How AI Products Make Money: Models, Pricing, Margins

Enroll

SaaS pricing had a settled playbook and AI broke it. Freemium math collapses when every free user burns tokens, and the per-seat instinct ignores costs that scale with usage. This course teaches monetization for AI products end to end, covering goal setting, packaging, price metrics, rate-setting and discounting, and then adds the layer specific to AI: unit economics where inference is a real COGS line, consumption pricing set against seat pricing, and how to make the eval-backed case that a stronger and more expensive model pays for itself. It runs on case studies throughout, from DocuSign to Netflix to a RAG cost decision I had to make myself.

The outcome

How does this prepare you for a job?

Pricing is now a named function, with titles such as Pricing Manager and Price Intelligence, and every AI PM interview eventually reaches the margin question.

Built unit-economics models for AI products, covering inference COGS per request and gross-margin scenarios across different model choices and usage tiers.
Designed packaging and price metrics across seats, consumption, outcomes and hybrids, matched to the way value is actually delivered.
Ran a build-versus-buy analysis for a RAG system with a full cost model, comparing building, hosting and maintaining against vendor pricing.
Justified model selection with evals, quantifying when a stronger model's per-call premium nets out cheaper overall.
Presented monetization strategy to executive stakeholders with assumptions that could be defended and sensitivity ranges around them.
Applied a five-step pricing method covering goals, packaging, metrics, rate-setting and discounting to a live product.

Tools

AI tools learnt

Deliberately light on build tools. The instrument here is the model working inside the spreadsheet.

Hands-on skills

Model inference COGS from tokens per action, actions per user and cost per model, down to a gross margin you can defend
Choose the price metric between seat, consumption and outcome, knowing what each does to adoption and to margin
Package tiers and add-ons without freemium draining the margin
Run rate-setting using value-based anchors, competitor scans and signals of willingness to pay
Set discounting rules that a deal desk can actually hold to
Make the case for the stronger model, weighing eval-backed quality deltas against the per-call premium
Present pricing to executives, covering the room, the pushback and the one-pager

Syllabus

Modules

Level: Basic

Goals before numbers

Align monetization with strategy, choosing between growth, margin and land-and-expand
DocuSign, Zoom and Squarespace, and what each goal choice looked like in practice
Write the goal statement your pricing has to serve

Packaging and price metrics

Tiers, add-ons and bundling, and how to evaluate the packaging options
Seat pricing set against consumption pricing, using the Nosto and Kustomer/HelpShift cases
Implementability, meaning the metric your billing system can actually meter

AI unit economics

Modeling token COGS from first principles
Why freemium math breaks for AI products specifically
Margin scenarios across different model choices

Level: Advanced

Rate-setting and discounting

The essentials of value-based pricing, and the Netflix rate-setting case
Selecting a price point, and testing it
Discount rules, deal desks, and the international wrinkle, using Yousign

The AI-specific decisions

Build or buy a RAG system, using the real cost estimate I arrived at in practice
Cost-benefit across models, meaning the argument for a stronger model used fewer times
Evals as the layer that justifies spend, where no eval delta means no upgrade

Selling it inside

Present the strategy to executives and difficult stakeholders, and handle the silent killers of executive presence
Pricing research, covering both quantitative and qualitative techniques
Who owns pricing, which is an organizational question, using the Citrix case

Students can take the full course (both levels) or either level on its own.

Register

Waitlist

Join the course waitlist

No cohorts are scheduled yet. Join the waitlist and we’ll email you when registration opens.