Cursor was criticized for negative gross margins, but its hypergrowth in the massive AI coding market led to a $60B acquisition. In a risk-on environment, market leadership and growth potential can outweigh fundamental financial weaknesses, especially when the acquirer can solve the core problem.
The founder of Cursor likely preferred working for Elon Musk over Mark Zuckerberg, despite Meta also being a logical buyer. This "founder-acquirer fit" and the emotional appeal of landing in a desirable environment can be a critical, often underestimated, factor in high-stakes M&A, especially for founders not purely cash-motivated.
The acquisition of a market leader doesn't spark an immediate land rush. Instead, it forces potential buyers who lost out or were waiting to act, pushing them to acquire the #2 or #3 player because their primary target is gone and they realize their timeline has been accelerated.
Stripe's acquisition of OpenRouter highlights a paradox in large M&A. While a target's revenue is key for valuation, it's ultimately irrelevant to the acquirer. The real value lies in how the acquirer can leverage the asset to create a much larger revenue stream, often abandoning the original business model.
The value of AI model routers is challenged by enterprise behavior. Similar to "multi-cloud," companies find it impractical to manage dozens of models due to "model drift" and QA costs. They prefer to standardize on 2-3 qualified models, limiting the market for broad routing platforms.
When evaluating a hypergrowth company like Anthropic, the market will likely ignore massive off-balance-sheet compute commitments and unprecedented stock-based compensation (SBC). These negative financial indicators are deemed irrelevant as long as top-line revenue growth is explosive. The free pass is revoked the moment growth slows.
Projections of AI reaching trillion-dollar revenues based on a "billion knowledge workers" are flawed. A more realistic model is to take the US software budget as 50% of the world's total, as the rest of the world cannot afford the same software per capita. This grounds market sizing in actual spending power, not population.
The pace of AI-driven development has dramatically accelerated. High-performing companies are no longer planning quarterly; they have completed their 2024-2026 roadmaps and are now deep into 2027 planning. This sets a new, aggressive benchmark for what constitutes a top-tier engineering organization.
Silver Lake's bid for Workday exemplifies the mature SaaS endgame. Unlike venture capital, where market fit can forgive a high valuation, private equity returns depend on precise financial modeling. A 20% overpayment on price can destroy the IRR, making it a game of financial engineering, not growth speculation.
A closed ecosystem like Workday is a more attractive private equity target than an open one like Salesforce. Its limited interoperability makes it harder for third-party AI agents to extract value from its data, providing a stronger moat against disruption and making its revenue streams more predictable for an LBO model.
While being a system of record creates high switching costs and ensures retention, it doesn't translate to growth. "Prisoner" customers with no easy alternative are more likely to seek cost reductions from their vendor rather than increase their spend. Growth requires delivering new value, not just leveraging lock-in.
Initially, AI products like Lovable had minimal defensibility. Their moats were built not on a single technological breakthrough but by continuously adding features and complexity at a pace competitors couldn't match. In software, the most durable moat is often just being "faster and better" over a sustained period.