing $131.16 billion more in annual revenue. The company that reinvested has overtaken the company that bought back stock. And the market is rewarding the buyback company with a 33 times earnings multiple while punishing the reinvestment company at 23-24 times.

$81.3 Billion in Revenue, 47% Operating Margins, and $35.8 Billion in Operating Cash Flow — The Q2 Numbers Everyone Forgot

The quarter that triggered a 16% decline was, by any objective measure, a beat-and-raise. Revenue of $81.3 billion represented 17% year-over-year growth — extraordinary for a company generating revenue at this scale. Operating income increased 21% to $31.7 billion. Gross margins held at 68-69%. Operating margins came in at 47%, up from 45.5% in the year-earlier quarter. The business absorbed a 65% year-over-year increase in capex and still expanded its operating margin by 150 basis points. That's not a company struggling with returns on investment — that's a company whose revenue growth is outrunning even massive infrastructure spending.

Cash from operations hit $35.8 billion in the quarter, a 58.2% increase. Stripping out working capital movements, adjusted operating cash flow grew 34.1% to $45.39 billion. EBITDA jumped 23.5% to $47.38 billion. Net income surged 59.5% to $38.46 billion, though a significant portion of that spike came from a one-time accounting revaluation tied to the OpenAI recapitalization. Adjusting for that, non-GAAP pretax margins were approximately 49.7% and normalized EPS roughly $4.33 rather than the GAAP-reported $5.16.

For the first half of fiscal 2026, operating cash flow margins expanded from 42% to 51% year-over-year. Free cash flow margin was 23.3% for the first six months, up from 18.4% in the prior year period. Even with property, plant, and equipment expenditures doubling on the cash flow statement, free cash flow remained positive. The company projects that sequential capex will ease in the fiscal Q3 and Q4, which means the free cash flow margin inflection point is arriving — not in some speculative future quarter, but within the next six months.

Microsoft Stock (NASDAQ: MSFT) — The Three Segments Driving $305+ Billion in Annual Revenue

Intelligent Cloud: $32.9 Billion, Up 29%, and Still Capacity-Constrained

The Intelligent Cloud segment — encompassing Azure, SQL Server, and related server products — generated $32.91 billion in Q2 revenue, a 28.8% increase over the $25.54 billion from the year-ago quarter. Azure and other cloud services grew 39% year-over-year, and CEO Satya Nadella made a critical point on the conference call: if Microsoft had allocated all GPUs brought online in Q1 and Q2 solely to Azure rather than distributing capacity across first-party copilots and R&D, the Azure KPI would have been above 40%. The 39% headline number understates actual demand growth because Microsoft is deliberately routing capacity toward higher-lifetime-value products.

Segment operating margins were 42%, flat year-over-year, while gross margins declined from 63% to roughly 59%. That margin compression is entirely a timing phenomenon — new data centers begin generating depreciation and operating expenses immediately upon opening but take weeks or months to fill with paying customers. As capacity ramps to utilization, gross margins recover. The company has been on the receiving end of this dynamic for multiple consecutive quarters now, and yet operating margins for the overall business continue to expand. The revenue growth is simply overwhelming the margin dilution from new data center deployments.

Goldman Sachs projected in September 2024 that global cloud computing revenue would reach $2 trillion by 2030, a 22% annualized growth rate from the $496 billion base in 2023. As one of the two dominant platforms — alongside Amazon's AWS — Microsoft is positioned to capture a disproportionate share of that expansion. The backlog data confirms this: Azure has been capacity-constrained for multiple consecutive quarters. Current guidance calls for 37%-38% Azure growth in fiscal Q3, slightly above prior consensus. The 37-38% figure is a capacity allocation guide, not a demand guide. Actual demand is higher.

Productivity and Business Processes: $34.1 Billion, Up 16%, With a Pricing Catalyst Coming in July

The Productivity and Business Processes segment — Microsoft 365, LinkedIn, Dynamics 365, and Power BI — produced $34.12 billion in Q2 revenue, up 15.9% from $29.44 billion. This is now the largest segment by revenue at roughly 42% of total, and its operating margins of 60% versus 57.5% a year ago make it the most profitable. The margin expansion reflects the leverage inherent in software subscription businesses: incremental revenue from E5 tier upgrades and copilot seat additions drops almost entirely to the operating line.

Microsoft 365 commercial cloud revenue grew 17% while seat count expanded only 6% — the gap is a direct function of customers migrating to the E5 tier, the highest-priced all-in subscription that bundles advanced security, compliance, and analytics tools. Revenue per user is rising significantly faster than user count. And starting July 1, 2026, Microsoft is implementing a mid-to-high single digit price increase across multiple 365 SKUs: Business Basic rises 16.7% from $6 to $7 per month, Office 365 E3 climbs from $23 to $26 per month, and the frontline worker offering jumps 33%. Even government clients face increases. Given the hundreds of millions of 365 seats worldwide, these increases will generate billions in incremental annual revenue with zero associated capex.

On the consumer side, Microsoft 365 revenue jumped 27% as both subscriber count and revenue per user increased. LinkedIn posted 11% revenue growth driven by marketing solutions. Dynamics 365 grew 19% and continues to gain market share in the enterprise resource planning space. At roughly 3% of total revenue, Dynamics is small in absolute terms but significant as proof that Microsoft's enterprise SaaS franchise is accelerating, not deteriorating. The "SaaSpocalypse" narrative that has infected software valuations simply does not apply to a product that already embeds intelligent agents and copilots across sales, customer service, marketing, finance, and supply chain management.

More Personal Computing: $14.25 Billion, Down 3%, With Xbox Hardware Falling 32%

The weak spot. Revenue in More Personal Computing — Windows, Xbox, search and advertising — declined from $14.65 billion to $14.25 billion. Xbox hardware cratered 32% as the current console generation ages out and the Nintendo Switch 2 launch reshuffled competitive dynamics. Gaming content and services fell 5%. Segment operating profit slipped from $3.92 billion to $3.81 billion. Windows OEM remains close to 10% of total company revenue but isn't growing. At 17% of total revenue and shrinking, this segment is increasingly irrelevant to the investment thesis — it's the legacy business generating stable cash flows while the cloud and AI segments drive all incremental value creation.

The $625 Billion Backlog — Microsoft Stock's (NASDAQ: MSFT) Most Misunderstood Asset

Microsoft's commercial remaining performance obligations reached $625 billion at the end of Q2, more than doubling year-over-year with 110% growth. The weighted average contract duration is 2.5 years, unchanged even as the backlog has exploded in size. Approximately 25% of RPO — roughly $156 billion — is classified as current and expected to be recognized within the next four quarters. The current RPO grew 39% year-over-year, providing near-term revenue visibility that most companies can only dream about.

The controversy centers on composition. OpenAI accounts for approximately 45% of the total backlog, or about $281 billion, through a $250 billion infrastructure commitment plus additional contracted obligations. That leaves roughly $344 billion in backlog from Microsoft's broader customer base — enterprise customers already consuming Azure, Microsoft 365, and related cloud services. Even if every dollar of OpenAI's commitment were written off entirely, Microsoft would still have $344 billion in contracted revenue from diversified enterprise sources. The market appears to be applying a heavy discount to the entire $625 billion because of concentration risk from a single customer — a customer that just posted 233% revenue growth and is closing a $100 billion funding round.

The OpenAI commitment is structured as a ramp. Only about 14% of the current (next-12-month) backlog relates to OpenAI, meaning $14 billion of Azure infrastructure consumption is planned for the current calendar year. That jumps to $50 billion projected for 2027, with continued escalation through the life of the contract. The ramp structure means the risk is back-loaded — Microsoft has years to observe whether OpenAI's revenue trajectory supports the escalating commitment before the heaviest consumption periods arrive.

OpenAI — The $830 Billion Partner That Everyone Wants to Discount to Zero

Microsoft owns 27% of OpenAI. It was the exclusive cloud provider before OpenAI diversified infrastructure to include Oracle. Microsoft retains rights to OpenAI's IP models through 2032, allowing integration into its copilot products. And beyond the $250 billion infrastructure commitment, OpenAI has agreed to pay Microsoft 20% of its revenues through 2032 as a royalty — a payment stream that carries essentially 100% gross margin since it requires no incremental cost to Microsoft.

OpenAI's trajectory: ARR surpassed $20 billion in 2025, representing 233% year-over-year growth. The company burned $17 billion in cash last year against that revenue. CEO Sam Altman communicated to employees around the Super Bowl that monthly revenue growth had returned to 10%, implying roughly 150% annualized growth rather than the publicly guided 50% for the current year. If the 50% guidance holds, ARR reaches $30 billion in 2026. If the 10% monthly pace sustains, the number is significantly higher. OpenAI's intermediate-term forecast projects revenue reaching approximately $280 billion by 2030, based on capturing 25-30% of an estimated $900 billion total AI revenue market.

The $100 billion funding round — the largest private capital raise in history — is nearing completion with commitments from Nvidia ($30 billion), Amazon ($50 billion), SoftBank ($30 billion), and the Abu Dhabi sovereign wealth fund ($50 billion), plus multiple smaller participants. The post-money valuation is $830 billion. CFO Sarah Friar — previously Goldman Sachs software analyst and Square CFO — is preparing the company for an IPO at a projected valuation exceeding $1 trillion. The fact that Nvidia, Amazon, and a sovereign wealth fund are collectively writing $130 billion in checks to a company valued at $830 billion is perhaps the strongest possible signal that sophisticated capital allocators believe OpenAI can fulfill its infrastructure commitments.

The revenue share agreement is particularly valuable and chronically overlooked. If OpenAI reaches $100 billion in annual revenue by 2028-2029 — roughly consistent with its growth trajectory — the 20% royalty would generate $20 billion annually for Microsoft at 100% gross margin. Even if OpenAI defers some payments (the agreement reportedly allows this), the accrual alone would materially boost Microsoft's reported profitability. The market is assigning essentially zero value to this royalty stream.

15 Million Copilot Seats Against 450 Million MS 365 Users — The Seat Gap That Could Generate $7 Billion in Incremental Revenue

Microsoft's copilot ecosystem had 15 million paid seats at the end of Q2, up 160% year-over-year. MS 365 Copilot ARR exceeded $5 billion. GitHub Copilot has nearly 5 million users with ARR approaching $2 billion. These are impressive growth rates that are being entirely dismissed by a market obsessed with Azure capacity constraints and OpenAI risk.

Here's the math that matters: 15 million copilot seats represents 3.3% of total Microsoft 365 commercial seats. The remaining 96.7% — approximately 435 million seats — represents the "seat gap" that CEO Nadella has identified as a top priority. Microsoft 365 Copilot is priced at $30 per user per month. Converting even 10% of the remaining seat base would add approximately 43.5 million seats, or roughly $15.7 billion in additional ARR at current pricing. A 5% conversion adds $7.8 billion. The company's fiscal 2027 revenue target of $355 billion implies roughly $36 billion in dollar growth over the current year, and copilot adoption is expected to contribute approximately 20% of that increase based on the ARR disclosure trajectory.

More than 80% of the Fortune 500 now have active agents built using Microsoft's low-code/no-code Copilot Studio and Agent Builder tools. The company launched Agent 365 this quarter, extending governance, identity, security, and management controls to AI agents deployed on Microsoft's cloud or any other cloud. Partners including Adobe, Databricks, Nvidia, SAP, ServiceNow, and Workday are already integrating Agent 365. Microsoft positions itself as the first provider offering an agent control plane across multiple clouds — a platform lock-in strategy that, if successful, makes Microsoft the operating system layer for enterprise AI agent deployment.

Microsoft Stock (NASDAQ: MSFT) Balance Sheet — $75.38 Billion Net Liquidity and $12.7 Billion Returned to Shareholders in a Single Quarter

The balance sheet is a fortress. Cash and short-term investments: $89.46 billion. Long-term investments: $21.20 billion. Long-term debt: $35.38 billion. Net liquidity position: $75.38 billion. This is a company spending $37 billion per quarter on capex, returning $12.7 billion to shareholders through buybacks ($7.16 billion for 12 million shares) and dividends ($6.76 billion), and still accumulating cash. The buyback pace accelerated from $4.73 billion (8 million shares) in the year-ago quarter, indicating management's conviction that shares are undervalued at current prices.

The dividend continues to grow — $6.76 billion this quarter versus $6.17 billion a year ago. Combined with aggressive share repurchases, Microsoft's capital return program is reducing share count while increasing per-share cash generation. Management is not choosing between growth investment and shareholder returns — it's funding both simultaneously from operating cash flow that approaches $200 billion annualized.

Valuation — 22 Times Forward Earnings for 17% Revenue Growth, 116% EBITDA Expansion, and $625 Billion in Contracted Revenue

MSFT trades at approximately 22-24 times forward earnings depending on which consensus estimate is used. The forward P/E on fiscal 2026 estimates is roughly 23.7 times. On fiscal 2028 consensus EPS of $23.72, the multiple drops to 17.7 times — with 33.79% cumulative EPS growth expected over the next two fiscal years. The EV/S ratio is 7.8 times. These multiples are the lowest among the Mag 7 names. Apple trades at 33 times earnings with 27% EBITDA growth since 2021. Microsoft trades at 23-24 times with 117% EBITDA growth over the same period. The valuation gap makes no fundamental sense.

The first call consensus projects 16.4% revenue growth for the current fiscal year and 15.4% for fiscal 2027. Both estimates are likely conservative. The price increase taking effect July 1 isn't fully reflected in models yet. The copilot seat gap conversion is in early innings. OpenAI's revenue share payments haven't started flowing in meaningful amounts. And the 37-38% Azure growth guidance is a capacity-allocation number, not a demand number — actual demand growth is higher but constrained by infrastructure buildout timing. A reasonable independent estimate puts the three-year revenue CAGR at 17% or above, roughly 200 basis points above published consensus.

Free cash flow margin for the trailing twelve months is compressed by the capex surge — roughly 22% on a forward-four-quarter basis versus the 40% levels Microsoft historically achieved. But with the CFO signaling sequential capex easing and operating cash flow margins expanding from 42% to 51% year-over-year, the free cash flow inflection is imminent. As capex stabilizes and revenue continues growing at mid-to-high teens, free cash flow margins should expand materially starting in fiscal 2027. A return to 30%+ free cash flow margins on $355+ billion in revenue would generate well over $100 billion in annual free cash flow — a number the current market cap of roughly $2.95 trillion does not reflect.

Ratings — Wall Street at Strong Buy 4.71, Analysts at Buy 4.02, Quant at Hold 3.47

The consensus view from Wall Street analysts covering MSFT is Strong Buy with a score of 4.71 out of 5. Independent analyst ratings average Buy at 4.02. Only the quantitative models register Hold at 3.47, reflecting the recent price momentum damage and elevated capex ratios that quant screens flag mechanically. The disconnect between Wall Street's Strong Buy consensus and the stock's 28% drawdown represents either the analysts being wrong or the market being temporarily mispriced. Given the fundamental evidence — accelerating revenue, expanding margins, exploding backlog, and compressed valuation multiples — the market mispricing argument is far more compelling.

The Risks — OpenAI Concentration, GPU Useful Life, Copilot Execution, and the ¥160 Question Nobody's Asking

The risks are real and worth cataloguing honestly. OpenAI concentration at 45% of the total backlog is genuine concentration risk. If OpenAI's revenue growth decelerates sharply — say to single-digit monthly growth rather than the current 10% — its ability to ramp Azure consumption to $50 billion annually by 2027 becomes questionable. The infrastructure commitments span at least 8 years with a steep ramp profile, and the back-loaded structure means the largest payments come during the most uncertain forecast period.

GPU useful life assumptions carry risk. Microsoft depreciates short-lived assets over a 6-year period, consistent with Google, Meta, and Oracle (Amazon uses 5 years). If the pace of AI hardware innovation renders current-generation GPUs obsolete faster than the depreciation schedule assumes, the company would face write-downs that compress margins. Microsoft and Nadella have argued that software optimization extends the inference processing life of Nvidia chips, and the evidence supports this claim so far — but it requires continuous validation as new GPU architectures launch.

Copilot execution is an execution risk, not a market risk. The 3.3% seat penetration against the total 365 base could stall if enterprises determine that copilot ROI doesn't justify the $30/user/month price point. Competition from Anthropic's Claude Cowork and OpenAI's own consumer and enterprise products creates channel confusion — Microsoft's copilots are built using OpenAI's models, but OpenAI is simultaneously launching competing productivity tools. The competitive dynamics of being both partner and competitor with OpenAI are inherently unstable.

The gaming segment is in decline with Xbox hardware down 32% and no clear catalyst for reversal in the near term. While immaterial to the investment thesis at 17% of revenue and shrinking, it represents a drag on reported growth rates that feeds negative sentiment.

The Verdict — Microsoft Stock (NASDAQ: MSFT) Is a Buy at $397

MSFT at $397 is a buy. The stock is priced as though the AI infrastructure buildout will produce zero returns, the $625 billion backlog is fabricated, the copilot opportunity is dead on arrival, and OpenAI is going bankrupt. None of these things are happening. Revenue grew 17% last quarter at $81.3 billion. Operating margins expanded to 47%. EBITDA has more than doubled since 2021 to $175.26 billion. The balance sheet carries $75.38 billion in net liquidity. The backlog provides multi-year revenue visibility that no other technology company can match. And the stock trades at the lowest forward P/E among its Mag 7 peers.

The market's fixation on capex spending levels is a classic case of confusing investment with waste. Microsoft is not building data centers speculatively — it is capacity-constrained and losing revenue because it cannot deploy infrastructure fast enough. The $37 billion quarterly capex is a response to demand that exceeds supply, not a bet on demand that may never materialize. The profitability data since 2021 proves that Microsoft's capex converts to EBITDA at a ratio that justifies every dollar spent.

Buy MSFT at $397 as a core holding. The 28.5% drawdown from $555.45 has created the most attractive entry point in Microsoft shares since the Liberation Day trough at $359.84. The downside is defined by the $360 floor — a level that held during the worst market stress of 2025. The upside is defined by normalized valuation: at 25 times the fiscal 2028 consensus EPS of $23.72, the stock reaches $593. At 28 times — still below the multiple it carried for most of 2024-2025 — the target is $664. The risk-reward from $397 is asymmetric in the most favorable sense: roughly 10% downside to the tested floor versus 50-67% upside to fair value on consensus estimates that are probably too low. The free cash flow inflection in fiscal 2027, the July pricing increase, the copilot seat gap conversion, and the OpenAI revenue share payments represent four distinct catalysts that the current price assigns negligible probability. That's not risk management — that's mispricing. And mispricing in a $3 trillion company doesn't persist forever.

That's TradingNEWS