3 Reasons Why The Trade Desk Is a Screaming Buy Right Now

CEO Jeff Green purchased roughly $150 million of The Trade Desk stock on the open market. The stock trades at ~14x forward earnings versus the S&P 500 at 21.7x, while revenue growth slowed to 14% in Q4 and is projected at 10% in Q1. Reports that The Trade Desk may place ads on OpenAI/ChatGPT represent a potentially material growth catalyst that could accelerate user monetization and meaningfully re-rate the stock if realized.

Analysis

If generative-AI surfaces a new, high-attention ad surface, the obvious winner is the party that controls neutral marketplace rules and measurement primitives rather than raw inventory. That shifts the economic battleground from pure CPM arbitrage to take-rate economics, attribution fidelity, and latency/serving cost — areas where platform-agnostic DSPs can extract higher margins if they own identity and deterministic measurement hooks.

Compute and infrastructure are a second-order beneficiary: sustained AI ad load increases demand for low-latency inference and privacy-preserving measurement (server-side tagging, on-device features). That flow benefits GPU/accelerator suppliers and cloud partners, while also raising the variable cost base for ad-serving; margin capture will depend on contract terms with the AI host and whether the marketplace can keep a visible, auditable auction rather than a revenue-share black box.

Key risks are structural: (1) an AI host taking the role of both inventory owner and auctioneer (margin extraction, opaque pricing); (2) rapid advertiser reticence over brand safety/measurement leading to slower-than-modeled CPMs; and (3) regulatory moves restricting personalized advertising or forcing greater transparency. Time-to-materiality is not instantaneous — expect announcement-driven price moves within months and revenue/GM impact to crystallize over 12–24 months as buyers test and scale AI-native campaigns.

The consensus prize is exclusivity and outsized top-line upside; the blind spot is execution friction — integration, measurement standards, and revenue-share negotiation — any of which can compress forward EBITDA by 200–400bps versus a clean win. Position sizing should therefore target asymmetric optionality while protecting downside through collars or selling premium around known event windows.

Key Decisions for Investors

• Long TTD via a 12–24 month call spread (buy long-dated OTM calls, sell higher strike calls) sized to 1.0–1.5% of fund NAV. Rationale: asymmetric upside if AI-native inventory scales; target 2x payoff if cadence of advertiser testing converts to +5–8% incremental revenue within 18 months. Risk: limited to premium paid, edge case loss if the hosting platform captures economics.
• Buy-and-collar TTD equity for a core position (hold 12–18 months): purchase shares, buy 12-month put ~25% OTM, sell 12-month call ~40% OTM to finance protection. Rationale: captures ongoing cashflows while limiting downside to ~25% over 12 months and capping upside near 40%, appropriate given execution/regulatory risks.
• Small, tactical long NVDA (12-month calls, 0.5% NAV) as a correlated hedged upside to AI compute demand. Rationale: if AI monetization drives sustained inference load, NVDA captures disproportionate infrastructure upside — provides positive convexity if ad monetization accelerates but TTD execution lags. Risk: NVDA correlation means this is not a hedge to TTD downside arising from ad-market contraction.
• Event-driven short idea: if an AI host announces an exclusive, opaque revenue-share model that meaningfully reduces marketplace take-rates, buy puts on TTD (6–12 month) to hedge or scale short exposure opportunistically. Trigger: announcement with >20% estimated take-rate concession; target payoff if EBITDA margin contracts by 200–400bps. Risk: false signal if announcement is non-exclusive or accompanied by favorable measurement terms.