# Sell slow stock to likely buyers before discounting

> Before marking down slow stock, tell the people who bought that style before. Connect Shopify and Klaviyo, ask Claude to list products with more than six months of stock and the cash tied up, and find subscribed past buyers of the same style, color or category. Email that narrow group first, then discount only what is left.

Type: Play (a chance to grow)
Run it: Quarterly
Source: https://gooseworks.ai/playbooks/sell-slow-moving-stock-before-discounting

## 1. Connect

Give your AI assistant read-only access to Shopify and Klaviyo.

## 2. Ask

```text
List products with more than 6 months of stock at the current sales rate, with the cash tied up in each. For each, find past customers who bought the same style, colour or category, and how many are still subscribed.
```

## 3. Act

Email that narrow group first, at full price or a small offer. Discount only what's left.

## Steps

1. Connect Shopify and Klaviyo read-only through a Claude connector or MCP server, or export inventory, orders and subscriber lists and upload them.
2. Ask the playbook prompt to list slow stock with cash tied up and the subscribed past buyers who match each product.
3. Check stock counts are current and that the products are not slow because of a fault or bad reviews.
4. Email the narrow segment first at full price or a small offer, holding back a slice to compare.
5. Count orders against the held-back slice, discount only what remains, and re-run each quarter.

## What it tends to find

$40K sat in slow colours. A few thousand past buyers of that colour family had never been told. (Illustrative example, not a result from a named store.)

## Watch out

Send to a narrow segment first, and count orders against a held-back slice.

## FAQ

### How do I know which stock is slow?

Divide units on hand by your recent sales rate. If it would take more than six months to sell at that pace, it is slow. Claude can do this per product from your inventory and orders.

### Why email past buyers before discounting?

People who bought the same style or color already like it, so some will buy at full price. A markdown gives that margin away to everyone, including buyers who would have paid more.

### How do I measure whether the email worked?

Hold back a random slice of the matched segment. Compare orders from those who got the email with those who did not over the same days.

### Can I use this prompt with Muse or ChatGPT?

Yes. Any assistant that can read your inventory, orders and email lists can run it, either connected to the tools or from files you export.