CRWD · Analyze

Price at analysis: $227.96 (post-print, +20.5% same-day, new all-time high; prior ATH $227.50). position. First full-scored /analyze for this name — no prior conviction or trim level existed.

0. Knowledge check

python .mcp/kb.py find CRWD returned no matches — nothing on file for this ticker. This report is the first record. Two field-level pitfalls applied below on general vendor-data grounds (forward-EPS fiscal-year mismatch; TTM-as-sum-of-quarters), both already logged in Knowledge/Playbook/.

1. General Health (Fundamentals)

Guard applied — TTM is the sum of four quarters, not one column, and the freshest quarter (FQ2 FY27, ended 2026-07-31, reported 2026-08-26) is not yet in the Yahoo/fin.py feed. All TTM figures below are hand-rolled: the prior TTM (through FQ1 FY27, Apr-26) from Yahoo's quarterly statements, with the oldest quarter (FQ2 FY26, Jul-25) dropped and the just-reported FQ2 FY27 figures (from the earnings release / call) substituted in.

Metric	Value	Basis
Revenue (TTM, through Jul-26)	~$5.40B	$5.094B (Yahoo TTM thru Apr-26) − $1.169B (Jul-25) + $1.47B (Jul-26, company-reported)
Revenue growth (latest qtr, YoY)	+26%	Company release; accelerating for the 5th straight quarter
Revenue 3yr CAGR (annual FY23→FY26)	29.0%	fin.py
ARR (ending, Jul-26)	$5.84B, +25% YoY	Accelerating for the 4th straight quarter
Net new ARR (FQ2 FY27)	$333M, +51% YoY	All-time record; beat guide midpoint by >$45M
FCF (TTM, through Jul-26)	~$1.523B	$1.431B (sum of 4 Yahoo quarters thru Apr-26) − $285.0M (Jul-25) + $377M (Jul-26, company-reported)
FCF margin (latest qtr)	26%	Guide for FY27: ≥30% full-year
FCF 3yr CAGR (annual)	22.5%	fin.py
Non-GAAP operating margin (latest qtr)	25%, +350bps YoY	Record $372M non-GAAP operating income
GAAP net income (latest qtr)	+$5M	3rd consecutive quarter of positive GAAP earnings (Jan-26, Apr-26, Jul-26); Oct-25 was still a loss
Non-GAAP net income (latest qtr)	$323M / $0.31 diluted EPS	+34% YoY split-adjusted
SBC (TTM, through Apr-26)	~$1.147B, ~22.5% of TTM revenue	Sum of 4 quarterly SBC lines — this is the entire GAAP-to-non-GAAP bridge
Debt / Assets	~7.3%	$821M debt / $11.27B assets (Apr-26)
Cash	$5.01B (post-print) vs. $821M debt	Net cash ~$4.2B — fortress balance sheet
Diluted shares (latest qtr)	1.032B, +3.2% YoY	SBC-driven issuance partly offset by opportunistic buybacks ($176M in the Apr-26 quarter)
FCF/share (TTM)	~$1.48	$1.523B ÷ ~1.03B diluted shares

Data trap caught mid-calc: Yahoo's get_stock_info snapshot field freeCashflow reports $1.926B, but summing the four quarterly Free Cash Flow columns in the same feed gives $1.431B (through Apr-26) — a ~$495M (35%) gap on the same ticker, same pull. operatingCashflow in the same snapshot, by contrast, ties out exactly to the quarterly sum ($1.819B both ways). Used the quarter-sum figure throughout; see pitfall note below.

Capital allocation: cash flows to (1) growth R&D/S&M — the dominant use, funding the AIDR/next-gen-SIEM/identity build-out, (2) modest opportunistic buybacks (~$176M/qtr, well below SBC issuance, so net dilution persists at ~3%/yr), (3) tuck-in M&A (Apr-26 quarter alone: $881M, largely the pending XM Cyber technology-asset deal — explicitly excluded from FY27 ARR/revenue guidance per the CFO), (4) zero dividends. No debt paydown needed — leverage is immaterial.

Verdict on health: genuinely strong and accelerating — revenue, ARR, net new ARR, operating margin, and FCF margin are all inflecting upward simultaneously, which is a real signal, not just a comp effect. The one asterisk is SBC: at ~22% of revenue it is the entire reason GAAP is barely breakeven while non-GAAP prints a fat margin. That gap is normal for this company's history and is narrowing (three straight GAAP-profitable quarters), but it means the non-GAAP multiple below should be read with that discount in mind.

2. Moat & Competitive Advantage

Quantitative base. Gross margin 75% (TTM) trending toward the low-80s (non-GAAP subscription gross margin 81%, +90bps YoY, against an explicit FY29 target of 82–85%). ROIC is not yet a meaningful metric — GAAP earnings are near breakeven — so the better lens here is Rule of 40: revenue growth (26%) + FCF margin (26–30%) = 52–56%, well above the 40 threshold and rising. That is the moat signal a pre-full-margin platform business should be judged on.

Revenue-stream map. Endpoint (core, re-accelerating for the 4th straight quarter) → Next-Gen SIEM ($695M ending ARR, +60% YoY, displacing legacy Splunk/QRadar-style incumbents using CrowdStrike's already-resident telemetry as the wedge) → Identity ($585M+, +34% YoY) → Cloud (~$905M, +29% YoY — note: the call transcript garbled this figure as "$105 million" in one spot vs. the CFO's explicit "$905 million," used the CFO figure) → the new AIDR (AI Detection & Response) line, which management says is growing faster than any prior product launch and which Kurtz argues could eventually exceed core EDR in size. Falcon Flex — the platform bundling/consumption vehicle — is the structural story: ending ARR >$2.29B, +101% YoY, 34% of Q2 net-new-ARR came from new-logo Flex deals, and customers converting from point subscriptions to Flex show a >40% average ARR uplift. Module depth (51%/35%/26% of customers on 7/8/9+ modules) is rising, which is the actual switching-cost moat — not brand, a genuine multi-product lock-in.

Adversarial stress-test — "you are a well-funded rival, how do you attack this?" The honest answer is: Microsoft already is. Microsoft Defender for Endpoint is bundled at zero incremental cost inside M365 E5 licensing — for any organization that has already paid for E5, Defender is "free," and for the large mid-market segment that doesn't need best-of-breed efficacy, that procurement math is close to unbeatable on price. Independent 2026 competitive coverage frames the market as three converging forces: Microsoft winning enterprise RFPs on bundle economics, SentinelOne winning switchers still reacting to the July 2024 CrowdStrike outage, and CrowdStrike holding the strategic high ground on threat-intelligence depth and detection efficacy. This is a real, well-funded, structural threat — not a hypothetical one — and it caps how much pricing power CrowdStrike has at the low end of the market indefinitely. CrowdStrike's counter is genuine (first-party telemetry from the largest EDR install base compounds into better AI models for AIDR/SIEM, and Falcon Flex raises the effective switching cost of leaving), but "we win on quality, they win on price" is a durable equilibrium, not a knockout.

Disruption forecast (5–10yr arc). The AI-driven attack-surface expansion Kurtz calls the "Mythos moment" is a genuine secular tailwind — more agents, more non-human identities, more attack surface, and cybersecurity spend becomes closer to non-discretionary as a result. That argues for durability of demand. It does not argue for durability of CrowdStrike's specific share of that demand against a hyperscaler that can subsidize security as a loss-leader for its core franchise indefinitely.

Evergreen assessment. Yes on the category (cybersecurity doesn't go away), qualified on the company (best-of-breed product moat, not a structural one — vulnerable over a long horizon to "good enough and free," the same dynamic that has compressed margins in other software categories bundled into hyperscaler platforms).

Post-outage recovery. Notably, the July 2024 outage was not raised once by any analyst on the FQ2 FY27 call — two years on, the recovery in growth, retention, and pipeline metrics looks structurally complete at the enterprise level, even as SentinelOne continues to harvest some switchers at the margin.

3. Valuation

Per analysis_notes.md — Graham and dividend-based models are explicitly the wrong tools here; apply the fair-value approach conditionally.

• Graham IV ($2.48 vs. $227.96, from fin.py) — discard. Trailing GAAP EPS ($0.06) and BVPS ($4.55) are both artifacts of a heavily-SBC'd, barely-GAAP-profitable balance sheet; the formula is not meaningful for this name and is reported here only to show it was checked, not ignored.
• DYT / DDM — N/A, no dividend.
• Bogle's Expected Return — 0% yield + strong underlying earnings/FCF growth (accelerating into the 25–34% range on the metrics that matter) minus a very large expected multiple-normalization drag. The growth term is strongly positive; the multiple-change term is the dominant risk. Net expected return over a multi-year horizon is genuinely uncertain and could easily be negative even with continued 25%+ growth, purely from multiple compression — this is the central tension in this valuation.
• FCF multiple (primary method, per the guard's instruction to weight FCF/ARR/DBNR over Graham). FY27 guide implies ~$1.8B FCF (≥30% margin on $5.99–6.01B revenue) ÷ ~1.04B diluted shares ≈ $1.73/share forward FCF. Applying a demanding-but-bounded growth-FCF multiple of 70–100x (justified by Rule-of-40+ economics and genuine margin-expansion runway, well above what a mature SaaS name commands, well below today's implied multiple) gives fair value of $121–$173.
• Forward non-GAAP EPS cross-check. FY27 guide midpoint is $1.255 (range $1.25–1.26). At a 100–140x multiple (still a rich growth premium, PEG ~4–5.5x on 25% growth) that is $126–$176. The two methods converge tightly.

Fair value range: $130–$180.

Guard fired — vendor forward P/E is the wrong fiscal year. Yahoo's forwardPE (143.17x) implies a forward EPS of $1.59, which does not match the company's own FY27 guide of $1.25–1.26 — it is one full fiscal year further out (consistent with an FY28 consensus estimate). The field that actually lines up with the current-year (FY27) guide is priceEpsCurrentYear / epsCurrentYear ($1.234, implying 184.67x) — close to, and more reliable than, forwardPE for this name. The correct current-year forward multiple, computed off the company's own guide, is $227.96 ÷ $1.255 ≈ 182x — not the 143x a careless read of forwardPE would suggest. That 39-point gap (143x vs. 182x) is exactly the trap the guard warned about, and it makes the stock look meaningfully cheaper than it is if you grab the wrong field.

At 182x forward non-GAAP EPS, ~31x forward revenue (EV $188.9B ÷ FY27 revenue guide midpoint ~$6.0B), and a PEG of 6.0–6.15x, CRWD is priced for several more years of uninterrupted 25–30%+ growth with no multiple normalization. The market may get that — the fundamentals genuinely are accelerating — but the fair-value work above puts intrinsic worth $50–100 below the current print, and the framework's stated bias is "rather miss an opportunity than overpay."

Entry (add zone): $120–$140 — roughly the FCF/EPS-based fair-value floor, a level last seen before the current run. Trim: 145x forward non-GAAP EPS (≈$182 on the $1.255 FY27 guide midpoint) — just above the top of the fair-value band. Current price ($227.96, ~182x fwd) is already above this level, i.e. the name sits in Trim Zone today, not on the edge of it.

4. Sentiment & Intelligence

The 8/26 print, in full. Beat and raised on every guided line: net new ARR $333M (+51% YoY, a company record, beating the high end of guidance by $45M+), ending ARR $5.84B (+25% YoY, 4th straight quarter of acceleration), revenue $1.47B (+26% YoY, 5th straight quarter of acceleration), non-GAAP operating margin 25% (+350bps YoY), record FCF $377M (26% of revenue), and a third consecutive quarter of positive GAAP net income. Management raised the FY27 net new ARR growth outlook by 630bps to ~34% YoY at the midpoint — a $116M raise to the full-year net-new-ARR guide in just two quarters, and the second raise this year.

Management tone is unambiguously confident and thematically consistent: the "Mythos moment" framing — AI expanding the attack surface, agents "going rogue," and cybersecurity spend rising because of AI rather than being deflated by AI (Kurtz: "AI is driving more cyberattacks. AI is driving more cyber spending") — is the same narrative that drove CRM, CRWD and OKTA to all beat-and-raise on the same day, and TEAM three weeks earlier. This is a genuine sector-level inversion (AI as a demand driver for software, not a disruption threat) that shows up as 15–35% single-session moves rather than the 3% drift a normal beat produces.

Stock reaction. +20.5% on the day to a new all-time high ($227.96, prior ATH $227.50), part of a simultaneous cybersecurity-complex re-rating (CRWD, FTNT, DT, MANH, OKTA, TEAM, RBRK all hit 1-year highs together per the prior watchlist scan). Analyst consensus has not caught up: the mean target price ($210.53) sits below the current spot despite an "average rating: Buy" (1.7) — the rating is stale relative to the print, and estimate revisions have not yet fully absorbed the raise.

Insider activity. CEO Kurtz has sold shares on a near-weekly cadence for many months — consistent ~20,000-share tranches (roughly $3–4.5M each at prevailing prices), a pattern strongly consistent with a scheduled 10b5-1 diversification plan rather than a reactive signal. Other execs (CFO Podbere, President Sentonas, director Gandhi) show similar routine, small-relative-to-holdings sales. No insider buying observed in the transaction history pulled — normal for a name at all-time highs, not itself a red flag, but also not a golden flag either.

Data gap, flagged per the framework's skepticism default: management stated dollar-based net and gross retention rates "improved sequentially" on the call but did not disclose the absolute percentage; it was not surfaced in the press-release excerpts searched either. Treat DBNR as directionally positive but unquantified until the 10-Q/next disclosure — do not assume a specific number.

5. Synthesis & Weighted Verdict

Business quality vs. market pricing — the framework's first-principles test. This is squarely the "Great company, Expensive price" cell: the fundamentals case is arguably the strongest it has been in the company's history (five straight quarters of accelerating revenue growth, four of accelerating ARR growth, margin expansion across GAAP and non-GAAP, a real platform-consolidation flywheel in Falcon Flex, and a secular AI-driven demand tailwind that is, for once in 2026, working for a software name rather than against it). Set against that: a ~182x current-year forward earnings multiple, ~31x forward revenue, and a PEG above 6 — pricing that assumes several more years of uninterrupted 30%+ compounding with no multiple normalization, on top of a structural, well-funded competitive threat (Microsoft's bundle economics) that the company has not eliminated, only out-executed against so far.

Weighting. For a high-growth SaaS platform name, Moat and Sentiment carry more weight than a static Graham valuation per the house conflict-resolution rule — and both are genuinely constructive here. But the Valuation Analyst's FCF/EPS-anchored fair-value work ($130–180) sits meaningfully below spot ($227.96) even after giving full credit to the accelerating growth, and the conservative bias in analysis_notes.md ("rather miss an opportunity than overpay") points toward treating this as a valuation-driven profit-taking signal rather than a business-quality downgrade. This is not a thesis break — nothing in the print, the moat, or the competitive picture argues for selling the position down to zero — it is a call to trim into a 20%+ single-day pop to an all-time high, on a multiple that has run well ahead of even a generous read of the fundamentals.

Conviction: 6.0/10. Held back from higher purely by price — the underlying business, on this print, would earn a materially higher score at a lower entry.

Key risks: - Valuation/multiple-compression risk — ~182x forward non-GAAP EPS, ~31x forward revenue, PEG ~6x prices in years of flawless execution; CRWD's own history (2022 de-rating, the 2024 outage crash) shows this multiple is not sticky when growth or trust wobbles. - Microsoft Defender XDR bundling — a structural, well-funded threat that wins on procurement economics at the middle market regardless of CrowdStrike's product efficacy edge; SentinelOne continues to absorb post-outage switchers at the margin. - SBC-driven GAAP/non-GAAP gap — SBC at ~22% of TTM revenue is the entire bridge between a barely-breakeven GAAP business and a fat non-GAAP margin; the non-GAAP multiple ignores a real, recurring economic cost. - Consensus lag — analyst price targets have not caught up to the post-print level, meaning the stock now needs continued upside surprises just to keep the "Buy" rating's math intact.

Data Traps Hit (for orchestrator write-back to Knowledge/Playbook/)

1. pitfall-vendor-forward-eps-is-the-wrong-fiscal-year — CONFIRMED, same pattern. CRWD's forwardPE (143.17x) / forwardEps ($1.59228) maps to what is almost certainly an FY28 consensus estimate, not the company's own just-issued FY27 guide ($1.25–1.26 non-GAAP EPS). The field that actually tracks the current fiscal year is priceEpsCurrentYear/epsCurrentYear (184.67x on $1.234 EPS) — close to the correct hand-computed 182x off the company's own guide midpoint. Recommend adding CRWD as a named example in the existing note: this is now confirmed on 20/20 names checked in the 8/26–8/27 batch.

2. New candidate pitfall — Yahoo snapshot freeCashflow diverges from the sum of its own quarterly Free Cash Flow columns, independent of the JV/lease-related causes already on file. On CRWD, get_stock_info's freeCashflow field returns $1.926B while summing the four quarterly Free Cash Flow values in get_financial_statement(quarterly_cashflow) gives $1.431B (same ticker, same pull, through the same quarter, Apr-26) — a $495M / 35% gap. By contrast operatingCashflow in the same snapshot ties out exactly to the quarterly sum ($1.819B both ways), isolating the divergence to the FCF field specifically (i.e., the snapshot's capex assumption disagrees with the quarterly-statement capex). This is a second, independent confirmation of the existing pitfall-yahoo-snapshot-fcf-field-diverges-from-quarter-sum note (logged 2026-08-27, currently medium confidence) — recommend raising that note's confidence to high given a second clean, large, reproducible instance on a different name the same day.

3. Correction, not a new trap — CRWD's dual-class structure is retired, not live. The task guard flagged "founder/dual-class structure (Kurtz)" as a risk to check. Confirmed via WebSearch: CrowdStrike's Class B super-voting shares (10 votes/share) were fully converted to Class A and the dual-class structure was sunset in FY2025 — the company has traded one-share-one-vote since. sharesOutstanding and heldPercentInsiders (1.47%) in the Yahoo feed are not subject to the dual-class-FPI corruption pattern in pitfall-yahoo-share-count-dual-class-fpi for this name going forward. Worth a note only if that pitfall's text implies CRWD is still dual-class anywhere — it is not, as of this analysis.

Sources

CrowdStrike FQ2 FY2027 earnings call transcript (roic.ai, 2026-08-26); CrowdStrike Q2 FY2027 press release coverage — Seeking Alpha, 24/7 Wall St., Yahoo Finance UK, CNBC; competitive landscape — Palo Alto Networks Cyberpedia, Decryption Digest; Yahoo Finance MCP (get_stock_info, get_financial_statement, get_holder_info); .mcp/fin.py CRWD --news.