Several creatives compete for the same audience and the campaign history is on file, but no test has isolated which one wins. In this case, the immediate task is to plan a controlled test of creative concepts. The workflow should return a test matrix holding budget, audience and landing page constant so differences trace back to the creative.
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 test of creative concepts.
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 creatives to compare, the shared audience and placement set, the primary decision metric, the merchant's reported conversion lag, and the campaign history export with its currency and attribution setting]. Hold budget, audience, landing page and bid strategy identical across variants so the only difference is the creative, allocate delivery through one shared audience rather than overlapping ones, and fix the start date, end date and decision rule before launch, marking any variant that receives almost no delivery as unevaluable instead of a loser.
Return the test matrix with variant, window and decision rule, confirmed against the supplied campaign history 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.
List the creatives to compare, the single audience and placement set they will share, the primary metric that decides the result, and the campaign history export covering the conversion lag the merchant reports, with the currency and attribution setting that export uses.
Hold budget, audience, landing page and bid strategy identical across variants so the only difference is the creative, allocate delivery through one shared audience rather than overlapping ones, and fix the start date, end date and decision rule before launch, marking any variant that receives almost no delivery as unevaluable instead of a loser.
Deliver the test matrix naming each variant, the creative element it changes, the shared constants, the fixed window and the decision rule, then reconcile the number of variants in the matrix with the number actually receiving delivery in the ad account and note every variant whose results are still inside the conversion lag.
Cover decision rules, frequent misunderstandings, and required inputs so the team aligns on prerequisites before execution.
One variable changes while budget, audience, landing page and bid strategy stay identical, because a second change gives the result two possible causes; the decision metric and end date are fixed before launch so the outcome cannot be read after the fact.
They are marked unevaluable rather than losing, since an undelivered variant carries no evidence either way; the fix is to confirm delivery settings and rerun that arm inside the same window instead of reading its empty result as a verdict.
Only when the merchant's stated stop condition is met, such as a clear breach of a spend or performance limit, because stopping at the first favourable day usually captures the conversion lag rather than a real difference between creatives.
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.