Method, algorithm, and computer program for targeting messages including advertisements in an interactive measurable medium

• US 8,260,663 B1
• Filed: 12/07/2001
• Issued: 09/04/2012
• Est. Priority Date: 12/07/2001
• Status: Active Grant

• Current Assignee: Carl Meyer
• Original Assignee: Carl Meyer
• Inventors: Ranka, Sanjay, Chang, E. Diane, Veiner, Daniel
• Primary Examiner(s): Weiss, John
• Assistant Examiner(s): Macasiano, Marilyn G
• Application Number:
  US10/017,074
• Time in Patent Office: 3,924 Days
• Field of Search: 705/10, 705/14, 705/7.33, 705/14.4, 705/14.49, 705/14.43, 705/14.66, 705/26.1, 707/102, 379/266.08, 709/224, 709/219, 709/223
• US Class Current: 705/14.49
• CPC Class Codes: G06Q 30/0241 Advertisements
• CPC Class Codes: G06Q 30/0244 Optimization

Abstract

Method, procedure, algorithm, system, and computer program and computer program product for improving and optimizing or attempting to optimize performance of messaging campaigns, particularly to marketing campaigns in which advertisements or other messages are distributed over an interactive measurable medium such as the Internet. When message is an advertisement, campaign involves a list of ad alternatives and a target customer population. Goal of message manager or marketing manager is to allocate ad alternatives to customer population to optimize business objectives such as maximizing the number of positive responses received. This is achieved at least in part by segmenting customer population into segments and then finding best allocation of ad alternatives for each segment. The number of segments and grouping of a customer population changes at different stages of the campaign. When the message is other than an advertisement, goal is to allocate messages to optimize analogous business or campaign objectives, typically measured by the number of successes or successful responses.

Claims

1. Apparatus, comprising:

a processor;

a computer memory holding computer program instructions which when executed by the processor comprise;

an optimization process programmed to receive message performance information and to generate recommended message allocations;

a segmentation process programmed to segment a target visitor population into a set of segments using one or more templates each comprising one or more cells, wherein the target visitor population comprises visitors defined by a set of visitor profile attributes, each visitor profile attribute comprising one or more portions, each portion corresponding to one or more values or range of values of an attribute, wherein values associated with one or more attributes corresponds to a cell, and the set of segments, collectively, comprise the target visitor population, wherein the segmentation process uses an algorithm to identify a segmentation by determining, recursively, beginning with a single segment that includes all of the set of segments representing the target visitor population, and continuing for each new segment resulting from a split until no new segments are produced, whether to split a segment into two or more disjoint smaller segments based on whether a weighted performance of the two or more disjoint smaller segments is better than that of the segment; and

an allocation process programmed to receive the recommended message allocations from the optimization process and to receive the set of segments from the segmentation process and, in response, generating message allocations.

2. The apparatus as described in claim 1 wherein the algorithm is a greedy algorithm.

3. The apparatus as described in claim 1 wherein visitors are classified by demographic or psychographic information along a number of attributes.

4. The apparatus as described in claim 1 wherein the segmentation process compares the set of possible segmentations to determine which segmentation will lead to a desired visitor response.

5. The apparatus as described in claim 1 wherein the segmentation process divides visitors into segments based on the performance of different messages for different visitors.