Revenue moved between two periods and the merchant cannot tell whether fewer orders, lower prices or a different mix caused it. In this case, the immediate task is to explain sales changes using supplied business evidence. The workflow should produce an explanation with the contributing factors ranked and the evidence behind each one.
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 explain sales changes using supplied business evidence.
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: [per-period order count, revenue before and after refunds, refunds mapped to their original order period, discount amounts with campaign tag, units and realized price per product, shipping and tax treatment differences, the promotional calendar, and the merchant's stated materiality threshold]. Decompose the revenue change into order volume, average order value split further into units per order and realized price per unit, and product mix, testing each against the supplied records and marking as noise rather than a driver any factor smaller than the merchant's stated materiality threshold.
Return the explanation with the contributing factors ranked and the evidence supporting each one 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.
Assemble for each period compared the order count, revenue before and after refunds, refunds mapped to the period that carried the original order, discount amounts with the campaign they belong to, units and realized price per product, any difference in how shipping and tax were treated, and the promotional calendar the merchant supplies.
Decompose the revenue change into order volume, average order value split further into units per order and realized price per unit, and product mix, testing each against the supplied records and marking as noise rather than a driver any factor smaller than the merchant's stated materiality threshold.
Deliver the driver list ranked by contribution with the arithmetic behind each factor, check that the volume, price and mix effects sum back to the observed total change, and list what the exports could not rebuild, such as discounts missing a campaign tag.
Cover decision rules, frequent misunderstandings, and required inputs so the team aligns on prerequisites before execution.
Attach each refund to the period carrying the original order, not the month it was issued. Using the issue date lets one period absorb the other's cancellations, which then reads as a demand decline that never actually happened.
Only once volume and price effects are separated. Recompute realized price per unit after discounts against the prior period; if units rose while realized price fell, the answer is promotion and mix, not weakening demand.
Payday cycles, weekends and campaign dates move volume independently of store performance, so window lengths and their calendar positions are agreed first; any mismatch between the two periods is recorded as a stated limitation of the analysis.
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.