CrowdStrike Reports $1.47 Billion in Quarterly Revenue and Raises Full-Year Outlook; Shares Rise About 11% After Hours

Two Numbers, One Quarter, and a Fight Over Which One Counts

CrowdStrike told investors on August 26, 2026, that it had just booked $1.47 billion in revenue for the three months ending July 31 — up 26% from $1.17 billion a year earlier^[1][2]. Adjusted profit came in at $0.31 a share, beating the $0.29 Wall Street expected^[1]. The company also raised its forecast for the rest of its fiscal year, and its stock jumped more than 11% in after-hours trading^[1][4].

By the numbers CrowdStrike chose to lead with, this was, in CEO George Kurtz's words, "the best quarter in CrowdStrike's history"^[1]. But there's a second number sitting in the company's own filings that tells a very different story about the same 12 months: a $0.65-per-share loss, under standard accounting rules, for the prior fiscal year^[18]. Both numbers are real. They're just measuring different things, and which one you trust says a lot about how you see this stock.

What "Record" Actually Means Here

The figure CrowdStrike wanted the headlines to carry wasn't revenue. It was "net new ARR" of $332.8 million^[2][20]. ARR stands for annual recurring revenue — the yearly value of all the subscriptions a company currently has under contract. Net new ARR measures how much that total grew in a single quarter, after accounting for any customers who canceled.

CrowdStrike says $332.8 million was a company record, and 51% higher than the same quarter last year^[2]. Total ARR now stands at $5.84 billion, up 25% year over year^[2][20]. Management liked the trend enough to raise its full-year net new ARR growth target by 6.3 percentage points, to about 34% at the midpoint^[1][2].

Why lead with ARR instead of revenue? Because ARR captures a new contract the moment it's signed, even if the cash arrives slowly over three years^[2][17]. It's also the number that best supports the growth story the company is selling. And that story matters more than usual right now, because CrowdStrike's stock price already has a lot riding on it.

A Price That Was Already Betting on a Blowout

Before the earnings came out, CrowdStrike was trading at somewhere between 153 and 175 times its expected future earnings, and close to 29 times its expected sales, against an industry average nearer 17 times sales^[11][12]. That's not a minor premium. It means investors were paying for years of continued fast growth, not just for what the company had already proven it could do.

Analysts at Bank of America had warned clients ahead of the report that a beat needed to be at least 3% on revenue to count as a real win, given how far the stock had already run in 2026 and how it reacted badly to a smaller beat the previous quarter^[21]. The actual beat came in around 2.1%^[4]. By that yardstick, CrowdStrike cleared its own bar but fell short of the one skeptics had set.

That's the tension running underneath this whole story. The quarter was genuinely strong by almost any normal measure. Whether it was strong enough for a stock already priced for near-perfection is a separate question, and reasonable people who agree on every number can still land on opposite sides of it.

Two Sets of Books, Two Honest Answers

There's a similar split hiding inside CrowdStrike's profit figure. The $0.31-a-share adjusted profit that beat estimates excludes stock-based compensation — the shares the company hands employees instead of cash^[1]. Under standard accounting rules, which do count that as a real cost, CrowdStrike actually lost $0.65 a share for the full prior fiscal year^[18].

Neither number is fabricated. Companies are legally allowed to report both, and doing so is standard practice in software. But the choice of which one to feature is still a choice, and it's central to why the same company can be described in one breath as wildly profitable and in the next as unprofitable^[1][18].

CrowdStrike's own defenders make a specific case for why the adjusted number is the fairer one: for a subscription business, cash flow and contracted revenue tell you more about the health of the company than one accounting line does, because a customer signing a long-term deal shows up immediately in ARR but only slowly in reported earnings^[2][17]. Skeptics counter that stock-based pay is still a real cost that dilutes existing shareholders, whatever column it sits in^[4][18]. High valuations also make that stock-based pay cheaper for the company in cash terms, which gives management an extra reason to keep the growth narrative fresh.

The AI Threat Behind the Sales Pitch

CrowdStrike and much of the market credit this quarter to a shift in how companies think about artificial intelligence and hacking. In April 2026, Anthropic previewed a model called Claude Mythos that reportedly could find and exploit unknown software vulnerabilities on its own^[10]. A U.S. official later said the model had found flaws in classified government systems^[9]. CrowdStrike says that anxiety is now translating into eight-figure contracts with AI labs, banks, automakers, and utilities^[1].

The company's pitch is that if AI can write attacks cheaply, defenders need AI that can respond just as fast — which is what it sells through a product called Charlotte AI, described as "agentic," meaning it can take multi-step action on its own instead of just flagging a problem for a person to handle^[1][5]. CrowdStrike also points to Falcon Flex, a contract structure where customers commit a pool of spending they can apply to any CrowdStrike product. Flex-linked ARR passed $2.29 billion this quarter, up 101% from a year ago^[1][13].

Not everyone agrees the threat itself is new. More than 100 security practitioners, including people at Adobe and Nvidia, said AI models like Mythos are good at finding exploits but not uniquely good at it — automated tools and well-funded state hackers were already doing this work^[8]. CNBC reported that some experts called the market reaction "hysteria," arguing the underlying danger predates the AI headlines^[8]. If they're right, some of this quarter's rush of security spending could be organizations pulling next year's budget forward rather than a lasting shift — the single biggest question mark hanging over CrowdStrike's raised forecast.

The Risk Built Into the Sales Pitch Itself

There's an irony worth naming. CrowdStrike's growth strategy depends on convincing customers to put more of their security tools onto one platform, which genuinely does cut costs and complexity for the buyer^[13][17]. But it also means that if CrowdStrike's software fails, it fails everywhere at once. That's exactly what happened on July 19, 2024, when a faulty CrowdStrike update crashed millions of Windows computers worldwide.

Delta Air Lines is still suing over that outage. A judge let part of the case proceed past a motion to dismiss in May 2025, and it remains in discovery^[15]. None of this quarter's coverage tied the record results back to that unresolved lawsuit — a gap that's easy to miss when the headline number is 11%.

Coverage of the earnings split along familiar lines. CNBC adopted the company's own "Mythos moment" framing in its headline^[1]. TechTimes led its preview with the highest valuation multiple available, 175 times earnings, rather than the lower end of the range other outlets cited^[11]. The Motley Fool published its piece as an openly bullish pre-earnings case for the stock^[16]. And the closest thing to international coverage found, an Investing.com recap on its India edition, mostly relayed CrowdStrike's own investor slides without adding a valuation counterpoint^[13] — a reminder that a dedicated outside skeptical voice on this story was largely missing.