01
Step 1
Collect the dated refund events with amounts and product identifiers, the reporting cadence of the refund source, the merchant's stated spike thresholds, the alert owner for each scope, and the baseline periods to compare against.
Back to topic
Spark playbook scenarios · Spark: AI Store Assistant
Detect unusual concentrations of refund activity
Refunds tend to arrive in bursts, and the team cannot tell a real shift from a reporting delay. In this case, the immediate task is to detect unusual concentrations of refund activity. The workflow should produce a refund breakdown and evidence-based action list, tied to the merchant's dated refund events and their agreed thresholds and owners.
Topic
Refund Analysis and Alerts
Target keyword
shopify refund spike alerts
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 detect unusual concentrations of 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: [dated refund events with amounts and product identifiers, the refund source's reporting cadence, the merchant's spike thresholds, alert owners per scope, and the baseline comparison periods]. Aggregate refunds per interval matching the reporting cadence, compare each interval against the matched baseline, require the elevation to persist across consecutive intervals before flagging, and rule out reporting delays by checking whether the following interval shows an offsetting dip.
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
Aggregate refunds per interval matching the reporting cadence, compare each interval against the matched baseline, require the elevation to persist across consecutive intervals before flagging, and rule out reporting delays by checking whether the following interval shows an offsetting dip.
03
Step 3
Deliver the refund breakdown with the alert list and action items, verify each flagged interval's counts against the raw refund events, and route alerts to the stated owner with the baseline, threshold and comparison window named in the message.
FAQ
Common questions when running this scenario
Cover decision rules, frequent misunderstandings, and required inputs so the team aligns on prerequisites before execution.
How is a reporting delay ruled out?
The interval after an apparent spike is checked for an offsetting dip, which is the signature of backfilled events arriving late. A real shift keeps the elevation across intervals. Intervals with a known source outage are excluded from both baseline and current.
How is the baseline chosen for comparison?
Matched periods with a comparable trading pattern, as supplied by the merchant. A mismatched baseline turns normal seasonality into false alarms. Intervals without a valid baseline are reported unjudged rather than flagged.
What must each alert message state?
The scope, the interval, the baseline used, the threshold applied and the persistence check that fired. The owner needs enough to act without re-running the analysis. Alerts missing the baseline reference are held back.
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.