A company buys software for ten employees: ten seats at thirty dollars each. Then it brings in one AI agent that does the work of five of them.
Next year, it needs five seats. The vendor’s software is doing more work than ever, and earning half as much.
The Arithmetic of the Squeeze
This publication’s own arithmetic: ten seats at thirty dollars is three hundred dollars, and five seats at thirty dollars is a hundred and fifty. The video gives no billing period for the thirty dollars.
The seat count fell by half, so the bill fell by half, even though the software now does more work.
At thirty dollars a seat, this publication’s own arithmetic puts the vendor’s revenue at $300 before and $150 after. The video gives no billing period.
What Usage-Based Pricing Is
So vendors change the meter. It is called usage-based pricing: you pay for how much you use, not for how many people have access.
The Idea on One Board
Per Seat Versus Per Use
Per seat
Per seat, the bill is easy to predict, but it follows headcount.
Per use
Per use, the bill follows the work, but it is harder to forecast.
Per Seat Versus Per Use
Per seat, the bill is easy to predict, but it follows headcount.
Per use, the bill follows the work, but it is harder to forecast.
Why It Matters
The video turns this into one question, and it is about the busiest month: what is the bill then?
Seats count people. Usage counts work. And agents work without a seat.
The key insight: Seats count people. Usage counts work. And agents work without a seat.
The honest limit
Per use, the bill follows the work, but it is harder to forecast.
Our report on the Microsoft and Hugging Face ThinkingBox post says the Microsoft and Hugging Face post reports that Claude Opus 5.5 and Claude Opus 5 each pass 241 of 507 tasks on all 20 attempts, at estimated costs of $7.80 and $13.30 per dependable task. The costs are the authors’ estimates at list rates. Counting a bill in units of work is the idea behind this lesson.
The Business Engineer Lens
Where this idea sits in the Business Engineer library:
The Business Engineer framework maps where each part of the AI stack sits and who holds the leverage. These pills are short lessons from the same library.
Outcome-Based Pricing contrasts with Usage-Based Pricing. Seats charge for access, usage charges for work, and outcomes charge for results.
Usage-Based Pricing builds on AI Agents. An agent doing the work of five people shrinks seat counts, so vendors move the meter to usage.
Run-Rate Revenue builds on Usage-Based Pricing. Much AI revenue is usage-based and may not repeat, which makes a month times twelve looser than recurring revenue.
Tokens is an example of Usage-Based Pricing. Models priced per token in and out put a meter on work, the usage-based model rather than seats.
Take-or-Pay(lesson unlocks soon) contrasts with Usage-Based Pricing. Take-or-pay fixes the buyer’s bill regardless of use; usage-based pricing makes the bill follow the work done.
Royalties(lesson unlocks soon) is an example of Usage-Based Pricing. Paying per loaf sold makes the bill grow with use, the per-use meter applied to someone else’s idea.
Cost per Task(lesson unlocks soon) contrasts with Usage-Based Pricing. Usage-based pricing charges per unit of use; cost per task asks what finishing the job costs, which can rise as unit prices fall.
Gennaro Cuofano is a CRO and tech executive who has worked in AI since late 2015, starting with bringing NLP, natural-language generation, voice and chatbot products to the marketing industry. His background is in law and finance: he holds a Master's degree in Law and an International MBA with an emphasis on corporate finance (LUISS Business School and the University of San Diego, 2012), and worked as a financial analyst at a real-estate investment firm in San Diego and as an assistant controller. His work focuses on business model strategy, business engineering and, more broadly, structural analysis: how companies make money, read from their own filings. He created FourWeekMBA and leads research there and at The Business Engineer, his newsletter on AI and business strategy, with over 95,000 subscribers and more than 1,000 published analyses. His writing has also appeared on Entrepreneur (for example, How to Digitize a Family Business in Europe) and HackerNoon. Find him on LinkedIn and Substack. See how we source.
Scroll to Top
Discover more from FourWeekMBA
Subscribe now to keep reading and get access to the full archive.
