Experiments · Hypotheses · Validation · Product Management · PTOS

The Ladder of Evidence: How to Choose the Right Test for Your Hypothesis and Not Deceive Yourself

Breaking down the 'Ladder of Evidence'—a hierarchy of signals from opinion to money that helps choose the cheapest and most honest way to test a hypothesis depending on the level of risk.

The Ladder of Evidence: How to Choose the Right Test for Your Hypothesis

In product development, all hypotheses are created equal, but some signals are 'more equal' than others. Teams often fall into the trap of accepting weak signals as irrefutable proof. 'Users liked the mockup' is not the same as 'users are willing to pay for this.'

To avoid deceiving yourself, the PTOS framework uses the 'Ladder of Evidence'—a hierarchy of signals that helps choose an appropriate method for testing a hypothesis. The higher the risk, the higher up the ladder the signal must be.

The Rungs of the Ladder of Evidence (From Bottom to Top: From Cheapest to Most Honest)

1. Opinion / Evaluation

What it is: A user says, 'Yes, that sounds useful,' or 'I like this design.'
Signal: The weakest. This is polite feedback that says almost nothing about future behavior.
When to use: In the very early stages, to weed out completely failed ideas.
Test: Survey, 5-second test.

2. Understanding

What it is: The user can explain the value of your proposition in their own words and what next step they should take.
Signal: Useful, but still doesn't prove a real need.
When to use: To test the clarity of your value proposition.
Test: Message test, interview.

3. Intent in Context

What it is: The user performs an action that requires a minimal 'stake' from them—attention or a click.
Signal: Much stronger than opinion. It shows that in the real context of the product, your offer has sparked interest.
When to use: To check if a feature is needed at all.
Test: A click on a fake-door button, starting to fill out a form.

4. Commitment

What it is: The user is willing to 'pay' with their time or reputation.
Signal: A strong indicator of real pain.
When to use: When you need to filter out those who are 'just curious' from those who are truly ready to act.
Test: Scheduling a call for a detailed discussion, filling out a long form with budget/timeline details, signing a Letter of Intent (LOI).

5. Behavior with Repetition

What it is: The user not only performed the target action (value-event) but also returned to do it again.
Signal: Very strong. Repeated use is the main sign that the product is truly solving a problem and becoming a habit.
When to use: To assess real adoption and retention.
Test: A pilot on a small group, Concierge MVP, Small safe release.

6. Money

What it is: The user is willing to pay real money for the solution.
Signal: One of the harshest and most honest indicators. Money is a concentrated expression of value.
When to use: To test the viability of the business model.
Test: Prepayment, paid pilot, selling before the product is fully ready.

How to Use the Ladder?

The rule is simple: the higher the risk of your hypothesis (new product, new segment, high price), the higher up the ladder of evidence you should aim.

Testing a new feature for an existing product? A signal from the 'Intent' (fake-door) and 'Behavior' (pilot) levels might be enough.
Launching a completely new product in a new market? You need a signal from the 'Commitment' (LOI) or 'Money' (prepayments) levels to avoid spending years building something nobody needs.

The Ladder of Evidence is your navigator in the world of uncertainty. It helps you not to waste resources on expensive tests where cheap ones will do, and not to mistake wishful thinking for reality.