Highly scalable internet-based randomized experiment methods and apparatus for obtaining insights from test promotion results

• US 10,176,491 B2
• Filed: 03/03/2015
• Issued: 01/08/2019
• Est. Priority Date: 03/13/2013
• Status: Active Grant

• Current Assignee: Maplebear Inc.
• Original Assignee: Eversight, Inc. (Maplebear Inc.)
• Inventors: Moran, David
• Primary Examiner(s): Brown, Luis A
• Assistant Examiner(s): Andrei, Radu
• Application Number:
  US14/636,543
• Publication Number:
  US 20170032406A1
• Time in Patent Office: 1,407 Days
• Field of Search: 705 1449, 705 1457
• CPC Class Codes: G06Q 10/063 Operations research, analysis or management
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• CPC Class Codes: G06Q 30/0242 Determining effectiveness of advertisements
• CPC Class Codes: G06Q 30/0244 Optimization
• CPC Class Codes: G06Q 30/0245 Surveys
• CPC Class Codes: G06T 1/0021 Image watermarking
• CPC Class Codes: G06T 1/0064 Geometric transfor invariant watermarking, e.g. affine transform invariant
• CPC Class Codes: G06T 3/02 Affine transformations for image registration G06T3/147; for image mosaicing G06T3/4038

Abstract

Methods and apparatus for conducting test promotions in a highly scalable and cost-effective manner using randomized experiment methodology are disclosed. Test promotions of interest are presented to visitors of a website in a randomized experiment manner wherein each page presents one test promotion of interest among other promotions. The other promotions presented in the same page may be randomized. The visitors'"'"' responses with respect to the test promotions of interest are then recorded and analyzed to determine the performance of each presented test promotion of interest.

Claims

1. A computer-implemented method for performing promotion optimization on a promotion aggregation platform, comprising:

generating a plurality of test promotions for different products and manufacturers, wherein each test promotion includes a different combination of promotion variables;

hosting a promotion aggregation platform wherein consumers have accounts;

segmenting consumers into subpopulations;

assigning a subset of the test promotions to each consumer of the segmented subpopulations of consumers;

dynamically presenting on a web application to each consumer at least one of the subset of test promotions assigned to the given consumer, along with un-assigned test promotions which are not part of the subset of the test promotions, and randomly selected from the plurality of test promotions, thereby mitigating effects of the un-assigned test promotions on responses of the consumer; and

administering a randomized order of promotion layouts in real-time for each consumer, enabling simultaneous viewing by all consumers.

2. The method of claim 1, wherein bibliographic information is collected for each consumer.

3. The method of claim 2, wherein the bibliographic data is used to assign product categories to each consumer.

4. The method of claim 2, wherein when no bibliographic data is available product categories are randomly assigned to each consumer.

5. The method of claim 1, further comprising tracking which promotions each consumer has been exposed to.

6. The method of claim 5, wherein the tracking is used during the administering to eliminate duplicate presentation of any given promotion to any given consumer.

7. The method of claim 5, wherein the tracking is used during the administering to eliminate inconsistent presentation of promotions to any given consumer.

8. The method of claim 1, wherein the responses include selecting the promotion within the consumer'"'"'s account.

9. The method of claim 1, wherein the responses include non-selection of the promotion presented to the consumer.

10. The method of claim 1, wherein the predicted future responses are insights communicated to manufacturers.