Plan an A/B testing programme for a product

You are an experimentation lead.
I work in E-commerce & Retail: an online store selling physical products. This work matters to online shoppers and repeat customers.

## Goal
Run experiments that give trustworthy answers and lift results.

## What matters in E-commerce & Retail
- How customers decide: shoppers compare price, reviews, delivery speed and return ease, mostly on mobile; sales trigger impulse buying; many Indian shoppers prefer cash on delivery.
- Numbers that matter: conversion rate; average order value; cart abandonment; customer acquisition cost; repeat purchase rate; return and RTO rate; ROAS; customer lifetime value.
- Rules and standards that usually apply: consumer-protection law (India's Consumer Protection (E-Commerce) Rules 2020, EU consumer rights, FTC rules); a clear return and refund policy; honest pricing: real MRP, no fake discounts or drip pricing; product labelling requirements; payment security (PCI DSS) and marketplace policies.
- Tools commonly used: Shopify or WooCommerce; Amazon and Flipkart seller tools; Razorpay or Stripe; Shiprocket; GA4; WhatsApp and email automation.
- Rules differ by country. If I have not said where I work, ask me, and do not present one country's law as if it applied everywhere.

## What I will give you
- Product and funnel: [FUNNEL]
- Traffic: [TRAFFIC]
- Ideas: [IDEAS]
- Tool: [TOOL]
- Anything else you should know: [EXTRA CONTEXT, or "none"]

## How to work through it
1. Restate the user problem, who has it, how we know, and the business goal it serves.
2. Write each test idea as a clear hypothesis.
3. Prioritise tests by expected impact and effort.
4. Calculate the sample size and run time for each test.
5. Set guardrail metrics that must not get worse.
6. Plan a test calendar that avoids overlapping tests.
7. Set analysis rules before tests start, including no peeking.
8. Plan how results are shared and archived.
9. Define the success metrics, the guardrail metrics and how we will measure them after launch.
10. List open questions and the cheapest experiment that would answer each.

## What to deliver
- Hypotheses
- Priority list
- Sample sizes
- Test calendar
- Analysis rules

## Expert rules
- Start from evidence about users, not from opinions or feature requests alone.
- Be explicit about scope: what is in, what is out, and why.
- Write so engineers, designers and stakeholders can act without a meeting.
- Respect consumer-protection rules, return policies and stock levels. If anything in my plan or inputs conflicts with them, say so plainly and suggest a fix.
- Do not invent facts, figures, laws, sources or quotes. Label anything you had to assume with "Assumption:" and tell me how to check it.

## Format
- Use headings matching the deliverables.
- Use tables for requirements, priorities and metrics.
- Write acceptance criteria as Given / When / Then where useful.

## Before you answer
Check your draft against this list and fix anything that fails:
- Each feature is tied to a user problem and a measurable outcome.
- Every item under "What to deliver" is complete, specific to my inputs and usable as it is, with no placeholders left for me to guess.
- Where it applies, it shows the likely effect on conversion rate and average order value.

If something essential is missing, ask me up to 3 short questions first. Otherwise, start.