Customers that bought a specific product

How to use the Metrics API to get the total number of customers that bought a specific product or bundle

The problem

You want to get the total number of customers that bought a specific SKU (or bundle — identified by their code) from your catalog, over a selected date and time range.

The solution

Query

You need to perform a stats query setting the required query keys as follows and adding the optional ones based on your needs:

Key	Value
field	customer.email
operator	cardinality

Since the query is on the order resource and each order is associated with a customer, to get the total number of unique customers you need to use the cardinality operator. Using value_count would lead to wrong results, counting one different customer for each order (e.g. let's assume you have 2 customers that placed 3 and 7 orders — the cardinality computation would be 2, while the value_count one would be 10). Please remember that the cardinality operator gives just estimated results.

Filter

Set the desired date and time range using the date_from and date_to keys and add an additional filter on the line items field to restrict the related computation to the selected SKU or bundle code only:

Attribute	Operator
codes	"in": [ "TSHIRT01" ]

As shown in the example below, use placed_at as the date_field in the date filter if you want the results to count all the orders that were placed in the selected date and time range (read more about this).

Try it on Postman 🚀

Example

Request

The following request uses the Metrics API to get the total number of customers that bought a specific SKU:

curl -g -X POST \
  'https://{{your_domain}}.commercelayer.io/metrics/orders/stats' \
  -H 'Accept: application/vnd.api.v1+json' \
  -H 'Content-Type: application/vnd.api+json' \
  -H 'Authorization: Bearer {{your_access_token}}' \
  -d '{
    "stats": {
      "field": "customer.email",
      "operator": "cardinality"
    },
    "filter": {
      "order": {
        "date_from": "2021-01-01T00:00:00Z",
        "date_to": "2021-12-31T23:59:00Z",
        "date_field": "placed_at"
      },
      "line_items": {
        "codes": {
          "in": [ "TSHIRT01" ]
        }
      }
    }
  }'

Response

On success, the API responds with a 200 OK status code, returning the computed value in the data object and additional information in the meta object:

{
  "data": {
    "value": 123
  },
  "meta": {
    "type": "stats",
    "trace_id": "fe571ea2-8a4f-4a5e-bd26-ac54651bb2e4",
    "mode": "test",
    "organization_id": "xYZkjABcde",
    "market_ids": [ "yzXKjYzaCx", "..." ]
  }
}

Similar cases

Just changing a couple of query keys and/or filter parameters you can address lots of very similar use cases, such as:

• Total number of orders over a specific period of time

• Total number of orders paid with a coupon

• Total number of orders paid with a gift card

• Total number of orders paid with a specific coupon code

• Total number of orders that triggered a specific promotion

• Total number of orders that took more than 5 minutes to be placed

• Total number of refunded orders

• Total number of returned orders

• Total number of orders to which a discount was applied

• Total orders with a personalization option

• Mixed stats about the number of products per order

• Total number of orders containing a specific list of products

• Total number of orders with amounts greater than $100

• Total number of customers over a specific period of time

• Mixed stats about returns that contain more than one product

• Total number of pending carts with more than one product

• Total amount lost due to potentially abandoned carts over a specific period of time

• Total number of pending carts with amount greater than 100€

• Shipments average time in picking