Skip to content

Returns and Refund Processor

Example prompt: "When a customer emails us about a return, check their order in Shopify, log the return request in our Google Sheet, and draft a reply based on our returns policy."

How to automate returns processing with GloriaMundo

The Problem

Handling return requests is one of the most time-consuming parts of running an online shop. Each request requires checking the original order details, verifying it falls within the returns window, looking up the returns policy for that product category, drafting a response, and logging the request somewhere the team can track it. For shops handling 10-20 return requests per week, this easily adds up to several hours of repetitive work — and slow responses lead to frustrated customers and negative reviews.

How GloriaMundo Solves It

We build a workflow that triggers when a return-related email arrives. An LLM step analyses the email to extract the order number, the product in question, and the reason for the return. An integration step looks up the order in Shopify to confirm the purchase date, items, and order value. A conditional step checks whether the request is within your returns window (e.g. 30 days). An LLM step drafts a reply to the customer — approving the return with instructions if it qualifies, or politely explaining why it falls outside the policy if it does not. An integration step logs the return request in a Google Sheet with the order number, product, reason, and status. Glass Box preview shows you the drafted reply and the Sheet entry before anything is sent, so you can adjust the response or override the policy decision.

Example Workflow Steps

  1. Trigger (integration): Fires when an email containing return-related keywords arrives in the support inbox.
  2. Step 1 (llm): Analyse the email to extract the order number, product name, and return reason.
  3. Step 2 (integration): Look up the order in Shopify to confirm purchase date, items, and value.
  4. Step 3 (conditional): Check whether the return request falls within the 30-day returns window.
  5. Step 4 (llm): Draft a customer reply — approve with return instructions, or explain why the request is outside policy.
  6. Step 5 (integration): Log the return request in a Google Sheet with order details, reason, and approval status.
  7. Step 6 (integration): Send the drafted reply to the customer via Gmail.

Integrations Used

  • Gmail — receives incoming return requests and sends drafted replies
  • Shopify — source of order details to verify purchase date and eligibility
  • Google Sheets — logs all return requests for tracking and reporting

Who This Is For

E-commerce store owners and customer service teams handling a steady volume of return requests who want to respond faster and more consistently without hiring additional support staff. Works well for shops processing 5-30 returns per week in which the policy is relatively straightforward.

Time & Cost Saved

Each return request typically takes 10-15 minutes to process manually — checking the order, reviewing the policy, drafting a reply, and logging it. At 15 returns per week, that is roughly 3 hours of work. This workflow handles the entire process in seconds, with human review only needed for edge cases. Response times drop from hours or days to minutes, which directly reduces the chance of a frustrated customer leaving a negative review. The workflow uses integration, LLM, and conditional steps, costing a few credits per return processed.