Going to the End

Don't stop until you've reached first-principles truth—expert answers are often incomplete

Ayo Omojola
Frameworks for product differentiation, team building, and first principles thinking

Going to the End

"You can't stop until you get to the end. And that's one of the reasons why being in a domain that matters is really important, because that's a very, very expensive activity to do if you're in a small team environment." - Ayo Omojola

What It Is

"Going to the End" is a first-principles approach to understanding complex problems. The core insight is that when you ask experts for answers, you often get incomplete or outdated information—not because they're lying, but because their understanding is limited by their experience. Real breakthroughs require pushing past expert answers until you reach fundamental truth.

This framework emerged from Ayo's experience at Cash App, where he discovered that experts often provided answers based on their tenure and past experience that turned out to be wrong. The team had to literally project regulatory text on screens and debate what it really meant, rather than accepting anyone's interpretation.

The framework acknowledges this is expensive—it takes time and energy. That's why it's only worth doing in "domains that matter" where the investment pays off.

How It Works

The Expert Problem: When you're not an expert, you typically ask the most tenured person in the domain for answers. But tenure and depth of experience vary wildly. Their answer reflects their mental model, which may be:

  • Based on how things worked at their last job
  • Outdated by new regulations or technology
  • Simply wrong but confidently stated

The "End" Solution:

  1. Don't accept first answers as final
  2. Push until you reach the actual constraint (contract, regulation, physics, etc.)
  3. Question "can't be done" with "what specifically will break?"
  4. Iterate until two incongruent things reconcile

The Cash Card Example: When building Cash App's card, Ayo visited card manufacturing factories, tested 1,000+ combinations of plastic, overlay, paper, envelope, and finishes. He learned about laser engraving settings that the vendors didn't initially present as options. Going to the end meant understanding the manufacturing process at the machine-setting level.

How to Apply It

  1. Don't stop at "can't" - When someone says something can't be done, ask: "What specifically will break? Will they fine us? Will the patient be harmed? What are the actual consequences?"

  2. Go to primary sources - Read the actual regulation, contract, or technical documentation. Project it on a screen and debate what it really means.

  3. Reconcile incongruencies - When two data points don't match, dig in. There's always a reason, and finding it often reveals important truths.

  4. Visit the factory - Literally or figuratively, go see how things are actually made. Understanding at the physical/implementation level reveals possibilities that abstractions hide.

  5. Budget for the cost - This is expensive in time and energy. Only go to the end on things that truly matter for your business.

  6. Delegate the expertise - The person in the execution role must become the expert. They have to know that's what you're holding them accountable for.

When to Use It

  • Regulated industries: Where compliance constraints define what's possible
  • Technical innovation: When pushing beyond what vendors say is possible
  • New market entry: When expert opinion reflects old assumptions
  • Data analysis: When metrics don't reconcile and you need to understand why
  • Strategic decisions: When the answer matters enough to justify the cost

Source

  • Guest: Ayo Omojola
  • Episode: "Frameworks for product differentiation, team building, and first principles thinking"
  • Key Discussion: (32:27) - The Cash Card story; (35:55) - Application at Carbon Health
  • YouTube: Watch on YouTube

Related Frameworks

Back to Execution