System and method for assessing segmentation strategies

• US 20070226039A1
• Filed: 03/22/2006
• Published: 09/27/2007
• Est. Priority Date: 03/22/2006
• Status: Active Grant

• Current Assignee: SAS Institute Incorporated
• Original Assignee: SAS Institute Incorporated
• Inventors: Chu, Chengwen, Bulkley, Jack
• Granted Patent: US 8,000,995 B2
• Time in Patent Office: Days
• US Class Current: 705/10
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• CPC Class Codes: G06Q 30/0204 Market segmentation

Abstract

In a novel computer-implemented method and system for assessing segmentation strategies, at least two models are selected for a plurality of segments. Segment performance of the segmentation strategy segments according to selected models is measured. Aggregate segmentation strategy performance data is obtained by aggregating segment performance for each segmentation strategy. Segmentation strategy performance indicia are generated to compare the aggregate segmentation strategy performance data of at least two of the segmentation strategies.

Claims

1. A computer-implemented method for assessing segmentation strategies, comprising:

defining a plurality of segmentation strategies, wherein a segmentation strategy defines a plurality of segments;

selecting a model from a plurality of candidate models for each segment of each segmentation strategy so that at least two models are selected;

measuring segment performance for each segment by its selected model;

aggregating segment performance for each segmentation strategy to obtain aggregate segmentation strategy performance data; and

generating segmentation strategy performance indicia that compare the aggregate segmentation strategy performance data of at least two of the segmentation strategies.

2. The computer-implemented method of claim 1, further comprising:

selecting a first evaluation criterion;

wherein

measuring segment performance for each segment by its selected model comprises measuring segment performance with respect to the first evaluation criterion; and

aggregating segment performance for each segmentation strategy comprises aggregating segment performance for each segmentation strategy per the first evaluation criterion.

3. The method of claim 2, further comprising:

determining whether one of the segmentation strategies exceeds a performance criterion; and

if none of the segmentation strategies exceeds the performance criterion, then;

redefining the plurality of segmentation strategies.

4. The method of claim 2, further comprising:

determining whether one of the segmentation strategies exceeds a performance criterion; and

if none of the segmentation strategies exceeds the performance criterion, then;

selecting a second evaluation criterion;

measuring segment performance with respect to the second evaluation criterion; and

aggregating segment performance for each segmentation strategy per the second evaluation criterion.

5. The method of claim 1, further comprising:

identifying a most predictive variable among the segments.

6. The method of claim 1, wherein:

aggregating segment performance for each segmentation strategy comprises;

sorting segmentation strategy data by probability values associated with the segmentation strategy data to obtain sorted segmentation strategy data; and

binning the sorted segmentation strategy data.

7. The method of claim 1, further comprising:

integrating a decision matrix for the segmentation strategy performance indicia.

8. The method of claim 1, wherein:

each model selected for each segment comprises a champion model for that segment.

9. The method of claim 1, wherein:

the segmentation strategy performance indicia comprise a chart of the aggregate segmentation strategy performance data of the at least two of the segmentation strategies.

10. The method of claim 1, wherein:

the segmentation strategy performance indicia comprise a table of the aggregate segmentation strategy performance data of the at least two of the segmentation strategies.

11. The method of claim 1, wherein:

generating segmentation strategy performance indicia that compare the aggregate segmentation strategy performance data of at least two of the segmentation strategies comprises;

selecting model variables;

selecting model assessment criteria;

determining aggregate segmentation strategy performance data for each of the model assessment criteria for each model variable; and

ranking each segmentation strategy based on each of the model assessment criteria and each model variable.

12. The method of claim 11, further comprising:

determining a best segmentation strategy by generating a cumulative rank for each segmentation strategy, the cumulative rank based on the ranking of each segmentation strategy for each of the model assessment criteria for each model variable.

13. A graphical data display for facilitating analysis of a plurality of segmentation strategies, the graphical data display comprising:

segmentation strategy performance indicia that compare aggregate segmentation strategy performance data of at least two of the segmentation strategies;

wherein;

a segmentation strategy defines a plurality of segments;

segment performance for each segment according to a model is measured, and at least two models are used among the segments; and

aggregate segmentation strategy performance data for each segmentation strategy is based on the segment performance for the plurality of segments of the segmentation strategy.

14. The graphical data display of claim 13, wherein:

the segmentation strategy performance indicia comprise a chart of the aggregate segmentation strategy performance data of the at least two of the segmentation strategies.

15. The graphical data display of claim 14, wherein:

the chart comprises a lift curve for at least two of the segmentation strategies.

16. The graphical data display of claim 13, wherein:

the segmentation strategy performance indicia comprise a table of the aggregate segmentation strategy performance data of the at least two of the segmentation strategies.

17. The graphical data display of claim 13, wherein:

model variables are defined;

model assessment criteria are defined;

aggregate segmentation strategy performance data is based on each of the model assessment criteria for each model variable; and

the segmentation strategies are ranked based on each of the model assessment criteria.

18. A system for assessing segmentation strategies, comprising:

a data store for storing an input data source;

first software instructions stored in a computer-readable medium that upon execution facilitate defining a plurality of segmentation strategies, wherein a segmentation strategy defines a plurality of segments of the input data source;

second software instructions stored in a computer-readable medium that upon execution facilitate selecting a model from a plurality of candidate models for each segment of each segmentation strategy so that at least two models are selected;

third software instructions stored in a computer-readable medium that upon execution measure segment performance for each segment according to selected models for each segment;

fourth software instructions stored in a computer-readable medium that upon execution aggregate segment performance for each segmentation strategy to obtain aggregate segmentation strategy performance data; and

fifth software instructions stored in a computer-readable medium that upon execution generate segmentation strategy performance indicia that compare the aggregate segmentation strategy performance data of at least two of the segmentation strategies.

19. The system of claim 18, wherein:

the segmentation strategy performance indicia comprise a chart of the aggregate segmentation strategy performance data of the at least two of the segmentation strategies.

20. The system of claim 18, wherein:

the fifth software instructions comprise;

software instructions that upon execution facilitate selecting model variables and selecting model assessment criteria;

software instructions that upon execution determine aggregate segmentation strategy performance data for each of the model assessment criteria for each model variable and rank each segmentation strategy based on each of the model assessment criteria and each model variable.