The SaaSpocalypse That Wasn't: Salesforce, Booking, IBM Thrive with AI
Source: Fortune. Casualplayhub News adds summary, context, and editorial framing while linking back to the original report.
The famous quip attributed to Mark Twain—"the rumors of my death are greatly exaggerated"—aptly describes the current state of several software companies amid dire predictions of AI-driven disruption. Over the past year, roughly $2 trillion in software market capitalization has evaporated as investors feared that AI would make many software businesses obsolete. This wave of panic, dubbed the "SaaSpocalypse," has been premature and misguided. While some high-flying tech firms may indeed face threats from autonomous AI agents, the market has indiscriminately punished companies that are actually positioned to be AI winners.
Consider Salesforce. The leading customer relationship management platform was wrongly predicted to be replaced by large language models from OpenAI and Anthropic. Critics imagined that AI agents would manage customer relationships end-to-end, reducing Salesforce to a passive database. But that analysis reads the situation backwards. In the AI era, data is the new moat, and Salesforce has the data. As Wells Fargo analysts noted, "lower cost of intelligence increases value of incumbent data." AI agents need clean, unified, trusted data to operate effectively, and Salesforce holds that data for virtually every major company. This year alone, Salesforce processed over 216 trillion customer records. Its AI agent platform, Agentforce, grew from $100 million to $1.5 billion in annual recurring revenue in 18 months, with more than 30,000 deals closed. The acquisition of Slack, once underestimated, has become critical, as Slack provides human context that AI agents need. Salesforce also bought back $25 billion of its own stock this year, demonstrating confidence in its AI-driven future.
Booking Holdings tells a similar story. Earlier this year, some analysts argued that if travelers can ask a chatbot for a hotel, why would they need Booking.com? But that view mistakes Booking for a search engine rather than a differentiated travel transaction platform. Nearly three-quarters of its bookings are processed as merchant of record, handling payments across 100 methods and 50 currencies, plus managing complex disputes and cancellations that AI platforms have shown no interest in handling. Google itself has stated it has "no intention of becoming an OTA," and OpenAI retreated from in-chat checkout after a botched rollout. Booking’s competitive edge is reinforced by the fact that 90% of its room nights come from independent properties that lack the infrastructure for global payment processing. AI-driven traffic to Booking remains below 1% of room nights, but direct traffic holds steady in the mid-60% range. As AI increasingly drives discovery, Booking is best positioned to convert that traffic into loyal customers, thanks to its decades of experience in bidding for search traffic.
IBM presents a different but equally compelling case. When IBM stock fell 25% in a single day last month, bears thought they had found a winner. The drop was triggered by several large clients redirecting capital budgets toward memory amid a severe crunch. Yet a third of those supposedly lost deals closed within weeks. The bearish narrative that AI would disrupt IBM’s $21 billion consulting business and its legacy software ignores the reality: AI now accounts for half of all new consulting signings, at higher margins because IBM can bill for outcomes rather than hours. Red Hat, the software that enables AI agents to run across any cloud, grew 11%. IBM’s AI business has more than doubled over the past year. The company is being paid to build the AI transition, not run over by it.
In all three cases, the market has failed to recognize that these companies own assets that AI agents cannot operate without: Salesforce owns the data, Booking owns the transaction infrastructure, and IBM owns the underlying technology. Each of these moats becomes more valuable in an AI world. The SaaSpocalypse was a fiction, and the panic has created opportunities for those who see the real winners.
Article commentary
The market’s reaction to the rise of artificial intelligence has been marked by extreme volatility and a tendency toward binary thinking. The "SaaSpocalypse" narrative, while catchy, oversimplified a complex landscape. In reality, AI is not a monolithic force that will either destroy or elevate every software company; its impact depends on the specific assets and business models of each firm. The three examples highlighted—Salesforce, Booking Holdings, and IBM—illustrate how companies with strong data moats, transaction infrastructure, or platform dependencies can leverage AI to reinforce their competitive positions. A key insight from this analysis is that the value of data increases as AI lowers the cost of intelligence. This is a fundamental shift: instead of commoditizing data holders, AI makes their data more valuable because AI agents require high-quality, proprietary data to function effectively. Salesforce’s vast repository of customer records, Booking’s network of independent properties and payment systems, and IBM’s legacy enterprise software and consulting expertise all become more indispensable in an AI-driven world. The market’s initial panic overlooked this dynamic, focusing instead on the threat of AI agents replacing human tasks. Another critical factor is the distinction between discovery and transaction. Many investors conflate the top of the funnel, where AI can indeed disrupt search and recommendation, with the bottom of the funnel, where complex transactions, payment processing, and dispute resolution occur. Booking Holdings thrives because it owns the transaction layer, which AI platforms have shown little appetite or ability to handle. Similarly, IBM’s consulting business benefits from the need for human expertise to implement AI systems, not just the algorithms themselves. However, not all software companies will be winners. The article itself acknowledges that many high-flying tech firms face genuine threats. The key for investors is to differentiate between firms that own irreplaceable assets and those that are merely intermediaries. The market’s indiscriminate sell-off created opportunities for those who could see past the panic. The commentary also serves as a cautionary tale about the danger of simplistic narratives in financial markets. The fear of AI-driven obsolescence may have been overblown, but the need for careful analysis of business fundamentals remains as important as ever. Ultimately, the so-called SaaSpocalypse was a false alarm. The real story is the transformation of incumbent software companies into AI beneficiaries, a shift that will likely continue as AI adoption accelerates. This perspective offers a more nuanced and realistic view of the AI revolution, one that rewards patience and discernment over panic.