A long list of improvement ideas has accumulated, but nothing in it says which item should be done first or why. In this case, the immediate task is to build a prioritized store conversion improvement workflow. The workflow should produce a conversion diagnosis and experiment backlog, ordered by a priority rule the merchant's own records can support.
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 build a prioritized store conversion improvement workflow.
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: [each candidate improvement with its supporting page or segment counts and period, the merchant-assigned effort and owner, the metric each change would move, the testable traffic volume, the stated priority rule, and the items already in progress]. Score each candidate with that one rule only, comparing evidence strength against the effort recorded, and mark as unranked any item whose evidence cannot be recomputed from the supplied records or whose traffic is below the merchant's stated minimum for a test, listing those with raw counts instead of a rank.
Return a conversion diagnosis and experiment backlog with every priority verified against the merchant's records 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 every candidate improvement with the records it rests on, including the page or segment, the counts observed and the period they cover, together with the effort and owner the merchant assigns, the metric each change would move, the traffic volume available to test it, the stated priority rule, and any items already in progress.
Score each candidate with that one rule only, comparing evidence strength against the effort recorded, and mark as unranked any item whose evidence cannot be recomputed from the supplied records or whose traffic is below the merchant's stated minimum for a test, listing those with raw counts instead of a rank.
Deliver a conversion diagnosis and an experiment backlog ordered by that rule, each entry carrying its evidence, its success and guardrail metric, its owner and the record it can be checked against, then reconcile the backlog count with the candidate list the merchant submitted and note every item whose rank rests on a definition differing from the store-wide rate.
Cover decision rules, frequent misunderstandings, and required inputs so the team aligns on prerequisites before execution.
Each item must point to a record the merchant holds, such as the counts for the page or segment it concerns, the period those counts cover and the metric it would move. Evidence is what makes the ranking defensible rather than a matter of taste; an idea traced only to a preference is listed as unevidenced and is not ranked.
Effort is recorded in the terms the merchant uses, naming the owner, the work involved and whether the change can be tested on a subset of traffic. Recording it this way keeps the rule reproducible when the backlog is reviewed later; where effort is unknown, the item is marked as unestimated and ranked after the items with a recorded effort.
They are listed in a separate section with the reason, rather than dropped or ranked by enthusiasm. Such ideas often become testable once tracking covers the step they concern, so keeping them visible preserves them; they enter the ranked backlog only once a record can be produced for them.
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.