Pega

Their 'Next Best Action' decisioning and workflow automation commanded premium enterprise pricing because the logic was genuinely hard to build — until ChatGPT started doing business process automation from natural language descriptions. The consulting-heavy, multi-year implementations still have a structural moat, but the decisioning engine that justified the price is becoming a ChatGPT plugin.

Business process analysts, workflow designers, and the army of certified Pega developers who translate business requirements into rule configurations are staring down Claude's ability to generate decision trees and process flows from plain English requirements — no certification required.

The real moat isn't the software — it's the switching cost nightmare of ripping out years of embedded business logic from core enterprise systems. Banks and insurers have entire compliance frameworks built on Pega's case management; leaving means re-engineering every workflow simultaneously. It's ERP-level structural lock-in that buys them time, not permanence.

Their 'Pega GenAI Socrates' rollout shows they're rebuilding the platform around LLM integration rather than defending the old rules-based paradigm — smart strategy, but also an admission that their core IP advantage just became table stakes.

Enterprise workflow orchestration survives and probably grows — every company needs to automate more processes as AI capabilities expand. The question is whether they need Pega's proprietary rule engine to do it, or whether they can just describe the workflow to Claude and deploy it on standard cloud infrastructure.