Pricing systems for companies selling AI
Axion builds pricing decision systems that work inside your operations. They turn your own deals, customer and cost data into pricing guidance, with clear rules on what the system recommends and what a person approves.
Axion is in development and plans to begin client work in 2028. We welcome early conversations with pricing teams.
- 1
Question
How should we charge for our new AI assistant feature?
- 2
Evidence
Drawn from the company's own records and the market.
Past dealsCustomer segmentsCompetitor pricesCost to serve - 3
Guidance
A base subscription with included usage, a monthly cap, and a price range by segment.
- 4
Approval
Within agreed limits, the guidance applies. Outside them, the pricing lead reviews it before any customer sees it.
Why AI products are hard to price
Setting a price takes a clear view of what it costs to serve each customer, what customers will pay, and what competitors charge. AI makes all three harder.
Every use costs money
Each request runs on paid compute, and those costs change as models and infrastructure change.
Value varies widely
Two customers on the same plan can get very different results and use very different amounts.
Willingness to pay is unclear
The technology is new and moving fast, so there is little history to price from.
When these are missed, products lose money on heavy users, customers get bills they can't predict, and adoption stalls.
What Axion does
Axion's work has three parts, and stays with a client after the system goes live.
Pricing decision infrastructure
A system your team uses through an AI agent. It identifies which pricing questions to answer, combines your historical deals, customer segments, market trends and competitor information, estimates cost to serve and willingness to pay, and returns pricing guidance.
Design and implementation
Deployment in stages inside your own operations, including the rules on which decisions the system may recommend and which a person must approve.
Ongoing stewardship
Regular monitoring, recalibration and controlled updates, so pricing keeps up as costs, markets and customer behavior change.
Where Axion fits
Most pricing tools are built for large pricing departments, and most consulting projects end with a report.
| Option | Built for | After launch |
|---|---|---|
| Pricing software | Experienced teams that already know which questions to ask | Your team runs and maintains it |
| Pricing consultancies | One-time projects | Recommendations are handed over |
| Axion | Teams that are small, stretched or new to pricing AI products | Axion keeps monitoring and updating the system |
Who Axion is for
Companies that sell AI products, and companies that use AI at scale in industries where cost to serve drives every price.
- AI products and AI-enabled software
- Supply chain and fulfillment
- Manufacturing
- Healthcare technology
Founder
Xuerui (Rachel) Chen has worked on pricing for AI features in business software and on pricing in manufacturing and fulfillment, where cost to serve shapes every decision, as it now does for AI.
- More than eight years in pricing across software, manufacturing and logistics
- M.S. in Marketing Analytics, pricing concentration, University of Rochester Simon Business School
- Certified Pricing Professional, Professional Pricing Society
- Pricing coursework, MIT Sloan
Talk with us
If you price an AI product, or are about to, we'd like to hear what you're working on.
Email hello@axionpricing.com