SpaceX Stock Now Has a Price to Sales Ratio Over 115x. Is It Worth Buying Anyway?

SpaceX’s valuation is cited at ~116x trailing-12-month revenue ($19.3B), which remains “incredibly high” even after the post-IPO rally. However, three AI compute deals (Anthropic, Alphabet, Reflection AI) are expected to add about $2.32B/month—~$27.8B annualized—starting in October, supporting a 2027 revenue base case of ~$59B (~38x P/S on current valuation). Despite this growth setup, the article cautions that the stock still looks expensive and profitability in 2027 is uncertain.

Analysis

This is more useful as a sentiment read-through than a direct equity catalyst. The market takeaway is that private AI infrastructure can still re-rate on forward revenue visibility, but only if the revenue is sticky and not one-off capacity booking; that keeps the scarcity premium alive for compute, power, and networking suppliers. The public-market beneficiary with the cleanest second-order exposure is NVDA, but only if this kind of demand proves broad-based across customers rather than a single headline deal.

The risk is that investors extrapolate a 2027 revenue bridge that depends on multiple moving parts: customer retention, delivery timing, and whether these contracts translate into free cash flow after capex and operating costs. If any one leg slips, the valuation compresses fast because the starting multiple is so extreme that even strong growth can leave the stock expensive. For public names like GOOG/GOOGL, the direct financial impact looks immaterial; the only real read-through is that hyperscalers may continue outsourcing edge-case compute, which supports industry pricing but does not move Alphabet materially.

Contrarian view: consensus is likely underestimating how much of this is narrative-driven secondary-market pricing rather than audited earnings power. If the deal flow is real, it validates AI infra demand; if it is mostly internal transfer pricing or cancellable commitments, the multiple should come in hard over the next 3-6 months. The key falsifier is any evidence that those deals do not convert into recurring revenue or that forward margins remain negative despite the top-line step-up.

Key Decisions for Investors

• No immediate trade in GOOG/GOOGL off this article; treat any weakness as non-fundamental noise unless upcoming cloud-capex or margin guidance implies a real customer loss within 1-3 months.
• Add NVDA on pullbacks over the next 4-8 weeks only if broader AI capex data confirms sustained demand; risk/reward is favorable as a validation trade, but invalidate if hyperscaler capex growth decelerates next quarter.
• Do not chase SPCX secondary-market pricing on this headline; require proof of recurring revenue conversion and positive gross-profit trajectory before getting constructive, because the downside from multiple compression is large if execution slips.
• Set an alert for any follow-up filing or third-party verification of the AI compute contracts over the next 30-90 days; if the revenue is not recurring or is customer-concentrated, that is a de-risk event for the entire AI-infra complex.
• If exposed to private AI infra marks, trim into strength rather than adding: this is a good reminder that forward-sales multiples can stay high, but only until growth becomes less certain.