← Back to concept explainers

Concept Fable

Seat-based pricing

Conclusion first:Seat-based pricing charges by user, account, or headcount, and is the most common and easiest-to-understand procurement model in SaaS. It suits products where value mainly comes from people using the software and user count is roughly correlated with value and cost. But in enterprise AI / Agent services, an Agent may let a few people trigger a large amount of automated work; costs come from tokens, tasks, tool calls, and human escalation, while value comes from completed outcomes. In this case, seats can remain as basic access rights or administrative boundaries, but usage, service, or outcome pricing often needs to be layered on top.

Seat-Based Pricing, Per-Seat Pricing · Business model / pricing

Read the story first. There is no rush to give a definition here; let the question grow on its own first.

Fable

The academy charged lamp oil fees by seat, until the copying machine started doing the work itself

Songshi Academy had a night reading hall, and in the past it charged for lamp oil by seat. The more people came to study, the more lamp oil, tables, chairs, paper, and brushes were used, so charging by seat was simple and broadly fair.

Later the academy bought an automatic book-copying machine. One scholar could sit in the hall and have the machine copy ten boxes of books overnight; another student sat for only half an hour to check one page of annotations. Both occupied one seat, but their consumption and value were completely different.

At first, the steward still insisted on charging by seat. He felt procurement could understand it most easily, and the bill was easy to discuss. But large customers kept adding copying tasks, small customers complained that they were subsidizing others, and paper, ink, and maintenance costs rose rapidly.

Someone proposed a quota for each seat. Each person could copy only five books per day. As a result, heavy users felt the academy could not be used for real work, light users still found it expensive, and the manager was merely using restrictions to cover up the fact that the pricing unit had already become outdated.

A bookkeeping review found that the relationship between seat count and cost had broken down. What really ate money was pages copied, proofreading passes, rush jobs and manual error fixes; what really created value for clients was also this completed work.

The academy did not immediately abolish the seat fee. It kept the basic seat as access rights and a management boundary, but billed separately for large amounts of copying, rush orders, proofreading and dedicated organizing.

This change made sales more honest. For scenarios where people sit down and use it, seats remained clear; for scenarios where machines work on people's behalf, billing had to be close to the task, usage, or outcome. The manager also made a loss chart. Being fully seated did not necessarily mean profit, and empty seats did not necessarily mean no value; the key was how much copying, proofreading, and firefighting each seat triggered behind it. The old price list hid these real costs. He also showed the bill to customers. People who only came to read should not pay for large amounts of copying, and those who truly made machines work all night could not pay only for a lamp's worth of oil. Only when pricing units are close to value and cost will both sides avoid resenting each other after successful use. The academy also kept a simple package to make it easy for small customers to enter; but truly heavy work had to be billed separately, otherwise the larger the scale, the more damaging it became. The manager said, if the billing criteria are wrong, good customers will also turn into bad debt.

The manager understood that a seat can be an entry point, but it cannot pretend to represent all value. Otherwise, the more successfully customers use it, the more likely the academy is to lose money.

Reveal

This story is about: seat pricing

Seat-based pricing charges by user, account, or headcount, and is the most common and easiest-to-understand procurement model in SaaS. It suits products where value mainly comes from people using the software and user count is roughly correlated with value and cost. But in enterprise AI / Agent services, an Agent may let a few people trigger a large amount of automated work; costs come from tokens, tasks, tool calls, and human escalation, while value comes from completed outcomes. In this case, seats can remain as basic access rights or administrative boundaries, but usage, service, or outcome pricing often needs to be layered on top.

What matters is that it is not just a term, but a structural problem that repeatedly appears in real AI / Agent systems. Understanding it lets you judge when to add a model, when to change a process, and when to add governance.

Metaphor mapping

  • Night Reading Hall seats: traditional SaaS user seats
  • The automatic scribe: an Agent that can execute work on people's behalf
  • One person copying ten boxes of books overnight: a small number of seats generates a large amount of consumption and value
  • Per-seat limits: using restrictions to remedy pricing-unit mismatch
  • Paper and ink, repairs, proofreading: tokens, tool calls, human fallback and operations costs
  • Keep the base seat: seat as access right, permission and procurement anchor
  • Copying, rush handling, and proofreading billed separately: stack usage, SLA, or service fees
  • Machines doing the work: the fundamental reason seat pricing boundaries are narrowing in the AI era

Andy Ren Xu's judgment

The practical value of this concept is that it helps you break "AI features that look smart" into deliverable, verifiable, and governable units of work.

SaaS pricingsubscriptionnamed seatactive user pricingenterprise contract

Related distinctions

This concept is easily confused with adjacent concepts; it is recommended to continue to the comparison page.

京ICP备2026014272号