/
/
Know a WooCommerce customer's value and risk before you talk
For sales and service teams who want a WooCommerce customer's full order history summarized by their AI assistant.
On this page
Connector
Connect WooCommerce, pick your AI tool and paste the prompt.
The problem
Talking to a customer without knowing their history
Understanding a WooCommerce customer means pulling orders, what they buy, refunds, reviews and downloads, and some orders may sit under a different billing email. Building that picture by hand takes a long time.
Without PopdockAI
- Search orders by email and by account
- Work out spend and top products
- Check refunds and reviews separately
- Miss orders under another billing email
With PopdockAI
- The customer is found, or noted as guest only
- Orders, lines, refunds, reviews and downloads are read
- Value, top products and refund rate are worked out
- Risk flags are raised, and nothing is changed
The prompt
Use this prompt
Copy it into any AI tool connected to PopdockAI. Swap the values in braces for your own.
**Objective:** Build a complete account review for the customer with email `{customer-email}`, looking back `{lookback-days}` days (default 365).
**Step 1: Find the customer**
Call `woocommerce-get-customer` with `email` = `{customer-email}`. Keep `customer-id`, name, `created-date`, `role`, `paying-customer`, billing city, state and country. If nothing matches, the person may only have guest orders: say so and continue with the email-based reads below.
**Step 2: Pull their activity**
Work out the start date as today minus the lookback. Then call:
* `woocommerce-get-order` with `customer-email` and `created-date.from`. If a `customer-id` was found, also call it with `customer-id` and merge by `order-id`, because a registered customer can order under a different billing email.
* `woocommerce-get-order-line` with `customer-email` and `date-created.from`, for what they buy.
* `woocommerce-get-refund` with `customer-email` and `created-date.from`.
* `woocommerce-get-product-review` with `reviewer-email` = `{customer-email}`.
* `woocommerce-get-customer-download` with `customer-email`, only if any line looks like a digital product.
**Step 3: Work out the numbers**
Treat orders with status `completed`, `processing`, `on-hold` and `refunded` as sales. Exclude `pending`, `failed`, `cancelled` and `trash` orders from every total.
* Order count and value, split by status; average order value; first and last order date; days since the last order.
* Top 5 products by quantity, grouped by `line-sku` (fall back to `line-name`).
* Refunds: count, total, and refund rate as refunded amount / sales value.
* Reviews: count, average `rating`, any rated 2 or below.
**Step 4: Report**
1. Profile: name, email, customer id or "guest", registered since, location, role.
2. Summary table: orders, sales value, AOV, refunds, refund rate, net value, last order.
3. Orders table, newest first: number, date, status, total.
4. Top products.
5. Reviews and downloads (downloads with `downloads-remaining` 0 or a past `access-expires`).
6. Flags, each with the evidence: refund rate over 20%, two or more `failed` orders, orders stuck in `on-hold` or `pending`, low reviews, or no order in over 180 days.
**Rules & Guidelines:**
* This prompt is read-only. Do not call any write tool: `woocommerce-add-order`, `woocommerce-update-order`, `woocommerce-add-order-note`, `woocommerce-send-order-details`, `woocommerce-add-refund`, `woocommerce-add-customer`, `woocommerce-update-customer`, `woocommerce-add-product`, `woocommerce-update-product`, `woocommerce-update-product-variation`, `woocommerce-add-coupon`, `woocommerce-update-coupon`.
* Pass `max-records` explicitly (500 unless stated otherwise). If a read returns exactly that many rows, say the result may be truncated and that totals are a lower bound.
* If the rows carry more than one `currency`, report each currency separately. Never add amounts across currencies.
**Validation:**
* If there is no customer record and no orders for the email, say so and stop.
* If `{lookback-days}` is empty or not a number, use 365.Who it's for
One answer, three teams that use it
Sales reps
Prepare for a customer call
Know order count, value, top products and the last order date before you speak.
Give me an account review for priya@fabrikam.com in WooCommerce for the last 6 months.
Customer service agents
Spot a risky account
High refund rates, failed orders and low reviews are flagged before you make an exception.
Does tom@contoso.com have a high refund rate or failed orders in WooCommerce?
Account managers
Find customers going quiet
A long gap since the last order is flagged, so you can reach out in time.
When did orders@northwindtraders.com last order in WooCommerce, and what do they usually buy?
How it works
What PopdockAI does when you ask
You give a customer email and a lookback, and your AI assistant builds their profile from real order history. It is read-only and keeps currencies separate.
01
Ask
You give a customer email and a lookback period.
02
Find
It finds the customer or notes they order only as a guest.
03
Read
It reads orders, lines, refunds, reviews and downloads, counting real sales.
04
Report
You get a profile, the numbers and any flags, read-only.
What PopdockAI provides
The tools behind the answer
Ready made tools for WooCommerce and one secure connection to your AI tool. No code, no exports.
woocommerce-get-customer
woocommerce-get-order
woocommerce-get-order-line
woocommerce-get-refund
woocommerce-get-product-review
woocommerce-get-customer-download
This report is built from Get data tools only. PopdockAI applies the row, field and parameter rules your admin sets, and every query is written to the audit trail.
The outcome
A customer's value and risk before you talk
Related use cases
More ways to put this data to work
One prompt template with infinite possibilities.
Your AI assistant adapts its approach, tone, and capabilities to match each unique context while you maintain a single, manageable workflow.
Join forward-thinking companies that are transforming their operations with intelligent, secure AI automation.

.webp)
