SaaS-pocalypse isn't coming any time soon • The Register

The recent market turbulence sparked by a single blog post from Citrini Research has reignited debates about the future of software-as-a-service. The firm's dramatic prediction of a "SaaS-pocalypse" by June 2028, complete with 10 percent unemployment and Salesforce alumni driving for Uber, captured headlines and briefly rattled investors. But as the dust settles, industry leaders and analysts are pushing back against the doomsday narrative, arguing that the fundamentals of enterprise software remain remarkably resilient.

The core of the SaaS-pocalypse theory rests on the assumption that large language models and AI automation will dramatically lower the barriers to building and maintaining enterprise applications. If anyone can spin up a custom solution, the logic goes, why pay premium prices to established vendors? The Citrini post envisions a future where AI-driven development tools enable a flood of new competitors, forcing incumbents into a "race to the bottom" on pricing while differentiation evaporates.

However, this scenario overlooks several critical factors that have historically insulated enterprise software from disruption. First, there's the issue of data gravity. Oracle, Salesforce, SAP, and their peers have spent decades accumulating vast troves of customer data within their platforms. This isn't just about storage—it's about the complex web of integrations, workflows, and business processes that have evolved around these systems. Moving that data, retraining staff, and rebuilding those processes represents a monumental undertaking that few organizations undertake lightly.

Sridhar Ramaswamy, CEO of Snowflake, offered a more grounded perspective during recent investor discussions. "It's useful to step back and look at the impact that AI as a whole is having on software," he noted. "We spend a lot of time looking at this. We live this—and our take is that overall, the winners are going to be the companies that provide that single source of enterprise truth."

This emphasis on "single source of truth" gets to the heart of why enterprise software incumbents maintain their advantage. In complex organizations, data consistency isn't a luxury—it's a necessity. When financial reporting, supply chain management, and customer relationship data all live in different systems with conflicting information, decision-making becomes chaotic. AI models, no matter how sophisticated, cannot compensate for fundamental data governance problems.

Ramaswamy's point about built-in security, auditability, and governance resonates particularly strongly in an era of increasing regulatory scrutiny. Enterprise software isn't just about functionality; it's about compliance, risk management, and operational reliability. These requirements create natural moats around established vendors who have invested heavily in meeting industry standards and regulatory frameworks.

The inertia argument extends beyond technical considerations. Enterprise organizations are inherently conservative when it comes to mission-critical systems. The cost of failure—whether measured in lost revenue, regulatory penalties, or reputational damage—far outweighs the potential benefits of being an early adopter. This risk aversion creates a natural buffer against rapid disruption, even when new technologies promise significant advantages.

Gartner's analysis supports this more measured view. The research firm suggests that in the short to medium term, organizations will continue to deploy AI agents within their existing application environments rather than replacing those environments entirely. This approach allows companies to capture AI's benefits while minimizing disruption to established processes and data architectures.

There's also the question of what AI actually enables versus what it promises. While large language models have demonstrated impressive capabilities in code generation and automation, building enterprise-grade applications involves far more than writing code. It requires understanding complex business requirements, ensuring regulatory compliance, integrating with legacy systems, and providing ongoing support and maintenance. These challenges haven't disappeared simply because AI can generate Python scripts more efficiently.

The valuation concerns raised by the SaaS-pocalypse theory do warrant attention, however. Many pure-play SaaS companies have enjoyed extraordinary growth rates that may prove unsustainable as markets mature and competition intensifies. The law of large numbers suggests that maintaining triple-digit growth becomes increasingly difficult as companies scale, and the influx of new vendors will certainly put pressure on margins.

But this is a far cry from predicting the collapse of the entire SaaS model. What we're more likely to see is a market correction and consolidation, similar to what occurred in other technology sectors as they matured. Strong players with defensible positions—whether through data advantages, regulatory compliance, or deep industry expertise—will continue to thrive, while weaker competitors may struggle or be acquired.

The reality is that enterprise IT has always been, and likely will remain, a field where reliability and trust trump novelty. Organizations need systems that work consistently, protect sensitive data, and support critical business processes. The people responsible for maintaining these systems—the ones who ensure data coherence, manage governance frameworks, and keep operations running smoothly—aren't going anywhere.

In fact, their role may become even more critical as AI tools become more prevalent. Someone needs to validate AI-generated code, ensure compliance with regulations, manage the integration of new technologies with legacy systems, and maintain the human oversight that prevents automation from creating new classes of errors. Far from eliminating IT jobs, the AI revolution may shift their focus toward higher-value activities while maintaining the essential boring work that keeps enterprises functioning.

The market's overreaction to the Citrini post reveals more about investor psychology than it does about technological reality. In an environment where a single blog post can move billions in market capitalization, it's perhaps unsurprising that dramatic predictions gain traction. But for the people actually building and maintaining enterprise systems, the day-to-day challenges remain largely unchanged.

Data still needs to be accurate. Systems still need to be secure. Processes still need to be reliable. And someone still needs to do the often unglamorous work of making sure it all functions correctly. That work may evolve with new tools and technologies, but it's unlikely to disappear entirely.

The SaaS-pocalypse, if it comes at all, will be a slow evolution rather than a sudden collapse. Established vendors will adapt, incorporating AI capabilities into their platforms while leveraging their existing advantages in data, compliance, and customer relationships. New entrants will find niches but face significant barriers to displacing entrenched players in core enterprise functions.

For now, boring might indeed be the best way to be in enterprise IT. The systems that quietly and reliably support critical business operations will continue to generate substantial value, even if they don't make headlines or spark market frenzies. And that's probably a good thing for the organizations that depend on them.