Pricing Your AI Product — Lessons from 400+ Companies and 50 Unicorns

⬅️ Back to Podcasts

  • Podcast: Lenny’s Podcast
  • Host: Lenny Rachitsky
  • Guest: Madhavan Ramanujam — managing partner at Simon-Kucher, author of Monetizing Innovation and Scaling Innovation
  • Duration: ~1 hour
  • Listen: Apple Podcasts | YouTube

Madhavan Ramanujam has worked with over 250 companies including 30 unicorns on pricing strategy. His thesis: AI changes everything about monetization, and most founders are getting it wrong.

Why AI Pricing Is Different

Three things make AI fundamentally different from traditional SaaS:

  1. Immediate value creation — AI delivers quantifiable outcomes, not vague productivity improvements. The attribution problem is solved.
  2. Real cost structure — Unlike SaaS (near-zero marginal cost), AI has real computational expenses that scale with usage.
  3. Labor budget access — AI competes against human labor, not software. Labor budgets are 10x larger than software budgets.

The Dual Engine Strategy

Companies fall into two traps. Market share–only companies give away too much value acquiring customers they cannot expand. Wallet share–only companies price so high they limit their addressable market. You must master both simultaneously.

The Attribution-Autonomy Framework

The right pricing model depends on two dimensions:

  • Low attribution, low autonomy → subscription pricing (work on improving attribution)
  • High attribution, low autonomy → hybrid (base + consumption credits, like Cursor)
  • Low attribution, high autonomy → usage-based pricing (usage as value proxy)
  • High attribution, high autonomy → outcome-based pricing (the premium quadrant)

Only 5% of companies use outcome-based pricing today. Ramanujam predicts 25% within three years.

POC as Business Case

Stop treating proof of concepts as technical validations. The entire goal of a POC is to create a business case. Co-create the ROI model with the customer from day one. Agree on assumptions before demonstrating outcomes. Charge for POCs to qualify serious buyers.

The 20/80 Axiom

20% of what you build drives 80% of willingness to pay. And ironically, that 20% is often the easiest to build. Identify your highest-value features and monetize them properly instead of giving them away.

Price Increases Are Emotional, Not Logical

“Your reluctance to do a price increase is often internal and emotional, not external and logical.” Regular price increases tied to value delivery should be standard practice.

Outcome-Based Pricing Power

Companies in the outcome-based quadrant can capture 25-50% of the value they create, compared to 10-20% for traditional SaaS. The autonomous nature of AI and clearer attribution makes this possible.

Crepi il lupo! 🐺