The dot-com crash hit months after launch, enterprise buyers froze budgets, and "internet software" became a punchline in boardrooms. The question now is whether its massive bet on AI agents — Agentforce ARR grew 169% last quarter — can reignite growth or whether it'll expose the uncomfortable truth that much of CRM is glorified record-keeping. The land-and-expand math is relentless. If it doesn't, the seat-based model faces structural pressure from the very AI tools Salesforce is building. Strategic direction: Salesforce is focusing on profitable growth, Data Cloud, AI agents, automation, industry clouds, and cross-sell across its CRM portfolio. That's the pitch, and it's landing more often than Salesforce's investor presentations acknowledge. These companies grow. Salesforce's traditional pipeline of companies outgrowing simpler tools is narrowing. Investors decided they'd rather have margins than growth, and Salesforce obliged. Not cheap for a 10% grower, but not absurd given the cash flow profile and the optionality around AI monetization. The AI cannibalization question is the one that keeps the strategy team up at night. Elliott Management and Starboard Value forced margin discipline in 2023, which investors loved (stock up 90%+ since). Salesforce's growth story has narrowed to one question: can Agentforce become a $5-10 billion product line by FY2030? Agentforce is the only thing that could reaccelerate growth to 15%+ and justify the current valuation multiple. The cross-sell math remains the quiet growth engine. FY2027 guidance: $45.8-46.2 billion (10-11% growth). Growth stays at 10%, the seat-based model slowly erodes as automation reduces headcount, and Salesforce settles into the profile of a high-margin, low-growth infrastructure company trading at 5x revenue. Below that, the stock becomes a yield play, not a growth story. Parker Harris, Dave Moellenhoff, and Frank Dominguez — the three engineers Marc Benioff recruited to build his impossible idea — ran extension cords across the living room floor and coded on folding tables. Larry Ellison had been his mentor, his champion, even an early investor in the new venture. By 2003, Salesforce had enough traction to launch Dreamforce — initially a modest customer event that would eventually become the largest software conference in the world, drawing 170,000+ attendees. Anyone could build applications on top of Salesforce's infrastructure and sell them to Salesforce's customers. Suddenly, administrators, consultants, developers, and implementation partners had financial incentives to promote Salesforce adoption. That DNA still drives decisions today — including the bet on Agentforce, which is essentially the same argument Benioff made in 1999 applied to AI: what if it just worked, without requiring companies to build the infrastructure themselves?