McGraw Hill at Goldman Sachs Communacopia + Technology Conference: ai push

McGraw Hill said it is exceeding the growth and margin targets set at its IPO, with revenue growing about 2% annually, margins in the high 30% range, and digital products comprising roughly 70% of revenue. Management highlighted more than 100 million paid curriculum licenses and 7.5 million AI-product users, while piloting agentic-AI offerings with over a dozen universities and a pharmaceutical company, including potential 5%-10% pricing uplifts. Growth opportunities include literacy reform across 44 U.S. states, international education spending growth of 4.5%-6.5%, and expansion of Inclusive Access beyond its current 2,000 U.S. institutions. Key risks remain long adoption cycles, AI trust concerns—including recent classroom chatbot bans in Los Angeles and New York—and fragmented local education regulation.

Analysis

The investable issue is not whether MH can demonstrate AI functionality, but whether it can convert a regulated installed base into recurring ARPU expansion without sacrificing its historically defensive margin profile. A 5%-10% AI access surcharge, if adopted across even a modest portion of existing digital licenses, has materially higher incremental margins than new curriculum sales; however, pilots are not evidence of willingness to pay. The near-term valuation rerating therefore depends on disclosed attach rate, net revenue retention and realized pricing—not user counts or product-release cadence.

MH is relatively insulated from generic-AI disruption because district procurement, curriculum approvals and outcome accountability create switching costs that consumer-oriented education platforms cannot readily overcome. That moat also slows the upside: K-12 decision cycles and state-level adoption calendars make a rapid AI revenue inflection unlikely. Pearson (PSO) faces similar opportunities but has greater exposure to a broad digital-transition narrative, while MH's blended print capability could become a quiet share-gain lever if screen-time restrictions broaden in elementary education.

The underappreciated 6-18 month catalyst is literacy procurement rather than agentic AI. New pedagogical mandates can trigger replacement cycles, and MH's localization capabilities may improve bid economics in politically fragmented districts; California is a particularly binary share opportunity. The principal downside is that budget pressure, delayed district awards, or an AI product rollout that requires incremental service and cloud costs faster than pricing can offset would expose the gap between low-single-digit core growth and an AI-growth valuation.

Contrarian view: the market may over-credit a large license base as monetizable AI distribution. Education buyers frequently demand AI features be bundled into existing contracts, while institutions can use multiple chatbot vendors and retain content providers as low-priced inputs. A durable rerating requires proof that AI improves renewal rates or expands contract value; otherwise MH remains a steady curriculum asset with limited multiple expansion.

Key Decisions for Investors

• Watch-list MH for a post-earnings long rather than chase conference commentary. Initiate only if management quantifies paid AI attach rate or lifts organic-growth guidance; target a 10%-15% upside over 6-12 months from AI ARPU plus curriculum-cycle share gains, with thesis invalidated by flat/down net pricing or margin deleverage.
• Use a 3-6 month pair: long MH / short PSO only after evidence of K-12 win-rate acceleration or California order momentum. The spread expresses MH's potential localization and blended-format advantage while reducing broad education-sector and rate-duration exposure; exit if PSO demonstrates superior AI monetization or MH loses major state adoptions.
• Set an earnings alert for AI revenue, paid conversion, cloud/content costs, renewal metrics and bookings by K-12 versus higher education. Absent these disclosures, treat management's AI pricing experiments as optionality rather than underwriting revenue.
• Avoid DUOL as a direct read-through: consumer subscription behavior and language-learning engagement are weak proxies for regulated institutional curriculum purchasing. Any broad edtech AI rally without contracted revenue evidence is more likely to benefit high-beta software narratives than MH's slower procurement model.