Systems and methods for determining customer lifetime value

1. A system comprising:

one or more processors; and

one or more non-transitory computer-readable media storing computing instructions that, when executed on the one or more processors, cause the one or more processors to perform operations comprising:

storing, for each respective customer of one or more customers, respective customer information in a customer database, wherein the respective customer information comprises one or more interactions for each respective customer of the one or more customers with one or more channels;

predicting, using a machine learning lifetime value (LTV) update model, a respective LTV for each respective customer of the one or more customers based on historical interactions of each respective customer of the one or more customers within a period of time by:

predicting a respective retention probability for each respective customer of the one or more customers using at least logistic regression; and

determining when the respective retention probability for a respective one of the one or more customers is greater than a predetermined threshold value;

determining an online advertisement for each respective customer of the one or more customers using the respective LTV, as predicted, for each respective customer of the one or more customers; and

coordinating displaying the online advertisement for at least a portion of the one or more customers.

2. The system of claim 1 , wherein the machine learning LTV update model comprises a two-stage machine learning LTV update model.

3. The system of claim 1 , wherein determining the online advertisement for each respective customer of the one or more customers comprises:

applying one or more LTV decay functions to the respective LTV, as predicted, for each respective customer of at least a portion of the one or more customers to create a decayed LTV; and

determining the online advertisement based on the decayed LTV.

4. The system of claim 1 , wherein the computing instructions, when executed on the one or more processors, further cause the one or more processors to perform operations comprising:

determining, from the respective customer information in the customer database, whether each respective customer of the one or more customers had:

(1) a respective online store transaction with a retailer;

(2) a respective offline store transaction with the retailer; or

(3) a respective online interaction with the retailer within a first predetermined period of time;

categorizing one or more first customers of the one or more customers into a first customer group when the one or more first customers had:

(1) the respective online store transaction with the retailer;

(2) the respective offline store transaction with the retailer; or

(3) the respective online interaction with the retailer within the first predetermined period of time; and

categorizing one or more second customers of the one or more customers into a second customer group when the one or more second customers did not have:

(1) the respective online store transaction with the retailer;

(2) the respective offline store transaction with the retailer; or

(3) the respective online interaction with the retailer within the first predetermined period of time; and

determining the online advertisement for each respective customer of the one or more customers comprises:

determining the online advertisement for each respective customer of the one or more customers using the respective LTV, as predicted, for each respective customer of the one or more of customers and a respective categorization of each respective customer of the one or more customers.

5. The system of claim 1 , wherein the online advertisement is configured to increase the respective LTV, as predicted, for a respective customer of the one or more customers.

6. The system of claim 1 , wherein the machine learning LTV update model comprises one or more piecewise linear functions or a random forest model.

7. The system of claim 1 , wherein predicting the respective LTV comprises:

predicting a zero GMV value and a number of zero orders for each respective customer of the one or more customers.

8. The system of claim 1 , wherein predicting the respective LTV comprises:

predicting a gross merchandise volume (GMV) value or a number of orders for each respective customer of the one or more customers.

9. The system of claim 1 , wherein the machine learning LTV update model comprises aggregated models of:

a first machine learning LTV update model trained from data within a first predetermined period of time; and

a second machine learning LTV update model trained from historical data within a second predetermined period of time.

10. The system of claim 1 , wherein the computing instructions, when executed on the one or more processors, further cause the one or more processors to perform an operation comprising:

updating the machine learning LTV update model at least once every three months.

11. A method being implemented via execution of computing instructions configured to run on one or more processors and stored at one or more non-transitory computer-readable media, the method comprising:

storing, for each respective customer of one or more customers, respective customer information in a customer database, wherein the respective customer information comprises one or more interactions for each respective customer of the one or more customers with one or more channels;

predicting, using a machine learning lifetime value (LTV) update model, a respective LTV for each respective customer of the one or more customers based on historical interactions of each respective customer of the one or more customers within a period of time by:

predicting a respective retention probability for each respective customer of the one or more customers using at least logistic regression; and

determining when the respective retention probability for a respective one of the one or more customers is greater than a predetermined threshold value;

determining an online advertisement for each respective customer of the one or more customers using the respective LTV, as predicted, for each respective customer of the one or more customers; and

coordinating displaying the online advertisement for at least a portion of the one or more customers.

12. The method of claim 11 , wherein the machine learning LTV update model comprises a two-stage machine learning LTV update model.

13. The method of claim 11 , wherein determining the online advertisement for each respective customer of the one or more customers comprises:

applying one or more LTV decay functions to the respective LTV, as predicted, for each respective customer of at least a portion of the one or more customers to create a decayed LTV; and

determining the online advertisement based on the decayed LTV.

14. The method of claim 11 , wherein:

the method further comprises:

determining, from the respective customer information in the customer database, whether each respective customer of the one or more customers had:

(1) a respective online store transaction with a retailer;

(2) a respective offline store transaction with the retailer; or

(3) a respective online interaction with the retailer within a first predetermined period of time;

categorizing one or more first customers of the one or more customers into a first customer group when the one or more first customers had:

(1) the respective online store transaction with the retailer;

(2) the respective offline store transaction with the retailer; or

(3) the respective online interaction with the retailer within the first predetermined period of time; and

categorizing one or more second customers of the one or more customers into a second customer group when the one or more second customers did not have:

(1) the respective online store transaction with the retailer;

(2) the respective offline store transaction with the retailer; or

(3) the respective online interaction with the retailer within the first predetermined period of time; and

determining the online advertisement for each respective customer of the one or more customers comprises:

determining the online advertisement for each respective customer of the one or more customers using the respective LTV, as predicted, for each respective customer of the one or more of customers and a respective categorization of each respective customer of the one or more customers.

15. The method of claim 11 , wherein the online advertisement is configured to increase the respective LTV, as predicted, for a respective customer of the one or more customers.

16. The method of claim 11 , wherein the machine learning LTV update model comprises one or more piecewise linear functions or a random forest model.

17. The method of claim 11 , wherein predicting the respective LTV comprises:

predicting a zero GMV value and a number of zero orders for each respective customer of the one or more customers.

18. The method of claim 11 , wherein predicting the respective LTV comprises:

predicting a gross merchandise volume (GMV) value or a number of orders for each respective customer of the one or more customers.

19. The method of claim 11 , wherein the machine learning LTV update model comprises aggregated models of:

a first machine learning LTV update model trained from data within a first predetermined period of time; and

a second machine learning LTV update model trained from historical data within a second predetermined period of time.

20. The method of claim 11 , further comprising:

updating the machine learning LTV update model at least once every three months.