AI Is Ending Per-Seat Software Licensing

The era of per-seat software licensing - the model that has defined enterprise software for decades - is nearing its end as agentic AI reshapes how applications are built, delivered, and monetized. A new McKinsey analysis suggests the software industry is entering a “post-SaaS” era, where AI agents will interact directly with each other, driving the adoption of consumption-based and outcome-based pricing models over traditional licenses. 

According to McKinsey, 63% of software vendors believe AI will fundamentally transform their business model within the next three to five years, and 40% expect AI to unlock at least 20% additional revenue growth. As agentic architectures evolve faster than previous shifts such as client-server or cloud computing, companies that fail to adapt could face steep competitiveness challenges. 

Key trends shaping the shift include: 

• Outcome-based pricing: Vendors will increasingly charge for measurable results - such as productivity gains or successful transactions - instead of user seats. 

• Agent-to-agent transactions: AI systems will buy, sell, and execute services autonomously, potentially reducing seat counts by up to 70%, according to Partner Economics. 

• Customer churn and vendor switching: As AI reshapes delivery models, organizations will gain more flexibility but face risks of opaque pricing and locked “outcome bundles.” 

• Ethical and governance concerns: As AI agents act on users’ behalf, human oversight and transparency will become vital to ensure compliance, reliability, and trust. 

While pioneers like Salesforce and Intercom already monetize AI through usage-based models, experts warn that the supporting infrastructure for full agent-driven ecosystems remains immature. Still, McKinsey and industry leaders agree: the future of software lies in selling outcomes, not access. 

Source: 

https://www.zdnet.com/article/so-long-saas-why-ai-spells-end-of-per-seat-software-licenses-and-what-comes-next/