Customer Lifetime Value models

The Customer Lifetime Value model is a ready-to-use data science model that estimates a customer's value over a period of time based on profile and transaction patterns.

The Customer Lifetime Value model explained

The Customer Lifetime Value model predicts a customer's value to a company over a period of time. The lifetime value predictions are based on factors such as customers' history of transactions, the monetary value of those transactions, and the frequency of those transactions. These predictions provide an important metric in determining the cost/benefit or acquisition, retention, and personalized offers. The model can provide lifetime value predictions up to 24 months for a customer segment/profile.

Parameters of the model

When creating the model, you will need to define the following parameters for the model:

  • Algorithm: The algorithm is the piece of code that runs the model.

  • Queries: The queries selected for the model generate a dataset for model training and scoring.

  • Inputs: The inputs are query attributes from the Unity data model that are used for model training and scoring. You can't make changes to model inputs.

  • Outputs: The outputs are data objects and attributes from the Unity data model that are used to store the output values of the model. You can make updates to the default mapping of model outputs.

  • Model inputs

    To generate values, the Customer Lifetime Value model uses the following data.

Attribute from Unity data object Attribute name in the query Unity Data object Description of the attribute Data type Must have?
ID ID MasterCustomer Customer identifier String Yes
CustomerID CustomerID Customer SourceCustomerID String Yes
SourceID MasterCustomer Customer Source Identifier String Yes
SubType SubType OrderItem Type of the order String Yes
OrderEntryTS OrderEntryTS OrderItem The date and time when the order was placed Timestamp Yes
ExtendedPrice ExtendedPrice OrderItem Total amount paid by the customer for their orders Float Yes
OrderID OrderID OrderItem Order identifier String Yes

The following set of Intelligent attributes and Derived attributes are calculated at a customer level and also used as model inputs.

Attribute Attribute name in the query Level of calculation Description of the attribute Data type Must have?
first_purchase_date first_purchase_date MasterCustomer Intelligent Attribute - First purchase date for the customer; (if unavailable, model calculates internally) Timestamp Yes
last_purchase_date last_purchase_date MasterCustomer Intelligent Attribute - Latest purchase date for the customer; (if unavailable, model calculates internally) Timestamp Yes
total_order_amt total_order_amt MasterCustomer Intelligent Attribute - Total of 'ExtendedPrice' for a customer; (if unavailable, model calculates internally) Float Yes
total_order_rows total_order_rows MasterCustomer Intelligent Attribute - Total number of orders by the customer; (if unavailable, model calculates internally) Integer Yes
hist_order_amt hist_order_amt MasterCustomer Derived Attribute - Total value of orders older than 180 days Float Yes
hist_order_rows hist_order_rows MasterCustomer Derived Attribute - Count of orders older than 180 days Integer Yes
target_order_amt target_order_amt MasterCustomer Derived Attribute - Total value of orders between 180 and 90 days in the past Float Yes
target_order_rows target_order_rows MasterCustomer Derived Attribute - Count of orders between 180 and 90 days in the past Integer Yes
test_order_amt test_order_amt MasterCustomer Derived Attribute - Total value of orders from last 90 days Float Yes
test_order_rows test_order_rows MasterCustomer Derived Attribute - Count of orders from last 90 days Integer Yes
order_amt_3_m order_amt_3_m MasterCustomer Derived Attribute - Total value of orders before 90 days Float Yes
order_amt_6_m order_amt_6_m MasterCustomer Derived Attribute - Total value of orders before 180 days Float Yes
order_amt_12_m order_amt_12_m MasterCustomer Derived Attribute - Total value of orders before 360 days Float Yes

Model outputs

You can review values for the following output attributes in the CustomerLifetimeValue data object. You can review the predicted lifetime values and the predicted number of transactions over 3 to 24 month time periods

Attribute ID Attribute Name Attribute Description Data type
SourceCustomerLifetimeValueID Source CustomerLifetimeValue ID This attribute contains the Unique ID for the object or the Primary key. String
MasterCustomerID Master Customer ID This attribute contains the foreign key to the MasterCustomer Table. String
SourceMasterCustomerID Source Master Customer ID This attribute contains the original form of the MasterCustomer ID from the source data system. String
NumTransactions24m Number Transactions For 24 Months This attribute represents the predicted number of transactions over the next 24 months Integer
NumTransactions18m Number Transactions For 18 Months This attribute represents the predicted number of transactions over the next 18 months Integer
NumTransactions12m Number Transactions For 12 Months This attribute represents the predicted number of transactions over the next 12 months Integer
NumTransactions6m Number Transactions For 6 Months This attribute represents the predicted number of transactions over the next 6 months Integer
NumTransactions3m Number Transactions For 3 Months This attribute represents the predicted number of transactions over the next 3 months Integer
AverageMonetary Average Monetary This attribute represents the predicted average monetary transaction. Float
AverageCLV24m AverageCLV for 24 Months This attribute represents the predicted CLV over the next 24 months Float
AverageCLV18m AverageCLV for 18 Months This attribute represents the predicted CLV over the next 18 months Float
AverageCLV12m AverageCLV for 12 Months This attribute represents the predicted CLV over the next 12 months Float
AverageCLV6m AverageCLV for 6 Months This attribute represents the predicted CLV over the next 6 months Float
AverageCLV3m AverageCLV for 3 Months This attribute represents the predicted CLV over the next 3 months Float

data science, data science model, analyze data, create data science model, how to create a data science model, clv, customer lifetime value, clv model, customer lifetime value model