The “SaaSpocalypse” and what it means for ERPs and quality assurance

Key takeaways:

• AI is reshaping software in what investors are calling the “SaaSpocalypse,” and even ERP giants like SAP and Oracle aren’t immune to the pressure.
• ERPs have a paradoxical advantage. Their complexity is impossible to replicate, making them the “system of record” future AI agents will depend on.
• As ERPs evolve into orchestration layers for AI agents, robust quality engineering becomes essential.

In February, two CNBC journalists built a replica of Monday.com’s interface using Claude Code with no prior developer experience and less than $20 in credits. Software stocks at the time were already wobbly, with investors fearing that AI could erode seat-based pricing and undercut proprietary UI. To some, the CNBC news confirmed the speculation.

Monday’s stock fell sharply, and the market uncertainty has lingered as it tries to determine whether “software” as a category is facing an existential crisis. That anxiety has inevitably extended to the king of enterprise software: ERPs.

The ERP market is dominated by legacy companies like SAP and Oracle. Claims that these systems could be “vibe coded” away are easy to dismiss as market panic, but what traders have started calling the “SaaSpocolypse” points to something real. The way that software is built, tested, and used, is being reshaped by AI, and even the most entrenched platforms, not just productivity apps like Monday.com, will need to redefine their value in this new era.

AI agents and the system of record

The CNBC journalists noted that rapid, AI-assisted coding is most threatening to productivity apps like Monday.com, which sit on top of canonical data stored elsewhere. These tools derive much of their value from polished interfaces and pre-built workflows, making them vulnerable to ever-improving AI agents that can build and navigate workflows themselves.

In contrast, legacy ERP systems have a paradoxical advantage. The very traits that make ERP platforms like SAP difficult to implement and maintain – deep customizations, proprietary data flows, and decades of accumulated historical data – are exactly what keeps these systems from being easily replicated by AI agents, even the most advanced ones. This alone has protected ERPs from much of the current market disruption.

Perhaps their greatest advantage is that ERP systems like SAP house the canonical, domain-specific data that AI agents will ultimately depend on. At SAP Sapphire 2026, SAP CEO Christian Klein called the ERP “the brain of every company,” the central data repository that maps every part of a business’s operations. This data underpins the company’s new Business AI Platform, which customers can use to build customized agents across workflows using historical business data.

Watch the webinar: The new SAP QA playbook: The key Sapphire takeaways SAP users need to know

This signals where the ERP market is headed. Rather than a web of applications users must navigate, ERPs will instead need to become an orchestration layer: a system of record that agents draw from to retrieve context, validate actions, and complete tasks. As venture capitalist Jamin Ball put it, the ERPs that will win in the future will “build amazing agentic experiences on top of boring, rock solid sources of truth.”

This doesn’t mean that ERPs are totally immune from today’s trading chaos. ERP architectures were built for stability and control, which made them reliable systems of record. But the conditions that originally made ERPs successful as a one-stop-shop for enterprise software can become a constraint in a world where AI agents are expected to move fluidly across tools – pulling data from Salesforce, triggering workflows in SAP, updating records in Workday.

The challenge ahead is not whether ERPs will exist in the future — they most certainly will — but whether they can shift from proprietary systems of record into flexible control planes for agents.

What this means for testing ERPs

The SaaSpocolypse is not an end-of-life event for most SaaS companies. It is an active, still-evolving response to how AI is changing all parts of software, including software testing.

As ERPs evolve from business applications to orchestration layers for agents, quality engineering will undergo a similar shift. Software testing will move from fragmented, manual activities like test creation, execution, and maintenance into a unified lifecycle powered by coordinated quality agents. Like business users working through SAP’s new platform, testers will work through orchestration hubs that define and manage QE agents for various quality tasks.

Read the whitepaper: Why AI is the future of effective SAP releases

This will make quality work significantly more efficient, but it will also be essential to the future of ERPs. The success of initiatives like SAP’s Business AI Platform and the agents it oversees depends on the integrity of underlying enterprise data. Software testing must have its own agentic foundation to properly validate these new ERP agents and the data they are built on. Without robust, AI-enabled QE, future ERP systems like SAP risk becoming more autonomous but less trustworthy. This is why software testing is foundational to any ERP evolution emerging from the SaaSpocolypse: QE will help ERPs safely transform into something new and better.

This is not a hypothetical future. At SAP Sapphire 2026, Tricentis showcased its role as a quality layer inside SAP’s own Agent-led Toolchain, with capabilities purpose-built to generate, validate, and maintain tests across SAP environments as they evolve. When the business logic and the quality foundation are both in place, what enterprises can build with agentic AI, and how fast they can ship it, becomes a much more interesting question.