A per-seat price assumes a person sits behind the software. Once an AI agent handles the support ticket, nobody adds a seat, and the vendor’s revenue model stops matching the work being done. That mismatch explains why outcome-based pricing went from conference talk to published price lists in the past year. The shift is real, though it is messier than the headlines suggest.
Why seats stop making sense for agents
Seat pricing worked because value scaled with headcount: more users, more usage, more revenue. An agent breaks that link. A support team of ten can send thousands of conversations through an AI agent without hiring anyone, so a per-user fee either overcharges a small team or undercharges a busy one.
Venture firm Bessemer put it bluntly in its February 2026 AI pricing and monetization playbook: AI-native companies are abandoning seat-based SaaS pricing in favor of usage-, output- and outcome-based models. The same piece points to a cost reason too. AI companies see gross margins of 50-60%, against 80-90% for traditional SaaS, so pricing that ignores inference cost is risky for the vendor.
What outcome-based pricing looks like on a price list
Customer service is where the model is most visible, because a resolved ticket is easy to count. Three vendors show three slightly different takes:
- Intercom Fin charges $0.99 per outcome, with $9.99 for qualification outcomes and a 50-outcome monthly minimum when used with a non-Intercom helpdesk, according to the Fin pricing page.
- Zendesk bills $1.50 per committed automated resolution or $2.00 pay-as-you-go, and says you pay only for requests the AI agent resolved. Plans include a monthly allowance per agent, per its pricing page.
- HubSpot cut its Customer Agent from $1.00 per conversation to $0.50 per resolved conversation, and moved its Prospecting Agent to $1.00 per lead recommended, as HubSpot announced.
Notice that the unit differs: an outcome, a resolution, a recommended lead. There is no standard yet.
Most vendors are blending, not switching
If you read only the announcements, you would think seats are finished. The big platforms say otherwise. Salesforce sells Agentforce through three models at once, as SaaStr lays out: $2 per conversation, Flex Credits at $0.10 per action, and per-user licenses starting at $125 per user per month. SaaStr reports that buyers found the per-conversation model hard to forecast, and that Salesforce later framed the seat as "digital labor."
Bessemer’s playbook lands in the same place for younger companies. It describes hybrid pricing, a base subscription plus usage or outcome tiers, as the effective middle ground. My read is that pure outcome pricing suits narrow, countable jobs like tickets resolved. Everything broader tends to get a platform fee and a usage meter.
The catch: who defines an outcome?
Per-resolution pricing sounds fair until you ask what counts as a resolution. Fin’s page defines it as a conversation where no further help is requested after the agent’s last answer. HubSpot’s announcement says customers pay when agents complete their assigned task, but does not spell out a formal definition of resolved. Third-party guides such as HelpDesk point out that a customer who simply stops replying can still be billed, so the effective cost per truly solved problem may sit above the headline rate.
That also makes headline comparisons shaky. A $0.50 resolution and a $2.00 resolution are not the same product if each vendor counts differently.
What to check before you sign
- Ask for the written definition of a billable outcome, including what happens when a customer goes silent or gets escalated to a human.
- Model your bill at low, expected and peak volume. Outcome pricing trades seat predictability for a variable invoice.
- Look for minimums, included allowances and credit expiry. Several of the plans above bundle allowances or monthly minimums.
- Keep a seat-based or capped option on the table for work that is hard to count.
Where this leaves buyers
The useful question is no longer whether to pay per seat or per outcome. It is which parts of your stack have a countable result worth paying for, and which are just tools people use. Start with the first group, negotiate the definition before the rate, and expect vendor price lists to keep changing.
Planning an AI rollout and unsure how to budget it? Explore more on Nearsmarter or talk to us about structuring the build.