Predictive models

A predictive model is a feature of AmpIQ that predict customer behavior, such as predicted customer lifetime value (predicted CLV), churn propensity, product affinity, and lifecycle events.

Note

Predictive models require at least four years of historical data (five years or more is recommended). You should have at least 100,000 customer transactions on an annual basis, with at least a 10% retention rate.

The following tables must be configured in your customer 360 database prior to running predictive models:

1. Customer 360

2. Merged Customers

3. Transaction Attributes

4. Transaction Attributes Extended

5. Unified Itemized Transactions (updated to include product catalogs)

6. Unified Transactions

Common inputs to models

Columns from the Merged Customers, Unified Transactions, and Unified Itemized Transactions (including product catalogs) are used as inputs to predictive modeling. For multi-brand tenants, the amperity_id column from the Customer 360 table is also used as an input to predictive modeling.

The following columns are common inputs to predictive models:

Source table	Input columns
Merged Customers	The following columns are common inputs: amperity_id birthdate city email gender given_name phone postal state surname
Unified Itemized Transactions	The following columns are common inputs: is_return item_quantity item_revenue order_datetime order_id product_id
Unified Transactions	The following columns are common inputs: amperity_id order_canceled_quantity order_canceled_revenue order_datetime order_discount_amount order_id order_quantity order_returned_quantity order_returned_revenue order_revenue purchase_brand purchase_channel store_id

Available models

The following predictive models can be enabled for your tenant:

• Churn propensity

• Customer lifetime value

• Product affinity