Use Sensitivity to Decide What to Learn Next

Turning Fragility Into Learning Priorities

Once profit has been correctly interpreted, the question changes.

The question is no longer whether profit is possible or fragile.

It becomes:

Which assumptions matter enough to justify learning, delay, or redesign—before committing further?

This toolkit exists to answer that question.

It does not help you optimize profit.
It helps you decide what to pay attention to next.

This toolkit builds directly on the toolkit for reading a profit curve.

It assumes you can already recognize:

  • impossible profit,
  • fragile profit,
  • and robust profit.

Sensitivity does not replace that judgment. It sharpens it by revealing where being wrong would matter most.

Step 1: Treat Every Input as a Belief, Not a Fact

Profit estimates rest on assumptions.

Some are well supported.
Others are weak guesses.
All are uncertain.

Sensitivity analysis begins by reframing inputs as beliefs rather than facts:

  • demand shape,
  • willingness to pay,
  • variable cost per unit,
  • fixed commitments,
  • scale and access assumptions.

The goal is not to eliminate uncertainty.
It is to understand which uncertainty matters.

Step 2: Ask the Right Sensitivity Question

Sensitivity is often misunderstood.

The question is not:

Which variable moves profit the most?

The correct question is:

Which assumption, if wrong, would change the decision?

An assumption can be

  • highly uncertain but low consequence, or
  • fairly certain but catastrophic if wrong.

Sensitivity focuses on consequence, not likelihood.

Step 3: Vary One Assumption at a Time—Deliberately

To learn from sensitivity, change one assumption at a time while holding others fixed.

Typical candidates include:

  • demand slope (price sensitivity),
  • maximum willingness to pay,
  • variable cost per unit,
  • fixed cost magnitude or timing,
  • achievable scale or access.

For each variation, observe:

  • whether profit disappears,
  • whether fragility worsens,
  • or whether robustness remains.

The goal is not realism. The goal is exposure.

Step 4: Identify Decision-Critical Assumptions

An assumption is decision-critical if:

  • small changes reverse feasibility, or
  • reasonable error collapses profit, or
  • robustness depends heavily on it being “just right.”

These assumptions deserve attention.

Others—even if uncertain—may not.

Step 5: Sort Assumptions by Actionability

Not all sensitive assumptions should be treated the same.

Ask:

  • Can this assumption be tested further?
  • Can it be influenced by design?
  • Can commitment be delayed until it is clearer?
  • Is it largely outside our control?

This step prevents wasted effort.

Sensitivity without judgment leads to anxiety. Sensitivity with judgment leads to strategy.

Step 6: Decide the Next Responsible Move

Sensitivity analysis does not tell you what to do.
It tells you what it would be irresponsible to ignore.

Common responses include:

  • Experiment
    When a sensitive assumption can be tested cheaply.

  • Delay commitment
    When fragility is high but learning is ongoing.

  • Redesign
    When the structure itself creates unnecessary exposure.

  • Proceed deliberately
    When robustness exists despite uncertainty.

  • Stop
    When the decision fails under any reasonable assumption.

Each response is valid. What matters is that it is chosen consciously.

Step 7: Use Sensitivity to Allocate Attention, Not Precision

Sensitivity analysis is not about fine-tuning numbers.

It is about deciding:

  • what deserves more evidence,
  • what can be tolerated as uncertain,
  • and what must be resolved before commitment.

Time, money, and focus are scarce.

Sensitivity helps you spend them where they matter most.

What This Toolkit Refuses to Do

This toolkit does not: - forecast future outcomes, - or guarantee success.

It exists for a narrower—and more important—purpose:

To prevent commitment based on fragile assumptions that have not earned that privilege.

Where This Leaves the Decision

By the end of this toolkit, you should be able to say:

  • which assumptions your decision depends on most,
  • how much error you can tolerate,
  • and whether further learning is warranted before committing.

At that point, analytics has done its job.

What remains is judgment.

Sensitivity does not remove uncertainty.
It ensures you know where uncertainty still matters.

That is the discipline that makes pre-revenue decisions responsible rather than hopeful.