01
Step 1
Gather the refund records with reason codes, free-text notes, associated product identifiers and issue dates, together with the merchant's own reason categories for grouping.
Back to topic
Spark playbook scenarios · Spark: AI Store Assistant
Summarize recurring refund reasons
Refund reason codes and notes are scattered across individual orders, so repeated causes never surface as a pattern. In this case, the immediate task is to summarize recurring refund reasons. The workflow should produce a refund breakdown and evidence-based action list, grouped from the merchant's own refund records with their products and dates.
Topic
Refund Analysis and Alerts
Target keyword
shopify refund reasons report
Copyable AI prompt
A prompt you can drop directly into Spark
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.
ai-prompt.txt
Act as a Shopify operations specialist.
Help me summarize recurring refund reasons.
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 reason codes, free-text notes, associated product identifiers and issue dates, plus the merchant's reason categories]. Group each refund into the merchant's categories using the code where present and the note where absent, count refunds and refunded value per category, and mark unclassified entries as such rather than assigning a reason the records do not state, while checking whether categories concentrate on specific products or periods.
Return a refund breakdown and evidence-based action list 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.How to solve
A recommended resolution workflow
Split the work into input, judgment, and deliverable phases. Verify each step's output before moving to the next.
02
Step 2
Group each refund into the merchant's categories using the code where present and the note where absent, count refunds and refunded value per category, and mark unclassified entries as such rather than assigning a reason the records do not state, while checking whether categories concentrate on specific products or periods.
03
Step 3
Deliver the refund breakdown with an action list, verify that the per-category counts plus the unclassified count sum to the total refund count, and note any category whose concentration depends on a single product or a short period.
FAQ
Common questions when running this scenario
Cover decision rules, frequent misunderstandings, and required inputs so the team aligns on prerequisites before execution.
How are free-text notes grouped without inventing reasons?
A note is mapped to a category only where its wording supports it, otherwise the refund stays unclassified. Assigned reasons would distort the summary toward whatever the analyst assumed. The unclassified count is reported openly, never redistributed across categories.
What does the count reconciliation prove?
The per-category counts plus the unclassified count must sum to the total refund count in the window. Grouping steps can silently drop or double-count records. A gap blocks publication until it is traced and resolved.
How is a reason's concentration checked?
Each category is cross-checked against product and period, because a category can reflect one bad batch rather than a broad cause. A single-source concentration is labelled as such in the report. Only labelled concentrations become action items.
Related scenarios
Continue with adjacent or same-topic scenarios
Start with scenarios under the same topic, then expand into cross-topic matches suggested by similarity.
Next step
Want these AI scenarios embedded directly into daily operations?
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.