Down 22%, Is It Time to Buy CrowdStrike Stock on the Dip?

CrowdStrike reported fiscal Q4 revenue of $1.3B (+23% YoY) and record ARR of $5.2B (+24%), with Falcon Flex contributing $1.7B ARR (+120%). Adjusted (non-GAAP) profit rose 17% to $956.5M, while GAAP results showed a $162.5M loss for the year. Shares are ~22% below their all-time high and trade at a P/S of 22.3, but management forecasts ARR of $20B by FY2036 (a 284% increase), implying a forward P/S near 5.3 if achieved. Results and the long-term ARR target support upside for long-horizon investors, but high current valuation and the need to convert guidance into long-term execution present material risk.

Analysis

CrowdStrike’s scale in telemetry and model-trained detection is a real moat, but the incremental economics of a Flex-style modular subscription change the math: as customers dynamically toggle modules, ACV visibility falls and short-term ARR can swing more than traditional per-seat deals. That amplifies the importance of net retention and gross-margin per ingested event — if telemetry storage/ingestion costs rise faster than per-module pricing, margin expansion assumptions baked into long-range forecasts will be hard to realize.

A second-order beneficiary of CrowdStrike’s growth is the downstream ecosystem that stores and processes security telemetry (cloud infra, observability stores, GPU inference nodes for AI models); increased telemetry volumes drive more cloud spend and make pricing negotiations with hyperscalers a real operating lever and sourcing risk. Conversely, hyperscalers or large platform vendors bundling baseline detection could cap upsell or force CrowdStrike to trade higher subscription density for broader distribution.

Key tail risks are commoditization of AI-driven detection (faster replication by well-capitalized rivals), macro-driven enterprise IT belt-tightening that defers non-mission-critical modules, and the company’s continued reliance on non-GAAP adjustments to show profitability — any sustained uptick in stock-based comp or telemetry cost could re-rate margins downward. Near-term catalysts to watch are trends in net retention, module churn, and any changes in gross margin per ingest; those metrics will move the needle on whether long-term targets are credible or aspirational.

On the margin, the market is pricing a ‘perfect execution’ path for a decade; the current pullback is a partial repricing but not a bake-in of multi-year execution risk. That creates a tradeable asymmetric payoff for investors willing to take multi-year optioned exposure while protecting premium paid against near-term volatility tied to guidance and macro noise.

Key Decisions for Investors

• Long-duration, hedged CRWD LEAP: Buy Jan-2028 CRWD 2x-year LEAP calls (one OTM strike) financed by selling a higher strike to form a call spread; allocate 1.5-2% portfolio. Timeframe 24–36 months. R/R: limited premium outlay with 3x+ upside if net retention stabilizes and gross margins expand; stop/trim if two consecutive quarters show falling net retention or a >200 bps guide cut.
• Pair trade (quality skew): Long CRWD equity / Short ZS equity, equal notional, reweight 60/40 in favor of CRWD. Timeframe 12–24 months to capture differential in telemetry scale and product breadth. R/R: long CRWD captures optionality of platform monetization while short ZS hedges sector risk; unwind if CRWD’s churn rises >150 bps or ZS reports clear enterprise win-rate acceleration.
• Tactical short on PANW via put spread into next 2 earnings (6–12 months): Buy 6–9 month PANW put spread (OTM protection bought, further OTM sold) sized 0.5–1% portfolio. Thesis: slower cloud-native transition and hardware/Appliance exposure compress multiple if enterprise spend softens. Risk: sharp sector-wide re-rating higher; cap downside via spread structure.
• Thematic hedge with NVDA: Buy 12-month NVDA calls (smaller notional, 0.5% portfolio) to hedge upside in infrastructure demand should AI-driven security workloads materially outgrow expectations. Timeframe 12 months. R/R: asymmetric upside if GPUs see incremental demand from model inference/security workloads; loss limited to premium if AI security spending disappoints.