# Turn last week's Loop returns into a merchandising fix list

> Every Monday, see which products are driving preventable returns, why shoppers sent them back, and the one change that will fix each one.

- Workflow type: agent
- Services: Loop Returns, Google Sheets, Slack
- Categories: Operations, Marketing
- Published: 2026-08-07

## What it does

- Pulls the last seven days of returns from Loop Returns every Monday morning and groups them by product.
- Separates preventable returns, the ones caused by sizing, quality, or a product page that oversold, from the normal returns your bestsellers always get.
- Recommends one specific fix per flagged product, such as adding a sizing chart, reshooting photos, or rewriting a misleading description.
- Logs every flagged product to a Google Sheet so you can watch the trend week over week, and posts the top five to Slack.

## What you'll need

- A Loop Returns account with access to your returns and product data
- A Google account and a spreadsheet to use as the weekly tracking log, with a header row already in place
- A Slack workspace and the channel where your merchandising team wants the brief
- Enough weekly return volume for patterns to show up, generally a few dozen returns a week

## Prompt

Every Monday at 7am, review the last seven days of returns for my store and produce a merchandising brief that tells the team which product pages to fix. Start by pulling returns from Loop Returns with the Detailed Returns List action, setting the from and to range to cover the previous seven full days. Pass paginate=true and follow nextPageUrl until it comes back null so you have the complete week, because a week of returns will not fit on one page for most brands. Loop caps the from and to range at 120 days, so a seven day window is comfortably inside the limit.

Group every returned line item by product. For each product, count total returns and units returned, then sort the reason the shopper gave into preventable and not preventable. Treat sizing and fit, quality and defects, and not as described or different from the photos as preventable, since a merchandising change can actually move those. Treat changed my mind, no longer needed, found a better price, and arrived too late as not preventable. Reason wording varies by store, so match on meaning rather than exact strings, and put anything you cannot confidently classify in an unclear bucket instead of forcing it into one side.

Rank the products two ways: by raw return volume, and by the share of their returns citing a preventable reason. Work out the store wide preventable share for the week and use it as the baseline. Only flag a product when it had at least five returns in the window and its preventable share is either at least half of its returns or at least 1.5 times the store wide baseline. Deliberately skip products whose returns are just normal volume for a bestseller, meaning high raw counts with a preventable share at or below the baseline, and say in the brief that you skipped them. The point of this brief is preventable returns, not a leaderboard of popular products.

For each flagged product, enrich it with the Loop Returns List Products action to pull the title, description, images, and options, and use List Product Variants when you need the size or colorway breakdown. Use that detail to recommend exactly one concrete fix per product, tied to its dominant reason. Fit complaints concentrated in particular sizes point to a sizing chart update or a runs small note near the size selector. Not as described complaints point to better photography, a color or material correction, or clearer copy. Quality and defect complaints are not a page problem, so flag them for supplier or QA follow up and say that plainly. Be specific: name the size, the colorway, or the sentence in the description that is misleading, and skip generic advice like improve the product page.

Append one row per flagged product to my returns tracking sheet in Google Sheets using Append Values, so the team can watch the trend week over week. Use these columns in order: week ending date, product title, product id, total returns, units returned, preventable share as a percentage, top preventable reason, recommended fix, and a blank status column for the team to fill in. The sheet already has a header row, so append after the last row and never overwrite existing rows.

Then post the brief to my merchandising channel in Slack with Send a Message. Open with one line covering total returns for the week, the store wide preventable share, and how that compares to previous weeks, which you can read from the tracking sheet with Get Values. Then list the top five flagged products, each with its return count, preventable share, dominant reason, and the single recommended fix, in a short scannable format. Note any bestsellers you skipped and why, then link the tracking sheet at the end. Keep the whole message to roughly 20 lines so it reads in the channel without expanding.

If no product clears the bar, do not append any rows. Post a short Slack note saying returns looked normal this week and include the store wide preventable share so the team still sees the trend. Loop allows 300 requests per minute per API key, so pace the pagination and enrichment calls, and if a call comes back with a 429, pause briefly and retry rather than dropping part of the week.

## How to customize

- Change the timing. Monday at 7am is the default, but any weekly or daily schedule works.
- Adjust the bar for flagging a product, such as the minimum number of returns it needs or how far above your store average its preventable share has to be.
- Choose which return reasons count as preventable, and point the brief at a different Slack channel or tracking spreadsheet.

## Example output

Returns for Aug 3 to Aug 9: 412 returns, 58% preventable (up from 51% last week)

1. Linen Wide Leg Pant, 34 returns, 79% preventable
Top reason: runs small, concentrated in sizes 6 and 8
Fix: add a sizing chart with real garment measurements plus a runs small note near the size selector

2. Ribbed Knit Midi Dress, 28 returns, 71% preventable
Top reason: color not as described, shoppers expected the sage in the hero shot
Fix: reshoot the sage colorway in daylight and add a true to color swatch image

3. Everyday Leather Tote, 19 returns, 68% preventable
Top reason: smaller than expected
Fix: add a scale photo of the bag worn on a model and put dimensions in the first bullet

Skipped: Classic Cotton Tee (61 returns, 24% preventable, normal for volume)

Full list logged to the returns tracker sheet.

## FAQ

### How does it know which returns were preventable?

It reads the reason each shopper gave when they started the return and sorts them into groups. Returns about sizing and fit, quality or defects, and items not matching the photos or description are treated as preventable, because a change to the product page or the product itself can reduce them. Change of mind, found a better price, and arrived too late are excluded, since no merchandising fix will move those.

### Will it flag my bestsellers just because they get a lot of returns?

No. High return counts on their own are not enough. Each product's share of preventable reasons is compared against your store average for the week, so a popular item with an ordinary mix of reasons is skipped and a lower volume item with a bad fit problem still gets flagged.

### What happens in a quiet week when nothing looks broken?

It posts a short note to Slack saying returns looked normal, along with your store wide preventable share so the team still sees the trend, and it does not add rows to the tracking sheet.

### Do we need a separate returns analytics tool for this?

No. It reads your returns and product details straight from Loop Returns, so there is nothing to export and no extra reporting subscription to buy.

### Why log to a spreadsheet if the brief already goes to Slack?

Slack tells you what to fix this week. The spreadsheet is the memory, so after a month you can see whether the sizing chart you added actually brought that product's returns down, and which fixes are still sitting untouched.

Use this prompt in General Input: https://www.generalinput.com/prompts/turn-last-weeks-loop-returns-into-a-merchandising-fix-list