Why some agent ideas are uninvestable

Madhavan Ramanujam stepped into the AI pricing conversation with a contrarian view: everyone is doing it wrong. As author of "Monetizing Innovation" and founding partner of 49 Palms Ventures, he's advised 250+ companies on pricing strategy. Now he's seeing founders leave millions on the table by using outdated SaaS playbooks for AI products.

> "Your moat is monetization and GTM. Can you get products in the hands of many people? Can you make it stick? And will they pay for the value?"

His message is clear: the two-year coding head start you're banking on is worthless. In the AI era, your competitive advantage isn't technology - it's how you price and distribute it.

The $50K to $500K Pricing Revelation

Madhavan shares a story that captures everything wrong with AI pricing today. A founder came to him charging $50K for an AI agent that was creating tens of millions in value. The founder thought it was "reasonable" and helped close deals faster.

"Put an option on the table. A $50,000 plus 10% of the outcomes that I generate or a $500,000 fixed fee."

This simple choice changed everything. The conversation shifted from price to value measurement. Customers chose the $500K option and negotiated down to $400K - a 10X increase with the same sales velocity.

"Which investor does not want that, right? That you could actually 10X your price and have the same sales velocity. Why wouldn't you do that?"

Why most AI companies are pricing wrong

The failing pattern Madhavan sees repeatedly:

Companies tie pricing to costs rather than value. Founders add 20% margin to token costs and call it pricing. As costs fall, so does revenue, even though customer value remains constant.

> "You just tied yourself to like a destiny that your pricing is going to keep coming down."

The labor budget opportunity. AI agents tap into labor budgets that are 10X larger than IT budgets, yet founders still price like SaaS:

> "200k to hire a salesperson and you charge $5,000 for a seat for a year. I mean like that doesn't make sense, right?"

Underestimating AI's value capture potential. Traditional SaaS captures 10% of value created. AI with high autonomy and clear attribution can capture 25-50%:

> "There is increased autonomy and there is increased attribution. And you can justify that."

The Human + AI Pricing Formula

Madhavan's framework for pricing AI that replaces human labor challenges conventional thinking:

> "If your AI can operate as the best salesperson, and is available 24/7, why wouldn't you think about it that way?"

His argument: It takes six months to hire someone, six months to train them. By the time they're productive, you've invested two years of salary. Your AI is productive on day one and works 24/7. Price accordingly - at or above human cost, not at 10% of it.

The Partnership Imperative

For established SaaS companies trying to add AI, Madhavan sees most getting stuck in seat-based pricing with no path to value attribution. His advice for companies like Slack:

> "I think it has to be first principles thinking, can I build some agents that actually sit on top of Slack and can do some meaningful work that I can monetize on it separately."

The key is finding the equivalent of Intercom's Fin.ai model - an agent that solves end-to-end workflows with clear value attribution.

Pricing in the Age of AI

Madhavan's framework for AI monetization starts before building:

> "Price before product. Period."

His approach:

• Have willingness-to-pay conversations before writing code
• Build only what people value enough to pay for
• Choose the right pricing archetype based on product characteristics
• Move from cost-plus to value-based pricing as quickly as possible

Leadership Lessons from the Pricing Frontier

Madhavan is juggling three major initiatives: running his fund (49 Palms), deploying capital, and launching his new book "Scaling Innovation." His thesis ties them together:

> "Monetization is the key to winning in AI."

His investment philosophy focuses on durable monetization rather than growth at all costs:

> "You need to have a clear conviction that you can actually, at the end of the day, build a profitable growth business."

Companies Mentioned

• 49 Palms Ventures
• Delphi
• Sierra
• Intercom (Fin.ai)
• ServiceNow
• Slack
• Workday
• Superhuman
• Gmail
• DocuSign
• Zapier
• YC (Y Combinator)
• First Round Capital
• NFX
• Stanford

See Privacy Policy at https://art19.com/privacy and California Privacy Notice at https://art19.com/privacy#do-not-sell-my-info.