Predicting marketing campaigns having more than one step

• US 8,521,579 B2
• Filed: 05/27/2003
• Issued: 08/27/2013
• Est. Priority Date: 04/14/2003
• Status: Active Grant

• Current Assignee: SAP SE
• Original Assignee: SAP AG (SAP SE)
• Inventors: Witting, Thomas
• Primary Examiner(s): Choi, Peter
• Assistant Examiner(s): CHONG CRUZ, NADJA N
• Application Number:
  US10/445,722
• Publication Number:
  US 20040204982A1
• Time in Patent Office: 3,745 Days
• Field of Search: 705 7- 12
• US Class Current: 705/7.31
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• CPC Class Codes: G06Q 30/0202 Market predictions or forecasting for commercial activities
• CPC Class Codes: G06Q 30/0204 Market segmentation
• CPC Class Codes: G06Q 40/00 Finance; Insurance; Tax strategies; Processing of corporate or income taxes

Abstract

Methods and systems for assigning customers to steps in marketing campaigns. An assignment module assigns and reassigns the customers to the marketing activities. An evaluation module determines a predicted goal value of the marketing campaign for each assignment. The assignment module does not reassign a customer to a marketing activity that the customer has previously been assigned to. An execution module may execute campaign steps by performing marketing activities, and a response detection module may detect responses from the customers. The responses may be used in determining subsequent steps of the campaign. Creating sample target groups representative of the customers that are predicted to respond to a prior campaign step. The sample target group(s) may be used for predicting the outcome of subsequent campaign steps directed at the sample target group.

Claims

1. A method of predicting outcomes of marketing campaigns comprising more than one campaign step, by a computer system, the method comprising:

defining, by the computer system, a marketing campaign as including at least first and second campaign steps, the second campaign step configured to be performed after the first campaign step and only toward customers who responded to the first campaign step;

compiling first customer objects in a repository, the first customer objects corresponding to a first target group of customers registered in the computer system, at least one constraint taken into account when compiling the customer objects;

predicting, by the computer system and the first customer objects, a number of responses to be received if the first campaign step were performed toward a first target group of customers, each customer having a response probability;

compiling, by the computer system, second customer objects from the first customer objects, the second customer objects corresponding to a second target group selected from the first target group using respective response probabilities of individual customers in the first target group, the second target group being a subset of the first target group and substantially equal to the predicted number of responses;

predicting, by the computer system and the second customer objects, an outcome of performing the second campaign step toward the second target group; and

generating, by the computer system, a prediction for the marketing campaign using at least the predictions regarding the first and second campaign steps.

2. The method of claim 1, wherein the response probabilities are numbers within a range, and wherein selecting the second target group comprises:

randomly selecting a number in the range for a customer in the first target group;

comparing the randomly selected number to the customer'"'"'s response probability and including the customer in the second target group if the randomly selected number is less than the customer'"'"'s response probability; and

performing the selecting and comparing steps for all customers in the first target group.

3. The method of claim 2, wherein the range is 0 to 1.

4. The method of claim 1, wherein the second target group initially is not equal to the predicted number of responses, further comprising adjusting the second target group to be equal to the predicted number of responses.

5. The method of claim 4, wherein the second target group initially is smaller than the predicted number of responses and wherein adjusting the second target group comprises:

sorting the customers in the first target group that were not included in the second target group according to their response probabilities; and

adding to the second target group those customers having highest response probability so that the second target group becomes equal to the predicted number of responses.

6. The method of claim 4, wherein the second target group initially is greater than the predicted number of responses and wherein adjusting the second target group comprises:

sorting the customers in the second target group according to their response probabilities; and

removing from the second target group those customers having lowest response probability so that the second target group becomes equal to the predicted number of responses.

7. The method of claim 1, wherein the second campaign step comprises a plurality of alternative campaign elements, further comprising assigning the customers to the campaign elements using an optimizing algorithm.

8. The method of claim 7, wherein the optimizing algorithm assigns and reassigns the customers to the campaign elements while evaluating a predicted value of the second campaign step, but does not reassign a customer to a campaign element to which the customer has previously been assigned.

9. The method of claim 1, wherein determining the outcome of performing the second campaign step comprises predicting a number of responses to be received if the second campaign step were performed toward the second target group, further comprising:

selecting a third target group from the second target group using the response probabilities, the third target group being equal to the predicted number of responses for the second campaign step; and

predicting an outcome of performing a third campaign step toward the third target group.

10. The method of claim 1, wherein predicting the outcome of performing the second campaign step for at least one customer comprises using one selected from the group consisting of a predicted revenue per response, a predicted profit per response, a contact cost, a response cost, and combinations thereof, each being determined individually for the customer.

11. A system for predicting outcomes of marketing campaigns comprising more than one campaign step, the system comprising:

a processor;

program instructions tangibly embodied in a computer-readable medium and comprising a response prediction module that, when executed by the processor defines a marketing campaign as including at least first and second campaign steps, the second campaign step configured to be performed after the first campaign step and only toward customers who responded to the first campaign step, the second campaign step configured to be performed after the first campaign step and only toward customer who responded to the first campaign step;

program instructions tangibly embodied in a computer-readable medium and comprising an assignment module that, when executed by the processor, compiles first customer objects in a repository, the first customer objects corresponding to a first target group of customers registered in the computer system, at least one constraint taken into account when compiling the customer objects;

program instructions tangibly, embodied in a computer-readable medium and comprising a response prediction module that, when executed by the processor, predicts, using the first costumer objects, a number of responses to be received if the first campaign step were performed toward the first target group of customers, each customer having a response probability;

program instructions tangibly embodied in a computer-readable medium and comprising an assignment module that, when executed by the processor, compiles second customer objects from the first customer objects, the second customer objects corresponding to a second target group selected from the first target group using respective response probabilities of individual customers in the first target group, the second target group being a subset of the first target group and substantially equal to the predicted number of responses;

program instructions tangibly embodied in a computer-readable medium and comprising an evaluation module that, when executed by the processor, predicts, using the second customer objects, an outcome of performing a second campaign step toward the second target group; and

program instructions tangibly embodied in a computer-readable medium and comprising an evaluation module that, when executed by the processor, generates a prediction for the marketing campaign using at least the predictions regarding the first and second campaign steps.

12. The system of claim 11, wherein the response probabilities are numbers within a range and wherein the assignment module comprises a random-number generator capable of randomly selecting a number in the range for a customer in the first target group, wherein the assignment module compares the randomly selected number to the customer'"'"'s response probability and includes the customer in the second target group if the randomly selected number is less than the customer'"'"'s response probability, and wherein the assignment modules performs the selecting and comparing steps for all customers in the first target group.

13. The system of claim 11, wherein the response prediction module predicts a number of responses to be received if the second campaign step were performed toward the second target group, wherein the assignment module further selects a third target group from the second target group using the response probabilities, the second target group being equal to the predicted number of responses for the second campaign step, and wherein the evaluation module further predicts an outcome of performing a third campaign step toward the third target group.

14. The system of claim 11, wherein the evaluation module predicts the outcome of performing the second campaign step for at least one customer using one selected from the group consisting of a predicted revenue per response, a predicted profit per response, a contact cost, a response cost, and combinations thereof, each being determined individually for the customer.

15. Computer software, tangibly embodied in a non-transitory computer-readable medium, for predicting outcomes of marketing campaigns comprising more than one campaign step, the software comprising instructions, when executed by a processor, to perform operations comprising:

define a marketing campaign as including at least first and second campaign steps, the second campaign step configured to be performed after the first campaign step and only toward customers who responded to the first campaign step;

compile first customer objects in a repository, the first customer objects corresponding to a first target group of customers registered in the computer system, at least one constraint taken into account when compiling the customer objects;

predict, using the first customer objects, a number of responses to be received if the first campaign step were performed toward a first target group of customers, each customer having a response probability;

compile second customer objects from the first customer objects, the second customer objects corresponding to a second target group selected from the first target group using respective response probabilities of individual customers in the first target group, the second target group being a subset of the first target group and substantially equal to the predicted number of responses;

predict the second customer objects, an outcome of performing the second campaign step toward the second target group; and

generate a prediction for the marketing campaign using at least the predictions regarding the first and second campaign steps.

16. The software of claim 15, wherein the second target group initially is not equal to the predicted number of responses, further comprising:

adjust the second target group to be equal to the predicted number of responses.

17. The software of claim 16, wherein the second target group initially is smaller than the predicted number of responses and wherein adjusting the second target group comprises:

sort the customers in the first target group that were not included in the second target group according to their response probabilities; and

add to the second target group those customers having highest response probability so that the second target group becomes equal to the predicted number of responses.

18. The software of claim 16, wherein the second target group initially is greater than the predicted number of responses and wherein adjusting the second target group comprises:

sort the customers in the second target group according to their response probabilities; and

remove from the second target group those customers having lowest response probability so that the second target group becomes equal to the predicted number of responses.