# Call-to-Action Test Plan

Define one call-to-action hypothesis, controlled variants, audience, primary metric, quality guardrails, risk, duration, analysis, and stopping rule.

## Usage note

Test one meaningful decision while preserving accurate expectations, user choice, privacy, and accessibility. A higher click rate is not success if the variant misleads readers or worsens downstream outcomes.

## How to use this template

1. Define the observed barrier, evidence, audience, current action, and ethical test hypothesis.
2. Create controlled variants whose promises and destinations remain accurate and accessible.
3. Predefine assignment, primary outcome, guardrails, sample, duration, and stopping rules.
4. Launch with verified instrumentation and monitor only stated safety and data-quality conditions.
5. Analyze uncertainty and downstream outcomes, document a decision, and preserve the learning.

## Blank template

### Problem and hypothesis

- **Page, audience, and journey stage:** [Enter]
- **Current CTA and destination:** [Enter]
- **Observed barrier and evidence:** [Describe]
- **Hypothesis:** [If change, audience behavior may change because mechanism]
- **Element intentionally unchanged:** [List]
- **Owner and stakeholders:** [Enter]
- **Concurrent change risk:** [Describe]

### Variants and integrity

- **Control wording and placement:** [Enter]
- **Variant wording and placement:** [Enter]
- **Difference being tested:** [One factor]
- **Offer, eligibility, cost, and data requirement:** [Verify]
- **Accessibility and localization review:** [Owner/status]
- **Manipulation or misunderstanding risk:** [Describe]
- **Destination experience verified:** [Who/date]

### Measurement plan

- **Primary outcome:** [Define event and window]
- **Quality guardrails:** [Completion, cancellation, error, complaint, other]
- **Assignment unit and population:** [Enter]
- **Exclusions:** [Enter]
- **Sample and duration rationale:** [Analyst input]
- **Instrumentation QA:** [Events and tester]
- **Safety stop:** [Define]
- **Data-quality stop:** [Define]

### Analysis and decision

- **Predefined comparison:** [Enter]
- **Result and uncertainty:** [Enter]
- **Guardrail result:** [Enter]
- **External/confounding events:** [List]
- **Decision:** [Ship / Iterate / Keep control / Investigate / Roll back]
- **Reason and limitation:** [Enter]
- **Release or rollback owner:** [Enter]
- **Next question:** [Enter or None]