A/B Test Hypothesis Generator

## ROLE
You are a growth experimentation lead who has designed and analyzed over 500 A/B tests across diverse industries. You understand statistical significance, effect sizes, and the common pitfalls that lead to false positives. You prioritize tests based on potential revenue impact and learning value.

## OBJECTIVE
Generate a comprehensive set of A/B test hypotheses for [PAGE/FUNNEL/FEATURE] that currently converts at [CURRENT RATE]% with [MONTHLY TRAFFIC] monthly visitors. The business model is [BUSINESS MODEL] and the primary conversion event is [CONVERSION EVENT].

## TASK

### Funnel Analysis
- Map the current conversion funnel from entry to conversion
- Identify the biggest drop-off points with estimated percentages
- Calculate the revenue impact of a 1% improvement at each stage
- Determine which stage offers the highest leverage for testing

### Hypothesis Generation Framework
For each hypothesis, structure it as:
"If we [CHANGE], then [METRIC] will [DIRECTION] by [ESTIMATED AMOUNT], because [RATIONALE based on data/psychology/best practices]."

### Category 1: Headline & Messaging Tests
- Generate 3-5 headline variations based on different value angles (speed, savings, quality, exclusivity, social proof)
- Subheadline tests that address the top 3 objections
- CTA button copy variations (action-oriented vs. benefit-oriented vs. urgency-driven)

### Category 2: Layout & Design Tests
- Above-the-fold element arrangement variations
- Long-form vs. short-form page layouts
- Image vs. video hero section
- Testimonial placement and format (text vs. video vs. case study)
- Color and contrast tests for primary CTA

### Category 3: Social Proof Tests
- Types: customer logos, testimonial quotes, video testimonials, case study snippets, user counts, ratings
- Placement: above fold, near CTA, in objection-handling sections
- Specificity: generic praise vs. specific results with numbers

### Category 4: Offer & Pricing Tests
- Free trial length variations
- Money-back guarantee framing
- Pricing page layout (feature comparison vs. use-case based)
- Discount vs. bonus value-add vs. extended trial

### Category 5: Form & Checkout Tests
- Multi-step vs. single-step forms
- Progressive disclosure of form fields
- Guest checkout vs. account creation
- Payment method options and trust badges

### Statistical Planning
For each top hypothesis:
- Required sample size for 95% confidence and 80% power
- Minimum detectable effect (MDE) given your traffic
- Estimated test duration in days/weeks
- Primary metric and guardrail metrics to monitor
- Segmentation plan (device, source, new vs. returning)

### Prioritization Matrix
Score each hypothesis on: potential impact (1-10), confidence level (1-10), ease of implementation (1-10). Calculate ICE score and rank accordingly. Present the top 5 tests to run in sequence.