The Crazy Ideas Pipeline
by Eeke de Milliano • Head of Product at Retool
Eeke is a product leader who spent six years at Stripe, starting as one of their first employees and PMs where she helped build foundational products like Stripe Connect and Radar. She now leads product at Retool, bringing deep expertise in scaling technical products and high-performance cultures.
🎙️ Episode Context
Eeke de Milliano dissects the tension between process and innovation, arguing that while process reduces variance, it often stifles the highest performers. She shares actionable frameworks for scaling Stripe-like cultures, including how to structure 'crazy ideas' into launched products, how to build a diversified talent portfolio rather than hiring clones, and why companies should aim for a 'Minimum Viable Process.'
Problem It Solves
Overcomes the natural tendency of organizations to gravitate toward safe, incremental work and ignore high-risk/high-reward innovation.
Framework Overview
A structured mechanism to solicit, fund, and incubate non-obvious ideas that would typically be killed by standard prioritization frameworks. It treats the company as a VC and product teams as startups.
🧠 Framework Structure
The 10% Probability Prompt: Create a ...
VC-Style Funding: Fund selected ideas...
Organizational Separation: Keep incub...
Review and Revive: Annually review th...
When to Use
During annual planning cycles or when the team feels stuck in an incremental optimization loop.
Common Mistakes
Integrating the 'crazy idea' team back into the core org too quickly, where they get crushed by technical debt and bug fixes before finding PMF.
Real World Example
Retool Workflows, a major product launch, started as a 'Crazy Idea' submission because scheduling queries wasn't part of the core UI builder value prop.
Crazy ideas are ideas that we shouldn't, obviously, do. There's a 90% chance that they make no sense. But in the 10% chance that they do, they will make a 10x to 100x difference.
— Eeke de Milliano