Hypothesis Canvas

Purpose

The Hypothesis Canvas turns vague beliefs into structured, testable hypotheses. By forcing the team to articulate what they believe, why it matters, how they will test it, and what success looks like, the canvas prevents wasted effort on experiments without clear learning goals. It is a core tool in Lean Startup and Lean UX workflows.

At a glance

How to run

1. Select the assumption (3 min): Pick the riskiest assumption from an Assumption Map or backlog.
2. Fill the canvas (20 min): Work through each section:

• We believe: State the hypothesis clearly.
• To verify, we will: Describe the experiment or test.
• We measure: Define the metric or observable.
• We are right if: State the success criterion (quantitative threshold or qualitative signal).
• We will learn: Articulate what a positive and negative result each mean for the strategy.

1. Challenge (5 min): Another team member plays devil's advocate — is the hypothesis falsifiable? Is the metric measurable? Is the threshold meaningful?
2. Commit (2 min): Assign an owner, timeline, and resources for the experiment.

Materials needed

• Hypothesis Canvas template (printed A3 or digital)
• Markers
• The assumption or belief to test
• Access to data or experiment tools

Pitfalls & common mistakes

• Unfalsifiable hypotheses: If no outcome can disprove it, it is not a hypothesis. Rewrite.
• Vanity metrics: Choose metrics that reflect real behaviour, not feel-good numbers.
• Moving goalposts: Define success criteria before running the experiment, not after.
• Hypothesis overload: One canvas per hypothesis. Do not cram multiple beliefs onto one.

Inclusion & accessibility

• Provide the template in advance for participants who need processing time.
• Use plain language — avoid jargon like "falsifiable" without explanation.
• Allow asynchronous contribution for distributed teams.

Variations

• Lean Hypothesis Statement: A one-sentence format: "We believe [change] will result in [outcome]. We will know we are right when [metric] reaches [threshold]."
• Experiment Card: A lighter version with just Hypothesis, Method, Metric, and Deadline.
• Double Diamond Hypothesis: Pair with discovery research for richer framing.

When NOT to use this

• When the decision is already made and the experiment would not change it.
• When the team lacks resources to run even a lightweight experiment.
• For well-established facts that do not need testing.

Source & further reading

• Jeff Gothelf & Josh Seiden, *Lean UX* (2013)
• Eric Ries, *The Lean Startup* (2011)
• Strategyzer: Test Card