When AI agents do the work of ten employees, paying per seat makes no sense. The software industry is scrambling to find new pricing models before revenue collapses.
The per-seat pricing model has been the economic engine of the software industry for over two decades. From Salesforce's early days charging per user per month to the modern SaaS playbook of "land and expand," the assumption has been straightforward: as companies hire more people, they buy more software seats. Revenue scales with headcount. In 2026, that assumption is breaking. AI agents are performing the work of multiple employees, and organisations are discovering they need far fewer software seats to accomplish the same, or greater, output.
The Math That Terrifies SaaS CEOs
Consider a sales team of 100 representatives, each with a Salesforce licence at roughly $300 per month. That is $360,000 in annual recurring revenue for Salesforce from a single customer. Now imagine that customer deploys AI agents that handle lead qualification, data entry, follow-up scheduling, and CRM updates, allowing the same pipeline to be managed by 20 human sellers augmented by AI. The seat count drops from 100 to 20. Revenue drops from $360,000 to $72,000 per year, an 80% reduction for the same business outcome. Multiply this across thousands of enterprise customers, and you understand why investors are panicking.
The New Pricing Models Emerging
Software companies are experimenting with several alternatives. Usage-based pricing charges for API calls, transactions processed, or compute consumed rather than seats occupied. Salesforce has pioneered an Agentic Enterprise Licence Agreement offering fixed-price access to its Agentforce AI platform. ServiceNow is moving to consumption-based and value-based pricing for AI agent offerings. Outcome-based pricing, where the vendor shares in the measurable business value their software creates, is emerging as the most radical alternative. Gartner predicts that by 2030, at least 40% of enterprise SaaS spend will have shifted to these new models.
Winners and Losers in the Transition
Companies whose products become infrastructure that AI agents use will thrive. Data platforms, API services, identity providers, and core systems of record are becoming more valuable as the number of AI agents accessing them grows. Companies whose products replicate work that AI agents can perform natively, such as basic workflow automation, scheduling tools, and simple data entry applications, face existential risk.
What This Means for Buyers
For enterprise buyers, this transition is overwhelmingly positive. The cost of software as a share of revenue should decline as per-seat models give way to usage-based alternatives. Organisations can reallocate budget from software seats to AI infrastructure, investing in agentic AI platforms that deliver greater automation at lower per-unit costs. The key is to start renegotiating contracts now rather than waiting for renewal cycles to catch up with the new reality.
At QverLabs, we designed our compliance platform from the start around outcome-based value rather than seat counts. Our clients pay for compliance coverage, not for how many humans log into a dashboard. As AI agents take on more of the execution work, this model scales naturally: more automation means more value delivered, not more seats sold. This alignment between vendor incentives and customer outcomes is what the next generation of software pricing should look like.
