Two dashboards show different conversion rates for one store because each divides a different numerator by a different base. In this case, the immediate task is to analyze store conversion rates with aligned denominators. The workflow should produce a conversion diagnosis and experiment backlog, each rate recomputed from the merchant's own figures.
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 analyze store conversion rates with aligned denominators.
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 disputed rate with its numerator, denominator, window, timezone, source and attribution rule, the raw sessions with their session rule, the orders with their status filter and currency, and the definitions used by each dashboard]. Recompute each rate from those raw components with one numerator and one denominator applied everywhere, express the result with the window and the rules used, and set aside any rate whose components are absent from the supplied records instead of substituting the figure from a dashboard.
Return a conversion diagnosis and experiment backlog, noting any rate that cannot be recomputed from the supplied 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.
Record for every rate in dispute its numerator, its denominator, the window and timezone it covers, the source it came from and the attribution and deduplication rules behind both sides, then pull the raw components from the merchant's own records, meaning sessions with their session rule, orders with their status filter and the currency applied.
Recompute each rate from those raw components with one numerator and one denominator applied everywhere, express the result with the window and the rules used, and set aside any rate whose components are absent from the supplied records instead of substituting the figure from a dashboard.
Deliver a conversion diagnosis and an experiment backlog in which each rate appears recomputed beside the figures it replaced with the difference explained by the rule that caused it, then reconcile the recomputed order count with the order export and the session count with the visit record for the same window and timezone.
Cover decision rules, frequent misunderstandings, and required inputs so the team aligns on prerequisites before execution.
A rate is trusted only after its numerator, its denominator, the window and timezone, the source and the status filter behind each side are written down. Those items are what let another person reproduce the figure; where one is unknown, the rate is labelled provisional, because a number that cannot be reproduced cannot be compared with anything.
Both sides are recomputed from the same raw records with one attribution rule, one session rule and one window, and the remaining difference is attributed to the rule that changed, such as the denominator or the status filter. Reconciling at the component level shows which choice moved the number, so the merchant can decide which definition to keep.
A rate whose numerator or denominator is absent from the supplied records is reported as not recomputable, with the missing component named, rather than filled from a dashboard figure. Substituting a value from another tool reintroduces the very mismatch being investigated; the rate is returned once the merchant supplies the missing component.
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.