Ciwi
Back to topic
Spark playbook scenarios · Spark: AI Store Assistant

Prepare refund decisions from policy and order evidence

Support agents handle each refund request from memory, so similar cases end with different amounts. In this case, the immediate task is to prepare refund decisions from policy and order evidence. The workflow should produce a return or refund decision draft and processing steps, using the merchant's order records, policy text and prior refunds.

Topic
Returns and Refund Automation
Target keyword
shopify refund automation
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 prepare refund decisions from policy and order 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: [the order record with line items and prices, the request reason with claimed returned quantities, the policy text with time and condition limits, and prior refunds on the order].

Check eligibility line by line against the policy, compute the refundable amount from returned quantities and line prices only, hold shipping charges unless the stated policy covers them for that request type, and route unclear cases and amounts above the stated limit to the named approver.
Return a return or refund decision draft and processing steps 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.
Use in Spark
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.

01

Step 1

Gather the order record with line items and prices, the request reason with claimed returned quantities, the policy text with its time and condition limits, and all refunds already issued on that order.

02

Step 2

Check eligibility line by line against the policy, compute the refundable amount from returned quantities and line prices only, hold shipping charges unless the stated policy covers them for that request type, and route unclear cases and amounts above the stated limit to the named approver.

03

Step 3

Deliver the decision draft citing the policy clause applied per line, then verify the proposed amount equals the order totals minus prior refunds before any payment step.

FAQ

Common questions when running this scenario

Cover decision rules, frequent misunderstandings, and required inputs so the team aligns on prerequisites before execution.

How does the draft avoid refunding twice on one order?

Before proposing an amount, the workflow reads every refund already issued on that order and subtracts them from the refundable total, and the check runs ahead of any payment action. A second request for the same line shows what remains, not the original value. The draft always shows prior refunds beside its proposed amount.

How are shipping charges decided?

Shipping is refunded only where the merchant's stated policy covers it for that request type, and the policy clause is cited in the draft. Where the policy is silent, shipping is held for review with the reason recorded, rather than refunded by default or dropped silently. Partial returns follow the same stated rule as full ones.

What sends a case to approval instead of a draft amount?

Requests whose eligibility the policy does not clearly settle, line items outside the return window, and any amount above the merchant's stated approval limit route to the named approver with the evidence attached. The draft still records the policy clauses considered. Clear-cut cases within the limit proceed as drafts, keeping approval for genuine ambiguity.

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.

Open Spark product pageBack to Spark scenario library