Methods, systems, and articles of manufacture for analyzing initiatives for a business network

• US 8,224,686 B1
• Filed: 07/19/2011
• Issued: 07/17/2012
• Est. Priority Date: 01/30/2004
• Status: Expired due to Term

• Current Assignee: Applied Predictive Technologies Inc (MasterCard Incorporated)
• Original Assignee: Applied Predictive Technologies Inc (MasterCard Incorporated)
• Inventors: Bruce, Anthony, Manzi, Jim
• Primary Examiner(s): Boswell, Beth V
• Assistant Examiner(s): Anderson, Folashade
• Application Number:
  US13/185,559
• Time in Patent Office: 364 Days
• Field of Search: 705/7.29, 705/7.26, 705/7.33
• US Class Current: 705/7.29
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• CPC Class Codes: G06Q 30/0201 Market modelling; Market analysis; Collecting market data

Abstract

A system, method, and article of manufacture is disclosed for analyzing a business initiative for a business network including business locations separated into control group sites and test sites that have implemented the business initiative for a predetermined test period. Each of the sites have an associated set of attributes reflecting various characteristics corresponding to the respective site, such as geographical location, size of business location, number of employees, etc. In one aspect of the invention, a process is performed that collects a performance value for each of the test and control group sites reflecting a level of performance of each respective sites during the test period. The performance of the test sites is then measured relative to the performance of the control sites over the same time period. The process may segment the performance values for each test site attribute to identify those attributes that have a greater impact on the performance values of the test sites than other attributes. Further, the process configures a model for predicting the performance values of the test sites using the identified attributes and determines whether the model accurately predicts these performance values. If so, the process applies the model to the non-tested sites to predict the performance values of these sites. Based on the predicted performance values, a user may select one or more of the sites to implement the business initiative.

Claims

1. A method for analyzing a business initiative for a business network including business locations including test sites that have implemented the business initiative during a predetermined test period and non-test group sites that have not implemented the initiative, each of the sites being associated with a set of attributes, the method comprising:

generating test site fragments based on the value of each attribute and performance value for each test site during the predetermined test period for the business initiative;

generating control site fragments based on the value of each attribute and performance value for each control group site during the predetermined test period for the business initiative, where the non-test group sites include a set of control group sites;

determining the strength relationship of the between each attribute and the performance values of the test sites based on an analysis between selected test site fragments and corresponding control site fragments; and

generating, by a server, a list of the attributes ranked based on the strength of the relationship between the attribute and performance value for each control group site.

2. The method of claim 1, further comprising executing a model to the non-test group sites to predict the performance levels of the non-test group sites.

3. The method of claim 2, further comprising ranking a list of the non-test group sites based upon each non-test group site'"'"'s predicted performance level.

4. The method of claim 3, further comprising selecting a subset of the non-test group sites to implement the business initiative based on the ranked list of those sites.

5. The method of claim 2, further comprising:

selecting the model from a list of models provided by the server; and

selecting one or more parameters for the selected model.

6. The method of claim 2, further comprising:

reconfiguring the model with at least one new parameter based on a determination that the model does not accurately predict the performance levels of the test sites; and

instructing the server to execute the reconfigured model for the test sites.

7. The method of claim 6, further comprising:

repeating the reconfiguring and executing the reconfigured model until a quantitative measure reflects that the model accurately predicts the performance levels of the test sites.

8. The method of claim 2, further comprising selecting a number of the ranked attributes that the model should use.

9. The method of claim 2, further comprising receiving results of the model, wherein the results include quantitative measures of the model'"'"'s ability to accurately predict the performance levels of the test sites, wherein the quantitative measures includes a ranked list of selected attributes that the model considered during its execution and data values assigned to each of the selected attributes by the model.

10. The method of claim 9, wherein the data values includes a coefficient data value for a mathematical function used by the model to generate the results.

11. The method of claim 1, wherein the list of the attributes ranked based on each attribute'"'"'s impact on the test site performance values is generated by the server based on comparisons between test site fragments and corresponding control group site fragments.

12. The method of claim 1, further comprising:

instructing the server to execute the model for the test sites; and

receiving results of the executed model, wherein the results include quantitative measures of the model'"'"'s ability to accurately predict the performance levels of the test sites.

13. A computer program product, comprising a non-transitory computer readable media having computer program code therein, said computer program code adapted to cause a computer to perform the following steps of:

analyzing a business initiative for a business network including business locations having sites including test sites that have implemented the business initiative during a predetermined test period and non-test group sites that have not implemented the initiative, each of the sites being associated with a number of attributes, wherein the non-test group sites includes a set of control group sites;

determining, the impact of each of the attributes on performance values of the test sites during the test period by;

segmenting the test sites into fragments based on a selected attribute and the performance value for the test sites during the predetermined test period for the business initiative,

segmenting the control group sites into fragments based on the selected attribute and the performance value for the control group sites during the predetermined test period for the business initiative, and

determining the strength of the relationship between the selected attribute and the test site performance values based on a comparison of the test site fragments and corresponding control group fragments.

14. The computer program product of claim 13, wherein the computer program code further includes program code for configuring a model, based on user input associated with the attributes, that predicts performance values of the test sites in relation to actual performance values of the test sites during the test period.

15. The computer program product of claim 14, wherein the computer program code further includes program code for executing the model for the test sites to produce results reflecting the model'"'"'s ability to accurately predict the performance values of the test sites.

16. The computer program product of claim 14, wherein the computer program code further includes program code for reconfiguring the model based on additional user input associated with the attributes.

17. The computer program product of claim 13, wherein the computer program code further includes program code for providing a user with a list of non-test group sites ranked according to each non-test group site'"'"'s predicted performance value.

18. The computer program product of claim 13, wherein the computer program code further includes program code for executing the model for the test sites includes program code for configuring the model using a subset of the test sites and testing the model'"'"'s ability to accurately predict performance values by predicting the performance values of test sites not included in the subset of test sites.