The 'Maximum Treatment' Protocol
by Various Product Leaders (Hosted by Lenny Rachitzky) • Product Executives at Airbnb, Stripe, Intercom, Quibi, Ramp, Duolingo, Toast
A compilation of elite product leaders including Katie Dill (Stripe), Paul Adams (Intercom), Tom Conrad (Quibi/Pets.com), Sri Batchu (Ramp), Jiaona Zhang (Webflow), Gina Gotthilf (Latitud), and Maggie Crowley (Toast) sharing their most significant career failures.
🎙️ Episode Context
This special compilation episode aggregates the most painful and instructive stories of failure from top product leaders. It shifts the focus from 'survivorship bias' success stories to the gritty reality of failed launches, leadership mutinies, and bankrupt companies, extracting the specific lessons that allowed these leaders to bounce back and succeed later.
Problem It Solves
Solves the problem of inconclusive experiments in low-volume (B2B) environments where statistical significance is hard to achieve with small changes.
Framework Overview
Instead of isolating variables (e.g., changing just the email subject line), this methodology argues for changing every possible variable at once to maximize the chance of moving the metric. If the metric moves, you can cost-rationalize and isolate variables later. If it doesn't move after maximum effort, you can conclusively abandon the hypothesis.
🧠 Framework Structure
Maximize the Treatment Effect: In the...
The 'Kitchen Sink' Approach: Combine ...
Fail Conclusively: If the most expens...
Cost-Rationalize Later: If the 'maxim...
When to Use
When testing broad strategic hypotheses (e.g., Account-Based Marketing) in a B2B setting or when user traffic is too low for granular A/B testing.
Common Mistakes
Testing small, incremental changes (like button color) in low-traffic environments, leading to false negatives where the signal is lost in the noise.
Real World Example
Sri Batchu at Ramp testing Account-Based Marketing by combining every possible touchpoint (email, design, triggers) rather than testing them sequentially.
If you have a hypothesis... just throw all of the possible tactics and resources that you think would move that needle because you can always cost rationalize later if it works.
— Various Product Leaders (Hosted by Lenny Rachitzky)