A new framework for AI agent pricing
Patterns from 60+ AI agent companies, the best single mental model for what you'll be charged and why.
Open growthunhinged.com →
Sales & Customers
Article
Thread
Article
Article
Podcast
Tool
Article
Guide
Article
Article
Article
Article
What does AI support actually cost, and how should a founder think about the ROI?
The short answer
Two pricing models dominate: per-resolution (Fin's $0.99, you pay only when the AI actually solves a ticket) and platform or usage pricing, with enterprise players like Decagon tying contracts to outcomes. The honest math compares it to your fully loaded cost per human-handled ticket (typically $5-10), but read the fine print: 'assumed resolutions', handoff fees, and volume minimums make bills less predictable than the headline price. Klarna's $40M savings claim and subsequent walk-back is the reminder that ROI includes retention, not just deflected tickets.
A quick orientation. The real value is below: resources worth your time, from people who've actually done it.
12 resources worth your time
The most disruptive AI pricing models so far
A rapid-fire comparison of Salesforce, Intercom and Zendesk AI pricing you can screenshot into your budget doc.
Open substack.com →How Intercom Reaccelerated Growth With Outcome-Based Pricing
A CFO's-eye view of why $0.99 per resolution worked, and what that means for you as the buyer.
Open mostlymetrics.com →AI Is Driving A Shift Towards Outcome-Based Pricing
Explains why support was the first category where pricing moved from seats to resolved outcomes.
Open a16z.com →Bret Taylor of Sierra on AI agents, outcome-based pricing, and the OpenAI board
Stripe's Collison pushes Taylor on whether outcome pricing is real economics or clever packaging.
Open cheekypint.substack.com →Intercom Pricing
The primary source: current plan plus per-resolution pricing to plug into your own cost model.
Open intercom.com →Intercom Fin AI Pricing Explained: Evaluating $0.99 Per Resolution
Breaks down what actually counts as a billable resolution, including assumed resolutions and handoff charges.
Open gleap.io →Fin AI Agent outcomes
Intercom's own definitions of the four billable outcome types, the fine print your invoice depends on.
Open intercom.com →Kyle Poyar: The State of AI Pricing in 2025
A concise state-of-play on AI pricing experiments and which ones are sticking.
Open schematichq.com →AI Agent Pricing Models Explained
A plain-language taxonomy (per-seat, usage, outcome, hybrid) for comparing vendor quotes apples to apples.
Open pickaxe.co →Klarna Reverses On AI, Says Customers Like Talking To People
The ROI counter-example: headline savings that ignored churn and quality costs.
Open forbes.com →How Much Does Intercom's Fin AI Cost?
A worked cost example at startup ticket volumes, useful for a quick back-of-envelope.
Open featurebase.app →More in Customer Support
- Why can AI now handle most support tickets, and should it? 13 →
- How do startups set up AI support (Intercom Fin, Decagon, custom bots) in a week? 12 →
- How do founders use support conversations as product research with AI? 11 →
- How do you stop an AI support agent from hallucinating and breaking customer trust? 12 →
- When do customers still need a human, and how do you design the AI-to-human handoff? 12 →