Two product-page variants are running together, but price, availability or seasonal demand shifted during the run. In this case, the immediate task is to plan a controlled comparison of product-page variants. The workflow should produce an A/B test plan and decision framework, validated against the merchant's own product records.
First replace every [bracketed] placeholder with real store scope and constraints, then add any actual records or data you have, and paste it into your preferred AI assistant or Spark agent. Spark will follow this page's workflow to produce structured output and verification steps.
Act as a Shopify operations specialist.
Help me plan a controlled comparison of product-page variants.
My store context is [describe the business, products and target audience].
My scope is [specify the affected products, pages, customers or reporting period]. Use these inputs: [the two product-page variants, the traffic split and assignment rule, the conversion metric in scope, the product record with price and availability across the whole window, and the merchant's stated observation window and minimum sample size].
Compare the variants only where both pointed at the same product at the same price and availability throughout the window, mark as contaminated any day on which price, stock or a promotion differed, and treat a comparison with fewer visitors than the merchant's stated minimum as inconclusive rather than as a loss.
Return an A/B test plan and decision framework for the specified task.
Show the relevant evidence or before-and-after examples and explain how I can verify the result.
If essential data is missing, ask for it and mark unsupported conclusions as unknown.
Do not invent metrics, product claims or customer facts.
For any publication, message, account or store change, use only connected tools within my explicitly authorized scope and report success only after verification; otherwise provide the draft or proposed steps.Split the work into input, judgment, and deliverable phases. Verify each step's output before moving to the next.
Collect the two product-page variants, the traffic split and assignment rule, the conversion metric in scope, the product record with its price and availability across the whole window, and the merchant's stated observation window and minimum sample size.
Compare the variants only where both pointed at the same product at the same price and availability throughout the window, mark as contaminated any day on which price, stock or a promotion differed, and treat a comparison with fewer visitors than the merchant's stated minimum as inconclusive rather than as a loss.
Deliver an A/B test plan with the decision framework, restate the result with the contaminated days excluded and the sample size shown, then reconcile the attributed orders against the order export and name the products the finding covers.
Cover decision rules, frequent misunderstandings, and required inputs so the team aligns on prerequisites before execution.
Any change to price, availability or an active promotion during the window contaminates the comparison, because the variant and the change cannot be separated in the result; those days are excluded and reported separately, and if too much of the window is contaminated the run is repeated rather than interpreted.
It must reach the merchant's stated minimum per variant before any difference is called a result, since a small sample turns normal variation into an apparent win; below that threshold the outcome is recorded as inconclusive and the window is extended instead of the weaker variant being removed.
It applies to the product tested and to products the merchant can show share its traffic and price behaviour, because a layout change that wins on a high-traffic product can lose on a slow one; the plan names the product scope explicitly so the finding is not generalised by habit.
Start with scenarios under the same topic, then expand into cross-topic matches suggested by similarity.
Ciwi Spark builds Shopify-native AI tools and agents around the workflows merchants actually run: research, ads, translation, and customer support. Stop rewriting the same prompts by hand.