Once-in-a-Decade Opportunity: 1 AI Software Stock to Buy While It's Still Down

Agentforce reached roughly $800M ARR (up 169% YoY) and Agentforce + Data 360 hit $2.9B ARR (up 200% YoY), with >60% of bookings from existing customers and premium AI bookings nearly tripling sequentially. Salesforce reported fiscal 2026 revenue of $41.5B, RPO of $72.4B (current RPO $35.1B, +16% YoY), and Q4 saw deals >$1M up 26% and >$10M up 33% YoY. Management guides fiscal 2027 revenue of $45.8B–$46.2B (implying ~10%–11% YoY growth) and expects subscription & support growth to reaccelerate in H2 FY27; the stock trades around 13x forward earnings, suggesting a modest valuation given improving AI-driven metrics.

Analysis

Salesforce’s AI products are acting less like standalone features and more like a gravity well for enterprise data — that raises ARPU through upsells and consumption, but also shifts margin exposure toward volatile GPU/cloud spend and usage-based revenue. Expect meaningful P&L leverage only if consumption growth outpaces incremental infrastructure costs over the next 6–18 months; otherwise revenue growth can look healthy while gross margins compress.

The most important second-order beneficiary is the enterprise data orchestration layer: vendors that make multi-cloud data consistent (integration, cataloging, governance) will see disproportionate demand as customers operationalize AI agents. Conversely, hyperscalers and GPU suppliers capture an outsized share of incremental economics; that raises counterparty concentration risk into Nvidia/hyperscaler supply chains over the medium term.

Key catalysts to watch in the coming 2–4 quarters are large-deal cadence (sustained >$10M deal flow), consumption versus subscription mix trends, and any visible signs of model-cost deflation from base-model commoditization. Tail risks that could unwind the story include rapid LLM price drops that commoditize the agent layer, EU/UK privacy/regulatory constraints that slow deployments, or a single quarter of materially negative gross-margin surprise tied to infrastructure costs.

The current valuation implies the market has priced in only mild reacceleration — that creates an asymmetric trade if execution holds. But sizing must explicitly account for two-way risk: execution (integration/salesforce delivery) on a long horizon and model-cost deflation on the short horizon; use option structures to capture upside while capping tail losses.

Key Decisions for Investors

• CRM — Core long (6–12 months): Allocate 2–3% of portfolio to CRM equity and hedge with a 12-month 10% OTM protective put or buy a put spread to cap downside. Risk/Reward: target 25–35% upside if enterprise consumption scales; capped downside to ~8–12% (cost of hedge). Enter on any <8% post-earnings gap-down or on confirmation of sequential large-deal growth.
• INFA — Tactical long (6–12 months): Buy Informatica shares or a 6–12 month call spread sized at 1% portfolio to play Data 360 integration wins. Risk/Reward: asymmetric upside (~40–60%) if integration leads to material wallet-share gains vs ~20% downside if competitive displacement stalls. Stop-loss at -18% or on loss of major account references.
• NVDA — Levered exposure (3–9 months): Buy 3–6 month call spreads to express continued GPU tailwinds tied to enterprise agent deployments (size as funded risk). Risk/Reward: high upside capture if hyperscaler/GPU demand remains tight; downside limited to premium paid. Avoid outright long-term naked calls due to inventory/cycle risk ahead of earnings.
• Hedge/Pair — Long CRM / protection via model-cost view (3–9 months): If worried about LLM deflation compressing margins, pair a CRM long with buying short-dated puts on GPU/infra suppliers or a small notional long-vol position tied to NVDA/hyperscaler event risk. Purpose: preserve CRM upside while hedging the specific risk that model-cost deflation or infra shocks reverse sentiment quickly.