Databricks in talks to raise capital at $134 billion valuation, The Information reports

Databricks is reportedly in talks to raise $5 billion at a $134 billion valuation, implying roughly 32x this year’s expected sales of about $4.1 billion. The company has raised its sales projections multiple times and now expects revenue to grow ~55% this year, but warned gross margin is slipping to 74% from an earlier plan of 77% due to heavier usage of its AI products; the size of the raise and lofty valuation highlight strong investor interest ahead of a potential IPO and may recalibrate private-market comps for AI-focused software firms.

Analysis

Market structure: A $134bn private valuation for Databricks at ~32x 2025E sales signals accelerating enterprise AI demand that directly benefits AI compute and infrastructure suppliers (GPU makers, server OEMs, SMCI) and hyperscalers that can monetize model run-time. It also pressures high-multiple SaaS vendors that rely on steady gross margins because usage-based AI will shift revenue mix toward variable cloud/compute costs, compressing software gross margins by several hundred basis points within 12–24 months.

Risk assessment: Tail risks include regulatory limits on AI model training (privacy/exports) and a macro IT-spend pullback; these could cut projected compute demand by 30–50% in a downside scenario. In the next days–weeks expect volatility around funding/IPO talk; over 6–18 months monitor gross-margin trends (Databricks cited a fall from 77% to 74%) as an early warning for wider margin compression across the sector.

Trade implications: Direct plays: overweight SMCI (server infra exposure) and NVDA (GPU demand) while underweight high-valuation SaaS/analytics names trading >20x sales. Use pair trades: long SMCI vs short IGV (iShares U.S. Software ETF) to capture relative re-rating; implement 3–9 month call spreads on NVDA to express compute upside with limited premium. Size ideas: 1–3% portfolio per idea, stop-loss 12–15%, target 20–40% upside in 3–12 months.

Contrarian angles: Consensus may underprice the margin squeeze — public SaaS multiples can compress before Databricks IPO, creating entry points in infrastructure names after a pullback. Historical parallel: Snowflake’s IPO re-priced cloud data stacks but benefited infrastructure suppliers later; if Databricks passes through cloud costs to customers, software margins could compress further, creating a longer runway for SMCI/NVDA but a shorter one for high-multiple SaaS.

Key Decisions for Investors

• Establish a 2–3% long position in SMCI (Super Micro Computer) within 2–6 weeks to capture server demand from enterprise AI; set a 12–15% stop-loss and target +30% price appreciation over 3–12 months.
• Buy a 6–9 month NVDA call spread (bull call spread sized to 1–2% portfolio risk) to express GPU demand upside while capping premium; exit or roll if NVDA rises >40% or implied vol >80% (close to earnings/event risk).
• Implement a pair trade: go 2% long SMCI and 2% short IGV (iShares U.S. Software ETF) to capture relative outperformance of infra vs high-multiple software over 3–9 months; close if spread narrows by 50% or within 9 months.
• Trim/underweight positions in SaaS names trading >20x forward sales by reducing exposure 20–40% within 30 days (e.g., reallocate proceeds to infra names), because margin compression from usage-based AI could hit revenue multiples.
• Avoid meaningful pre-IPO exposure to Databricks; instead, monitor sequential gross-margin disclosure and customer cloud pass-through language over next 90–180 days and only consider public-comp arbitrage after S-1 transparency or post-IPO two-quarter performance.