All data that Predictable ingests to train models and score the target population.

Known Customers

All customers in client universe

COLUMN NAME TYPE DEFINITION
CUSTOMER_ID VARCHAR unique customer identifier

Transaction Events

All transaction events with relevant discount and spend data. This table should be at the transaction level – one row per transaction. For transactions with multiple products purchased, pass product SKUs in an array in the product column.

COLUMN NAME TYPE DEFINITION
CUSTOMER_ID VARCHAR unique customer identifier
PURCHASE_DATETIME NUMBER unix timestamp in seconds of when transaction event occurred
DISCOUNT_CODE VARCHAR discount code, if anything, used on purchase
SPEND FLOAT total amount of spend in a transaction
PRODUCT_COLUMN VARCHAR comma separated array of each product purchased in a specific transaction

Messaging Engagement Events

Email and SMS event engagement.

COLUMN NAME TYPE DEFINITION
CUSTOMER_ID VARCHAR unique customer identifier
DATE_EMAIL_SENT NUMBER unix timestamp in seconds of when message was sent
CAMPAIGN_NAME VARCHAR name of the campaign associated with the message
CLICKED_FLAG INT 1=message was clicked, 0 message was not clicked
UNSUB_FLAG INT 1=indicates an unsubscribe event, 0=all other events
MESSAGE_TYPE VARCHAR type of message sent (SMS, email, etc.)

Web / Other Engagement Events

Site, app, and all other engagement types. This data source is optional but will maximize the predictive power of the models if available.

COLUMN NAME TYPE DEFINITION
CUSTOMER_ID VARCHAR unique customer identifier
DATETIME NUMBER unix timestamp in seconds of when interaction occured
INTERACTION_TYPE VARCHAR type of interaction (page view, add to cart, etc.)
UTM_CAMPAIGN VARCHAR UTM campaign parameter associated with event
UTM_MEDIUM VARCHAR UTM medium parameter associated with event
UTM_SOURCE VARCHAR UTM source parameter associated with event