Part 2: The 25-Year Cycle of ERP

Historical Context

Enterprise Resource Planning has existed for decades, but it has gone through distinct cycles. Initially, software-based systems like SAP and Oracle dominated, requiring substantial on-premise infrastructure and extensive customization. They brought order and standardization to complex global operations, but at the cost of rigidity and high upkeep. In the late 1990s and early 2000s, companies such as NetSuite introduced cloud-based ERP, offering faster deployment and lower capital expenditure. This new model democratized ERP by making it accessible to smaller businesses, yet it still came with challenges like subscription costs and vendor lock-in.

What’s Next: AI-Based ERP

Now, we’re on the cusp of a third major wave—AI-based ERP. Rather than treat AI as a peripheral feature, forward-thinking businesses are starting to integrate machine learning and predictive analytics into core enterprise functions. From procurement to forecasting, AI algorithms can preemptively identify bottlenecks, optimize workflows, and reduce the burden of repetitive tasks. This evolution won’t merely enhance existing systems; it will redefine the fundamental architecture of ERP to prioritize real-time decision-making, adaptability, and continuous learning.

Building the System of Decision and Operation

To support these agent-driven processes, tomorrow’s ERP must be more than a static system of record. It needs to act as a dynamic system of decisions and operations. Here’s where domain-specific knowledge and company-specific context become critical. Agents might leverage large generic models—GPT, LLaMA, or others—for broad, foundational knowledge. But to excel, they need access to a proprietary, ever-evolving dataset capturing a company’s unique operational footprint. This insight is precisely what shaped Vertical Bar’s mission: collecting operational data not just to visualize the past, but to power the AI that will shape a company’s future.