Tech Graveyard/business-models
Per-Seat SaaS Pricing (1999 to Twitching)
Per-seat pricing made SaaS work: one login, one monthly fee. It is still the default on almost every price list. But when an AI agent does the work, there is no seat to charge for, and Intercom, Salesforce and Zendesk have started pricing the outcome instead.
Born 1999 · Still dying, projected end 2035 · Status: twitching
By Deepak Gupta · Published
Paired prediction: what replaces it in Future Tech
Certificate of Death
Name of decedent
Per-Seat SaaS Pricing
- Born
- 1999
- Died
- Not yet
- Age
- 27+
Cause of death
AI agents that do work without a login, priced per resolution, per conversation or per outcome instead of per user
Survived by
Outcome-based pricing, per-conversation and per-action credits, and hybrid plans that still sell seats alongside them
Invented by
No single inventor. It became the SaaS default; Salesforce, incorporated in 1999, still lists Sales Cloud prices per user per month
The hook
Salesforce still lists Sales Cloud at a price per user per month. It also sells Agentforce at $2 per conversation. The same company now prices software two ways, and only one of them assumes a person is sitting in the seat.
Thesis. Per-seat pricing is not dying yet. It is twitching. The model assumed every unit of value had a human login attached, and AI agents break that assumption. The first vendors to price agents per outcome are not abandoning the seat. They are hedging against the day it stops tracking value.
The story
Born with SaaS, 1999
Salesforce incorporated on March 8, 1999, and launched on February 7, 2000 at an event themed around "The End of Software". The name of the event was the pitch: software you sign into, not software you install.
Nobody can be credited with inventing the seat. It became the SaaS default because it was easy to explain: one login, one fee, every month. It was simple to buy, simple to forecast, and the vendor's revenue grew every time the customer hired.
The peak: headcount as a proxy for value
The seat works when value tracks people. A support team of 50 agents gets roughly twice the value of a team of 25, so charging per agent seems fair. It is still the default unit today: Salesforce's Sales Cloud editions are listed in USD per user per month, and Intercom's plans are priced per seat per month.
The seat also had the filings to itself. In a count of 750 US tech IPO final prospectuses filed 2004 to 2026 by guptadeepak.com's Rewired Metrics research (preliminary), no prospectus before 2025 described outcome-based pricing or the company's own AI agents. That makes 2024 the last year of the seat's monopoly, even as the first vendors began pricing agents differently.
The first twitches, 2024 to 2026
On August 28, 2024, Zendesk announced outcome-based pricing for its AI agents, saying customers would only incur costs for issues resolved autonomously by AI. On October 29, 2024, Salesforce made Agentforce Service Agent generally available, with pricing starting at $2 per conversation. In May 2025 it added Flex Credits at $500 per 100,000 credits, where each action consumes 20 credits, or $0.10 per action.
Intercom prices its Fin AI agent at $0.99 per outcome and charges at most one outcome per conversation. A resolution counts when no further help is requested after Fin's last answer. Its standalone Fin plan, sold for use with another helpdesk, says no seats are required.
On November 10, 2026, Zendesk went further. It introduced Zendesk Unlimited, described as moving away from traditional seat-based software consumption models, with new capabilities offered with unlimited seats. Customers set an Outcome Budget and pay for eligible issues resolved and business work completed.
Why the seat breaks
An AI agent that resolves a support ticket has no login, no salary and no seat. If a vendor's product resolves more tickets automatically, a customer needs fewer human agents, and a per-seat vendor earns less for doing a better job. Pricing the resolution instead lets the vendor's revenue grow with the work the software does, not with the customer's headcount.
The same count shows how early this is. Outcome-based pricing or the company's own AI agents appeared in 1 of 12 US software IPO prospectuses in 2025 and 2 of 16 in 2026. Usage or consumption wording is older and more common: 7 of 24 in 2020, 12 of 69 in 2021, and 5 of 16 in 2026, about 85% of them describing the company's own pricing.
Why it is twitching, not dying
Every vendor in this story still sells seats. Intercom's pricing FAQ describes two components: seats, paid per teammate, and usage, such as Fin outcomes. Salesforce's Agentforce pricing page lists per-user licenses next to the per-conversation and per-action options, and its May 2025 announcement offered employee-facing agents in a per user per month model. In that same release, a 451 Research analyst noted that the majority of businesses still prefer to buy technology on a license model, while the desire for outcome-aligned pricing is rising.
So the seat has not lost. It has lost its monopoly. Buyers still like a number they can budget. What has changed is that the seat now sits on the price list next to a meter that counts work, and the meter is where the new products go first. My projection is a long decline, not a cliff: seats for the humans who stay, outcomes for the work agents take over.
Key data points
- Salesforce incorporated March 8, 1999 and launched February 7, 2000 around "The End of Software"
- Zendesk announced outcome-based pricing for AI agents on August 28, 2024
- Salesforce Agentforce Service Agent, generally available October 29, 2024: from $2 per conversation
- Salesforce Flex Credits (May 2025): $500 per 100,000 credits, 20 credits ($0.10) per action
- Intercom Fin: $0.99 per outcome, at most one outcome per conversation
- Zendesk Unlimited (November 10, 2026): unlimited seats, customers pay against an Outcome Budget
- US software IPO prospectuses describing outcome-based pricing or own AI agents: 0 before 2025, 1 of 12 in 2025, 2 of 16 in 2026 (guptadeepak.com Rewired Metrics count, preliminary)
Contrarian angle
Outcome pricing is sold as fairer to buyers: pay only for what works. But the vendor defines what counts as an outcome. Intercom counts a resolution when no further help is requested after its agent's last answer, which a customer who simply gives up could also satisfy. A seat was a crude proxy for value, but both sides could count seats. The harder part of replacing the seat is not the price. It is agreeing who measures the result.
The flip side
What replaces it
The paired prediction in Future Tech.
Read the predictionFAQ
Is per-seat SaaS pricing dead?
No. It is still the default on most price lists, including Salesforce's Sales Cloud and Intercom's plans. What has changed is that AI agent products are increasingly priced per outcome, per conversation or per action alongside seats.
What is outcome-based pricing?
Pricing tied to a result instead of a user. Intercom charges $0.99 per Fin outcome, and Zendesk says customers only incur costs for issues its AI resolves autonomously. The vendor is paid when the work is done, not for access.
Why do AI agents threaten per-seat pricing?
An agent does work without a human login. If software resolves more tickets, the customer may need fewer human seats, so a per-seat vendor would earn less as its product improves. Pricing the outcome keeps revenue tied to the work.
Are buyers moving away from seats?
Slowly. In Salesforce's May 2025 announcement, a 451 Research analyst said the majority of businesses still prefer to buy technology on a license model, while interest in outcome-aligned pricing is rising.
What replaces per-seat pricing?
Most likely a hybrid. Salesforce sells per-user licenses next to per-conversation and per-action pricing. Zendesk's November 2026 plans offer unlimited seats with spend set by an Outcome Budget.
More from guptadeepak.com
Further reading
Related obituaries
More from the business-models graveyard.