Refund volume is rising overall, but nobody can say whether it is spread evenly or driven by a few items. In this case, the immediate task is to identify products with concentrated refund activity. The workflow should produce a breakdown that weighs each product's refund share against its order share and turns the outliers into an action list.
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 identify products with concentrated refund activity.
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: [refund records with product identifiers and refunded amounts, each product's order share for the same period, the merchant's chosen alert measure, concentration threshold, minimum count and alert owner]. Compute each product's refund share and order share, flag products whose refund share exceeds their order share by the stated threshold and whose absolute counts clear the stated minimum, and treat partial refunds by the one declared rule, either one event or a share of the order.
Return the refund breakdown and action list, then verify the findings against 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.
Collect the refund records with product identifiers and refunded amounts, each product's order share for the same period, the alert measure the merchant chose, counts or value, and the merchant's stated concentration threshold, minimum count and alert owner.
Compute each product's refund share and order share, flag products whose refund share exceeds their order share by the stated threshold and whose absolute counts clear the stated minimum, and treat partial refunds by the one declared rule, either one event or a share of the order.
Deliver the refund breakdown with an alert list naming each flagged product, its two shares and its raw counts, verify each alert by recomputing it from the records, and route the list to the stated owner.
Cover decision rules, frequent misunderstandings, and required inputs so the team aligns on prerequisites before execution.
A product with a large order share naturally carries a large refund share, so raw totals say nothing about concentration. Comparing the two shares isolates the products that refund out of proportion to how much they sell. Products below the stated minimum count are reported with counts, not flagged.
Both shares and the raw counts are recomputed from the records for every flagged product. Alert lists drive actions like delisting or description changes, so a false entry is costly. An alert that fails recomputation is removed before the list is routed.
The product, its refund share, its order share, the raw counts, the period and the measure used. The owner must be able to judge and act without re-pulling the data. Alerts with no stated owner are held rather than broadcast.
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.