US20220086517A1 - Systems and methods to select targeted advertising - Google Patents

Abstract

Systems and methods to select targeted advertising for display are disclosed. Example systems disclosed herein are to determine an opportunity metric for a first product based on a first quantity of the first product determined to be purchased by a first household during a first period of time and a second quantity of the first product expected to be purchased by the first household during the first period of time. Disclosed example systems are also to select, based on at least the opportunity metric, a first advertisement associated with the first product to deliver to the media device, and transmit the first advertisement to the address of the media device.

Description

RELATED APPLICATIONS
• This patent arises from a continuation of U.S. application Ser. No. 16/428,397, filed on May 31, 2019, which is a continuation of U.S. application Ser. No. 15/990,341, filed on May 25, 2018, which is a continuation of U.S. application Ser. No. 12/964,481, filed on Dec. 9, 2010, which is a non-provisional application of and claims the benefit of U.S. Provisional Application No. 61/355,882, which was filed on Jun. 17, 2010. Priority to U.S. application Ser. Nos. 16/428,397, U.S. application Ser. No. 15/990,341, U.S. application Ser. No. 12/964,481, and U.S. Provisional Application No. 61/355,882 is claimed. U.S. application Ser. No. 16/428,397, U.S. application Ser. No. 15/990,341, U.S. application Ser. No. 12/964,481 and U.S. Provisional Application No. 61/355,882 are hereby incorporated herein by reference in their entireties.
FIELD OF THE DISCLOSURE
• This disclosure relates generally to targeted advertising and, more particularly, to methods and apparatus to select targeted advertising.
BACKGROUND
• Product manufacturers and advertisers try to increase demand for their products by influencing the behavior of target consumer segments. Survey research is used to collect information about consumer attitudes and preferences. Behavioral information, whether observed directly or collected through survey research, can be used to predict demand. The manufacturers try to influence consumer preference through use of adverting strategies to increase demand. A manufacture will try to optimize its advertising spending by targeting specific consumer segments that represent a high opportunity for the manufacturer to influence consumer behavior by raising consumer awareness. Since consumer attitudes and preferences are constantly changing, manufacturers must continually monitor attitudes and preferences to predict demand, as well as continue to influence consumer preference through advertising.
BRIEF DESCRIPTION OF THE DRAWINGS
• FIG. 1 is a block diagram of an example system to provide individually-targeted advertising to different households based on an opportunity metric.
• FIG. 2 is a more detailed block diagram of an example implementation of the headend system of the cable provider described in connection withFIG. 1.
• FIG. 3A is a block diagram illustrating example data flows that may occur in the example systems illustrated inFIGS. 1 and 2.
• FIG. 3B illustrates an example advertisement saturation characteristic for an advertisement of a product with respect to a household.
• FIG. 4 is a table illustrating example weekly purchases for several households and products.
• FIG. 5 is a table illustrating example expected weekly purchases per household by segment for several products.
• FIG. 6 is a table illustrating an example advertising log for several households and several advertisements.
• FIG. 7 is a flowchart representative of example machine readable instructions which may be executed to select targeted advertising for display.
• FIG. 8 is a flowchart representative of example machine readable instructions which may be executed to determine one or more opportunity metrics for a household.
• FIG. 9 is a flowchart representative of example machine readable instructions which may be executed to select an advertisement for delivery to a household.
• FIG. 10 is a flowchart representative of example machine readable instructions which may be executed to determine a media response of a household to a displayed advertisement.
• FIG. 11 is a diagram of an example processor system that may be used to execute the example instructions ofFIGS. 7, 8, 9, and 10 to implement the example systems ofFIGS. 1 and 2.
DETAILED DESCRIPTION
• Although the example systems described herein include, among other components, software executed on hardware, such description is merely illustrative and should not be considered as limiting. For example, it is contemplated that any or all of the disclosed hardware and/or software components could be embodied exclusively in dedicated hardware, exclusively in software, exclusively in firmware or in some combination of hardware, firmware, and/or software.
• Cable providers typically sell broadcast advertising time to advertisers. Targeted advertising is a method of advertising performed by placing advertisements at times and locations where a particular type of consumer (e.g., a particular demographic segment) that is likely to be influenced to purchase the advertised product is likely to view the advertisement. Advertisers often look to increase the effectiveness of their advertisements by using targeted advertising.
• When considering how much money to spend on an advertisement, an advertiser will consider the reach of the advertisement (i.e., the number of viewers or households who view the advertisement in a given period of time) and the frequency with which the advertisement is shown. As the reach of the advertisement increases, more consumers may view the advertisement and the number of consumers who may be influenced to purchase a product is increased. However, as the frequency of the advertisement increases, the advertisement may lose its effectiveness over time. Specifically, repeated showings of an advertisement to a consumer tend to diminish the effect of the advertisement with each additional showing. Thus, additional resources spent on showing a “stale” advertisement may be better spent on other (non-stale) advertisements.
• The example systems and methods described herein are useful for effectively selecting and displaying advertising targeted to individuals and/or households. In some examples, an advertisement is selected for display to a household based on an opportunity metric generated for the household with respect to a product. Different advertisements may be selected for delivery to different households via, for example, a set top box installed at each of the households. The example systems and methods also consider the likely effectiveness of additional showings of a particular advertisement to a particular household (e.g., based on a number of prior showings and historical data showing the relationship(s) between the frequency of showings and effectiveness) and further adjust the selection of the advertisements to a household based on the predicted effectiveness. By controlling the selection of individually-targeted advertisements as described herein, content providers may increase the revenue generated by selling advertisement space and may improve the effectiveness of advertisement spending by advertisers.
• Example systems and methods described herein select advertisements for individual household delivery based on a highly individualized and dynamic opportunity metric or score with respect to a product. Previous systems use proxy methods to measure purchasing such as, for example, measuring television program audiences to approximate the number and types of households purchasing a product. However, such approaches do not adapt well in view of purchases of an individual household. In contrast, the example systems and methods disclosed herein can present the most effective advertisements to a particular household, because the selected advertisements change as the household adjusts its purchasing habits over time. The speed with which the example systems and methods adjust the opportunity calculations may keep pace with the rate at which consumers make their purchases and respond to advertisements.
• As used herein, the saturation of a household with respect to an advertisement refers to the effectiveness of the advertisement for future showings of the advertisement. Saturation may be limited to periods of time, after which the saturation of the household is reset, or may continue indefinitely. When a household is said to have “reached” saturation, future presentations of the advertisement generally have reduced effectiveness relative to earlier presentations.
• FIG. 1 is a block diagram of anexample system100 to provide individually-targeted advertising to different households A102 andB104 based on an opportunity metric. Theexample system100 includes acontent provider106 such as a cable television company to deliver programming and advertisements to a number of households A102 and B104. The households A102 andB104 are provided with respective settop boxes108 and110 coupled totelevisions112 and114. Thecontent provider106 may broadcast programming via aheadend system107 over a plurality of channels, to which the settop boxes108 and110 may tune to receive the broadcast programming. Additionally or alternatively, thecontent provider106 may provide a library or catalog of programming and/or advertisements that may be selected for on-demand delivery to either or both of the settop boxes108 or110. The tuned or delivered programming is displayed on therespective televisions112 and114. The television could alternatively or additionally be implemented by personal computer monitors or the like.
• In theexample system100, thecontent provider106 sells time on each of one or more channels to one or more advertisers. Thecontent provider106 may agree to broadcast particular advertisements to the settop boxes108 and110 and, thus, thetelevisions112 and114 for display to consumers.
• Advertisers seek to increase the effectiveness of their advertising expenditures by a) increasing the size of the audience that is exposed to their advertisements; and/or b) spending advertising resources in a manner designed to concentrate exposure of the advertisements on segments of consumers who are more likely to purchase the product being advertised. By increasing the audience that views an advertisement, the advertiser can increase the number of people who are influenced by the advertisement on the theory that a given advertisement will influence a percentage of people to purchase the product. However, by concentrating an advertisement's exposure on a particular class or segment of people who are more likely to be influenced than the public at large (e.g., because the product is of more interest to them), each unit of advertising resources can influence a relatively larger number of people to purchase the product being advertised.
• Traditionally, consumers have been grouped or classified by broad demographic and/or geographic segments. Surveys, shopper membership cards, and other data collection methods have been used to aid advertisers in determining when and where to place advertisements. However, these methods relied on massive generalizations of geodemographic data to purchasing habits. As a result, these prior methods may only yield nominal statistical improvements in some cases. In contrast, theexample system100 ofFIG. 1 includes anopportunity calculator118 to determine opportunity metrics for the individual households A102 andB104. The opportunity metrics are representative of the opportunity a household represents to advertisers of particular products based onpurchase data120 and122 for respective individual households A102 andB104 and expectedpurchase data124.
• Thehousehold purchase data120 and122 identifies purchases made by the respective example households A102 andB104. Thepurchase data120 and122 may be provided to theopportunity calculator118 by, for example, self-reporting purchases, using a purchase logging device to identify and/or log purchases' (e.g., a Nielsen HomeScan® device used in panel tracking), and/or any other reporting method and/or combination of reporting methods. Self-reporting may occur via filling out a survey and/or keeping a manual and/or electronic log of purchases.
• Another example method to collect purchase data by household may include collecting frequent shopper card data. Many retail stores currently issue frequent shopper cards to persons who volunteer personal information. The personal information may include any one or more of the person's address, the person's personal or household demographics, the person's personal preferences, and/or any other voluntarily-provided personal data. A frequent shopper card is scanned whenever the carrier of the frequent shopper card makes purchases at the issuing retailer, and the purchased items are logged and attributed to the carrier of the frequent shopper card. In exchange for the personal information and monitoring ability provided by the frequent shopper card, the retailer offers discounts on purchases to the carrier of the card. The retailer may then use the collected purchase data to identify patterns or perform other data operations to obtain useful shopper data.
• The expected purchases, as described in more detail below, may be determined by acollaborative filter126 that determines a quantity of a product that the households A102 andB104 might be expected to consume (e.g., per day, per week, per month, per year) based on the segment(s) in which the households A102 andB104 may be classified and what those segment(s) typically purchase. By determining the difference between what quantity of a given product the respective households A102 andB104 are expected to purchase (e.g., expectedpurchases124 ofFIG. 1) and what quantity of that product the households A102 andB104 actually purchased as reflected in thepurchase data120 and122, theopportunity calculator118 generates an opportunity metric. The opportunity metric is provided to thecontent provider106 to enable thecontent provider106 to select the most appropriate (e.g., effective) advertisements to be delivered to the households A102 andB104.
• To evaluate the relationships between different items, thecollaborative filter126 may be populated and/or updated with the purchasing relationships between different goods and/or brands. Thecollaborative filter126 may be maintained by, for example, a research service or agency that performs market research of consumer segments and behaviors.
• FIG. 2 is a more detailed block diagram of anexample headend system200 to implement theheadend system107 of thecontent provider106 described in connection withFIG. 1. Theheadend system200 may be used to select and display advertisements for individual households (e.g., the households A102 and B104) based on one or more opportunity metrics. For clarity and brevity, the following description ofFIG. 2 will reference an example where theheadend system200 considers thehousehold A102 ofFIG. 1. Theexample headend system200 includes anadvertisement selector202 to select an advertisement for delivery to a household via amedia deliverer204. Themedia deliverer204 may be implemented via, for example, a cable broadcast system. In the illustrated example, themedia deliverer204 is in communication with multiple subscriber set top boxes (e.g., the settop boxes108 and110 ofFIG. 1).
• Theadvertisement selector202 is provided with anadvertisement database206, from which theadvertisement selector202 may choose advertisements associated with one or more products. The advertisements may be provided to thedatabase206 by one or more advertisers (e.g., theadvertisers116 ofFIG. 1) who wish to have their product advertisement(s) targeted at households that have a high opportunity metric corresponding to the product(s) being advertised.
• Theadvertisement selector202 receives several inputs to determine an advertisement that should be delivered to a particular household. For example, theadvertisement selector202 receives the selection of ahousehold A102 from ahousehold selector208. The selection of the household focuses theadvertisement selector202 on thehousehold A102 that should be considered for individual advertisement delivery. As mentioned above, theadvertisement selector202 receives the opportunity metric(s) for one or more products. Because the opportunity metric(s) are specific to an individual household, theadvertisement selector202 considers the opportunity metric(s) corresponding to the household selected by thehousehold selector208.
• In the illustrated example, theadvertisement selector202 also receives programming association information from an ad/programming associator210. The ad/programming associator210 identifies advertisements for products in theadvertisement database206 which are appropriate for programming currently shown to thehousehold A102. For example, if a first advertisement and a second advertisement in the advertisement database correspond to equal or similar opportunity metric(s) for thehousehold A102, theassociator210 may identify an association between the first advertisement and the program being viewed at the household that causes theselector202 to select the first advertisement for delivery. For example, if a young children's program is being viewed at thehousehold A102, theassociator210 may indicate that a first advertisement for cereal has a more appropriate association with the program than a second advertisement for coffee. In such a case, theselector202 may give weight to the association and select the advertisement for cereal.
• Theexample advertisement selector202 further receives advertisement response data for thehousehold A102 from amedia response evaluator212. Themedia response evaluator212 monitors the advertisements that are shown to the household A102 (e.g., that are sent from theadvertisement selector202 to themedia deliverer204 and/or that are sent to thehousehold A102 during the course of regular programming and advertisement) and reports to theselector202 whether the advertisement being considered has been shown to thehousehold A102 enough to cause the advertisement to lose its effectiveness with respect to thehousehold A102. For example, when theadvertisement selector202 sends a selected advertisement to themedia deliverer204 for delivery to thehousehold A102, the selected advertisement (or, alternatively, metadata identifying the advertisement) is also sent to themedia response evaluator212. Themedia response evaluator212 uses one or more marketing mix models or saturation algorithms to determine whether thehousehold A102 has been presented with the advertisement often enough to reduce the effectiveness of the advertisement on future presentations.
• As an example of the operation of themedia response evaluator212, assume that theadvertisement selector202 is considering two advertisements: one for product X and one for product Y for delivery to thehousehold A102. The opportunity metric for thehousehold A102 associated with product X is higher than the opportunity metric associated with the product Y. However, the advertisement associated with product X has been shown to thehousehold A102 many times, while the advertisement associated with product Y has not yet been shown to thehousehold A102. As a result, themedia response evaluator212 determines that the effectiveness of the advertisement for product X is reduced because thehousehold A102 has become saturated with the message presented by the advertisement. In contrast, the effectiveness of an advertisement for product Y is still relatively high (given the lower opportunity metric for product Y). The calculated effectiveness of showing each advertisement may then be correlated to a price to charge an advertiser to show its advertisement.
• In operation, theexample headend system200 receives opportunity metric(s) that are individualized for several households. Thehousehold selector208 selects one of the households to be considered by theadvertisement selector202. Theadvertisement selector202 receives (e.g., requests, retrieves from storage, etc.) the opportunity metric(s) associated with the selected household. Theadvertisement selector202 evaluates the products for which an opportunity metric is provided and determines whether the household is saturated with the advertisement(s) for any of the product(s). Saturation information may be requested from themedia response evaluator212, which determines the saturation and/or the future media response of the household to the potential advertisement(s). Themedia response evaluator212 may perform the evaluation on request from theadvertisement selector202 and/or have previously prepared evaluations stored. For those advertisement(s) that have reached saturation, theadvertisement selector202 applies a penalty.
• Theadvertisement selector202 loads a pricing structure, pricing characteristic, or pricing framework from thepricing database214 and determines, based on the opportunity metric(s) and the saturation of the advertisements, which advertisement(s) may command the highest price from advertisers due to the likelihood that the advertisement(s) will be effective. Theadvertisement selector202 provides the selected advertisement(s) to themedia deliverer204, which determines the set top box address associated with the selected household from theaddress database216 and delivers the advertisement(s) to the address at the appropriate time. The advertisement(s) are further provided to themedia response evaluator212 to determine the response of the selected household to further presentations of the selected advertisement(s).
• FIG. 3A is a block diagram illustrating example data flows that may occur in the example systems illustrated inFIGS. 1 and 2. The example data flows illustrated inFIG. 3A show the exchanges of data between thehousehold A102, thecontent provider106, the settop boxes108 and110, and theadvertisers116. While theexample household A102 is illustrated inFIG. 3A, the example exchanges of data may be used in a similar or identical manner with respect to additional households (e.g., the household B104). The data flows300 ofFIG. 3A further detail the exchange of data between theadvertisement selector202, themedia deliverer204, theadvertising database206, thehousehold selector208, the ad/programming associator210, themedia response evaluator212, thepricing database214, and theaddress database216 within theexample headend system200 ofFIG. 2.
• For clarity and brevity, the example data flows300 ofFIG. 3A will refer to the selection of advertisement(s) for delivery to thehousehold A102 ofFIG. 1 without regard to thehousehold B104. However, the examples described herein may be extended to include any number of households.
• Theexample household A102 ofFIG. 1 is associated with purchase data302 (e.g., the products and quantities purchased by the household during a given time period, etc.) and geodemographic data304 (e.g., the household geographic location, the number of persons in the household, the number of children, the ages of the persons in the household, etc.). Thepurchase data302 and thegeodemographic data304 may be provided by, for example, self-reporting by thehousehold A102 and/or monitoring techniques practiced within thehousehold A102. Thepurchase data302 may additionally or alternatively be provided through the use of frequent shopper card data associated with frequent shopper cards carried by members of thehousehold A102.
• Based on thepurchase302 anddemographic data304, thehousehold A102 fits into a segment of consumers. Members of the consumer segment are identified as part of other marketing studies and their purchases are tracked over time.Consumer segment data306 may be developed and/or updated to reflect, for example, new product offerings, purchasing trends, and/or changes in membership characteristics. Additionally, retailerpurchase panel data308 is collected at different points of sale. In some examples, theproduct purchase data308 is collected from as many retailers or other points of sale as possible. Additionally or alternatively, the retailerpurchase panel data308 is received from a data processing facility that has collected, aggregated, and/or processed purchase data from multiple retailers (e.g., a set of retailers in one or more geographic areas). The processed purchase data may be filtered to provide data that is particularly useful to determining the expected purchases of thehousehold A102 and/or the segment.
• Thepurchase data302 and thegeodemographic data304 from thehousehold A102, the relevantconsumer segment data306, and the relevant retailerpurchase panel data308 are input to acollaborative filter126. Thepurchase data302 provided to thecollaborative filter126 may include product identifiers and quantities per time period. Thegeodemographic data304 of the illustrated example includes sufficient geodemographic information to place thehousehold A102 into a consumer segment. The exampleconsumer segment data306 of the illustrated example provides at least the segment purchase information of the products purchased by thehousehold A102. Theconsumer segment data306 may additionally provide available product purchase information for the consumer segment into which thehousehold A102 fits. The retailerpurchase panel data308 provides bills of sale that illustrate the combinations of products that are often purchased together.
• Thecollaborative filter126 receives thepurchase data302 and thegeodemographic data304 from thehousehold A102, theconsumer segment data306, and the retailerpurchase panel data308, and determines one or more opportunity metric(s) for one or more products based on the received information. For example, thepurchase data302 includes a quantity of a product X (e.g., a brand of a product) purchased weekly by thehousehold A102. Thegeodemographic data304 allows the collaborative filter126 (or some other classification device) to generate a purchase expectation for thehousehold A102 based on placement of thehousehold A102 into a segment (e.g., segment N). The placement of thehousehold A102 into the segment N may be based on other data besides geodemographic data, such as thepurchase data302.
• Theconsumer segment data306 provides the typical (e.g., average) weekly purchases of one or more products, including product X, by households in the segment N. Thus, thecollaborative filter126 may determine the amount by which thehousehold A102 is below or above the typical weekly purchases for the segment N. Retailerpurchase panel data308 allows thecollaborative filter126 to evaluate the combinations and quantities of products that are often purchased together, which may then be applied to thepurchase data302 to determine any other products that the household may be interested in based on its purchases of product X and the historical purchasing trends associated with the segment N.
• Unlike known collaborative filters, the examplecollaborative filter126 does not assume that once a product is purchased, the product no longer needs to be purchased again (or that the product does not need to be purchased long enough to assume no further purchase is desired). Such an assumption may be appropriate when recommending media, toys, consumer electronics, books, and/or other durable goods and/or items for which one purchase is often sufficient, based on previous purchases of such items. Instead, thecollaborative filter126 ofFIG. 3A determines recommended or high-opportunity items based on volumetric purchases such as foodstuffs, cleaning supplies, personal hygiene items, and/or other consumable items which are purchased and consumed with some regularity. However, thecollaborative filter126 may also determine or account for products that are generally purchased infrequently.
• Thecollaborative filter126 determines, based on thepurchases302 of thehousehold A102 and historical purchasing trends of the related segment N, items that may be desirable substitutes for and/or supplements to products thehousehold A102 currently purchases. By determining the segment (e.g., N) of thehousehold A102, thecollaborative filter126 determines the quantity of the product X that is typically purchased by households in the segment N, the additional quantity of the product X that thehousehold A102 should be buying based on its segment if the quantity purchased by the household A is below the typically purchased quantity, and similar products and quantities purchased by households in the segment N which have not been purchased or have been under-purchased by thehousehold A102. Thecollaborative filter126 further determines additional products that may be similar or dissimilar to the product X that are often purchased by households in the segment N based on the retailerpurchase panel data308. In some examples, the retailerpurchase panel data308 is at least partially used to determine theconsumer segment data306.
• Thecollaborative filter126 outputs one ormore opportunity metrics312 for each product (e.g., products X, B, C, and D) that are identified as having an opportunity associated with thehousehold A102. The opportunity metric of a product X is based on thepurchase data302 of the household A, the segment N in which the household A is located, and the typical purchases of the product X, the product's substitutes, and/or the product's complements by households in the selected segment N.
• For example, assume thehousehold A102 purchases thirty-six cans of soft drink D per week, households in the segment N purchase an average of twenty-four cans of soft drink X per week and twenty-four cans of soft drink D per week. Additionally, assume thehousehold A102 purchases zero ounces of potato chips B per week and members of segment N average purchases of sixteen ounces of potato chips B per week. In this example case, thecollaborative filter126 determines that the opportunity for motivating thehousehold A102 to increase its purchases of soft drink D is low because it currently purchases more of soft drink D than is typical for segment N. In contrast, the opportunity for soft drink X is high, because thehousehold A102 purchases none of soft drink X compared to the average of12 cans of soft drink X per week. Therefore, advertising soft drink X to thehousehold A102 may be highly likely to influence thehousehold A102 to increase its purchases of soft drink X. Additionally, thecollaborative filter126 determines that the opportunity for potato chips B is high because the average purchases for potato chips B among households in segment N is sixteen ounces per week and household A is currently not purchasing chips. Thus, an advertisement for potato chips B may have a relatively higher likelihood to influence thehousehold A102 to increase its purchases of potato chips B. If desired, the purchase data may be supplemented with user-specific data collected via, for example, a survey reflecting user preferences, dietary habits, medical conditions, allergies, etc. This supplemental data may be factored in by thecollaborative filter126.
• The opportunity metric(s)312 developed by theopportunity calculator118 ofFIG. 1 are input to theadvertisement selector202 ofFIG. 2. In addition to the opportunity metric(s)312, theadvertisement selector202 receives pricing information314 (e.g., from thepricing database214 ofFIG. 2), advertisement/programming associations316 (e.g., from the ad/programming associator210 ofFIG. 2), advertisements318 (e.g., from theadvertisement database206 ofFIG. 2), and household saturation information (e.g., a saturation metric) from an advertising response monitor320 (e.g., from themedia response evaluator212 ofFIG. 2). The selection of the household A102 (e.g., from thehousehold selector208 ofFIG. 2) is identified expressly or implicitly in the delivery of the opportunity metric(s)312 forhousehold A102.
• While the example opportunity metric(s)312 are shown inFIG. 3A as a score, the opportunity metric(s)312 may be presented or measured in any suitable manner. For example, the opportunity metric(s) of thehousehold A102 with respect to a product may be represented in terms of a normalized or gross score, a monetary amount (e.g., dollars/year), units of product (e.g., ounces/year), or any other appropriate unit or score.
• Thepricing information314 may include, for example, a function based on the opportunity metric(s) of a particular product, a pricing structure based on a client and/or volume of advertisements, or other pricing structure, characteristic, and/or factors. In some examples, thepricing information314 includes a function that increases the price of delivering a particular advertisement to a particular household or to a number of households based on the opportunity metric(s) associated with the advertised product and the household.
• In some example advertisement pricing models, a media provider (e.g., themedia provider200 ofFIG. 2) contracts with an advertiser (e.g., the advertisers116) to provide an advertisement with a certain number of viewers. The number of viewers may be calculated using, for example, reach and frequency numbers. Certain measures of advertisement exposure (e.g., gross ratings points) may consider showing an advertisement once to each person in a population to be equivalent to showing the advertisement twice each to half of the people in the population. In some examples, themedia provider200 may agree to provide an advertisement with a specified reach and frequency to households having a minimum opportunity metric for a price premium reflecting the improved advertising opportunity to theadvertiser116. Theadvertiser116 may determine that the premium is acceptable, or even a bargain, to target fewer households having a high opportunity metric instead of targeting a more general audience. However, many different methods and models of advertisement pricing based on the opportunity metric are available and are considered within the scope of the examples described herein. Pricing methods and models may be easily modified to improve revenue to both themedia provider200 and theadvertisers116.
• In some examples, themedia provider200 may determine the opportunity metric for a type of product as opposed to a particular brand. For example, cola is a type of product where Coke® and Pepsi® are particular brands. To increase revenue, themedia provider200 may solicit bids from the manufacturers of different colas on reach and frequency agreements for advertising priority to households having the highest opportunity metrics for cola.
• The advertising/programming associations316 may include, for example, broadcast programs that advertisers identify as preferable to identify with advertised products. For example, the ad/programming associations316 may include an association between a breakfast cereal C and a children's program specified by the advertiser or manufacturer of breakfast cereal C. These associations may be stored in a table and may be manually input based on data and/or requests from advertisers, broadcasters, and/or content creators.
• Theadvertisements318 include at least the advertisements for products having an opportunity metric provided by thecollaborative filter126. In some examples, theadvertisements318 include multiple different advertisements (e.g., variants of an advertisement or completely different advertisements) that may be shown to the household for the same product. Thus, theadvertisement selector202 may show different advertisements for the same product to thehousehold A102, thereby decreasing the saturation of thehousehold A102 to advertisements for a product. By changing the advertisements for a product, theadvertisement selector202 may maintain higher revenue for an advertisement mix sent to thehousehold A102 by maintaining a high effectiveness of the advertising mix.
• The ad response monitor320 monitors the advertising sent to thehousehold A102. The ad response monitor320 may monitor only household A and/or may be part of a larger section which monitors broadcast advertising sent to program viewers in a geographic area of interest. Broadcast advertisements may be identified in any manner, such as by reading an identification code embedded in (or otherwise broadcast with) the broadcast advertisement. Based on the advertising sent to thehousehold A102, the ad response monitor320 determines saturation metric(s) of thehousehold A102 with respect to one or more advertisements. As the number of times a particular advertisement is presented tohousehold A102 increases, the ad response monitor320 determines that the incremental effectiveness of that advertisement decreases with additional showings (i.e., the total effectiveness increases more slowly). The example ad response monitor320 thus provides a penalty to be applied tocertain advertisements318 that may be selected by theadvertisement selector202 when theadvertisements318 have been shown a sufficient number of times. Additionally, consumers with different demographics may tend to have different ad response characteristics or the ad response characteristics used by the ad response monitor320 may change over time. Therefore, the ad response monitor320 may be provided and/or updated by, for example, a consumer research service or agency that specializes in consumer behavior.
• As the number of times anadvertisement318 is shown increases beyond a saturation point, the ad response monitor320 increases the penalty. In some examples, however, the penalty may decrease from the first presentation to the second and/or third (and/or additional) presentations and then increase for presentations after the third (or later) presentation. One or more saturation characteristic(s)324 provide models for the ad response monitor320 to apply the penalty. An example of such a saturation characteristic is illustrated inFIG. 3B. Thus, while advertisements for products having mid- to high-range opportunity metrics with respect to thehousehold A102 may be sufficiently high to overcome the penalties and, thus, continues to be shown, after anadvertisement318 has been shown often enough the price for thatadvertisement318 will be overtaken by another advertisement offering a higher price because it is expected to exhibit higher advertising effectiveness.
• When theadvertisement selector202 has received thepricing information314, the advertisement/programming associations316, theadvertisements318, and the saturation metric(s) from the ad response monitor320, theadvertisement selector202 selects an advertisement for delivery and/or presentation to the household A102 (e.g., via the settop box108 and thetelevision112 ofFIG. 1). The selected advertisement and the destination for the advertisement (e.g., household A102) are provided to themedia deliverer204. Themedia deliverer204 also receiveshousehold address information322 corresponding to the settop box108 in thehousehold A102. The address information may include, for example, a media access control (MAC) address, an Internet protocol (IP) address, or any other type of network layer or other type of address that uniquely identifies the settop box108 ofhousehold A102. Themedia deliverer204 delivers the selected advertisement to the settop box108 inhousehold A102 at the appropriate time, such as shortly prior to the time space sold by thecontent provider200.
• Themedia deliverer204 may also be responsible for broadcasting (e.g., to a large portion of the possible audience) programming and/or advertisements to thehouseholds102 and104. In addition to providing the selected advertisements (e.g., advertisements individually selected for a household) and broadcast advertisements (e.g., advertisements not individually selected for a household) to thehousehold A102, themedia deliverer204 further provides and/or identifies the selected advertisements and broadcast advertisements to thead response monitor320. As mentioned above, the ad response monitor320 monitors the advertisements presented tohousehold A102 and provides household saturation information (e.g., a saturation metric) to theadvertisement selector202 to select future advertisements for delivery to thehousehold A102. Thus, the ad response monitor320 uses the identified advertisements to update the saturation metric or level.
• The advertisements provided to the settop box108 and/or presented to thehousehold A102 may be further fed back to the advertisement selector through advertisement-driven purchasing by thehousehold A102. For example, when an advertisement stimulates purchases of product X that previously had a high opportunity metric, thecollaborative filter126 may determine that the opportunity metric for thehousehold A102 for product X decreases because the quantity of product X that thehousehold A102 is expected to purchase has been constant for the relevant time period, unless thehousehold A102 changes segment or the behavior data associated with the segment changes as may happen over time (e.g., seasonally). As a result, the advertisements for product X become less effective and new products may be advertised to thehousehold A102 at a higher effectiveness and, thus, a higher price.
• The example data flows illustrated inFIG. 3A may be performed by any one or more parties. In some examples, a cable or other media provider (e.g., the content provider106) ofFIG. 1 may implement at least theadvertisement selector202, themedia deliverer204, and the ad response monitor320 to deliver media and advertisements to households and to maintain a high degree of responsiveness to saturation of households. In some examples, a media research organization may collect and/or process thepurchase data302, thegeodemographic data304, theconsumer segment data306, theretail purchase data308, and/or implement thecollaborative filter126 to generate opportunity metrics. The media research organization then provides the opportunity metrics to a content provider to select and deliver advertisements. In some examples, the media research organization may additionally provide and/or update the saturation characteristic(s)324 to improve the selection of advertisements. While the data flows may be implemented by any one or more parties, the examples described above may leverage existing expertise and relationships to improve service to advertisers and consumers.
• FIG. 3B illustrates an exampleadvertisement saturation characteristic324 for an advertisement of a product X with respect to theexample household A102. The horizontal axis is representative of the number of presentations of the advertisement to thehousehold A102. The vertical axis is representative of the expected likelihood that thehousehold A102 will exhibit a response or behavior (e.g., purchasing the advertised product). As illustrated inFIG. 3B, the expected likelihood of a response increases more rapidly between the second presentations and the third presentation than for other presentations (e.g., between the first and second presentations, between the fourth and fifth presentations, etc.).
• The example saturation characteristic324 may be provided to, for example, the ad response monitor320 to determine a saturation metric for the product X and thehousehold A102. The ad response monitor320 determines the number of presentations of the advertisement to thehousehold A102 and generates a saturation metric (e.g., a penalty). The ad response monitor320 provides the saturation metric to theadvertisement selector202, which may apply (e.g., subtract, multiply, etc.) the saturation metric to the corresponding opportunity metric to generate a net effectiveness metric. Different advertisements for the same product X may have different net effectiveness metrics depending on the number of times the respective advertisements have been presented to theexample household A102. Theadvertisement selector202 may then select an advertisement for delivery to thehousehold A102 by, for example, comparing the net effectiveness metrics to a threshold. In some examples, the threshold is determined by an agreement with an advertiser. However, the net effectiveness metrics may be used to identify an advertisement for delivery in any appropriate manner.
• According to theexample saturation characteristic324, the ad response monitor320 may cause the saturation metric of thehousehold A102 to the advertisement to be higher after thehousehold A102 has been presented the advertisement five times than after thehousehold A102 has been presented the advertisement two times. However, according to thesaturation characteristic324, the saturation metric after thehousehold A102 has been presented the advertisement once may be very similar to the saturation metric after thehousehold A102 has been presented the advertisement five times.
• The example saturation characteristic324 may be represented by the equation y=1+e^A+Bf, where y is the expected likelihood of response, f is the frequency with which presentations of an advertisement are presented to a household, and A and B are variables that may be determined empirically by, for example, a media research organization.
• While anexample saturation characteristic324 is illustrated inFIG. 3B,saturation characteristics324 may be additionally and/or alternatively represented by, for example, a mathematical algorithm, a lookup table, or any other appropriate representation. Further, thesaturation characteristic324 may differ between household, segment, and/or product combinations.
• FIG. 4 is a table400 illustrating example expected weekly purchases per household by segment. The example table400 may be representative of theconsumer segment data306 ofFIG. 3A for example segments N, Q, R, S, and T. The table400 includes the expected purchases of several products, including beer Z, chips B, bread H, bread J, and milk M. The expected purchases provided by table400 may be used by thecollaborative filter126 to determine opportunity metrics with respect to the products Z, B, H, J, and/or M for households belonging to the segments N, Q, R, S, and/or T. The example product references Z, B, H, J, and M are representative and would normally be replaced with brand names of corresponding products and/or other identifiers such as the UPC or the SKU of a product.
• FIG. 5 is a table500 illustrating example weekly purchases for several households by product. The example table500 may be provided to thecollaborative filter126 and used in combination with the example table400 to determine the opportunity metrics of the households 1-5 listed in the table400 with respect to the listed products. The example table500 further includes the segment that each household 1-5 falls into based on similar purchases and/or geodemographic information.
• To generate the opportunity metrics for the households 1-5, thecollaborative filter126 compares the purchases of each of the relevant products in the example weekly (or other time period) purchases table500 with the expected purchases of the corresponding products found in the appropriate segment row of the expected purchases table400. For example, thecollaborative filter126 compares the weekly purchases of chips B forhousehold 1, which is in segment N, with the expected weekly purchases of chips B by households in segment N in the expected purchases table400. If the weekly purchases and the corresponding expected purchases are similar or identical, the example opportunity metric corresponding to the product will be low. In contrast, if the weekly purchases of the product are lower than the expected purchases, the opportunity metric for that product with respect to that household will be high.
• FIG. 6 is a table600 illustrating an example advertising log for several households and several advertisements. The example table600 may be maintained by the examplemedia response evaluator212 ofFIG. 2 to evaluate the media response of different households 1-5 at the appropriate times. In the example table600, theadvertisements 1 and 2 represent broadcast advertisements that are shown to each of the households 1-5 during regular programming and advertising times, which may not be directly under the control of thecontent provider106. In contrast, theexample advertisement3 is targeted at thehouseholds 1 and 4 using the example system(s) and/or method(s) disclosed herein in addition to being shown as part of regular broadcast programming to all households 1-5. The remaining advertisements 4-11 are targeted advertisements presented to the individual households 1-5 during advertising time owned by thecontent provider106 and not presented as part of a blanket advertising campaign. Any number of advertisements and/or households may be stored in the example table600 by themedia response evaluator212 and used to evaluate media response to advertisements presented to the households.
• When evaluating the media responses according to the example table600, themedia response evaluator212 considers saturation to begin after three presentations of a given advertisement. However, any statistically or otherwise determined number of presentations may be used. Thus, in the example ofFIG. 6, when themedia response evaluator212 evaluates advertisements to be delivered tohousehold4, themedia response evaluator212 applies a penalty toadvertisements1,2, and9.
• While an example manner of implementing thesystem100 and theheadend system107 has been illustrated inFIGS. 1, 2 and/or 3A, one or more of the elements, processes and/or devices illustrated inFIGS. 1, 2 and/or 3A may be combined, divided, re-arranged, omitted, eliminated and/or implemented in any other way. Further, the example settop boxes108 and/or110, theexample content provider106, theexample headend system107, theexample opportunity calculator118, the examplecollaborative filter126, theexample advertisement selector202, theexample media deliverer204, theexample advertisement database206, theexample household selector208, the example advertisement/programming selector210, the examplemedia response evaluator212, theexample pricing database214, theexample address database216 and/or, more generally, theexample system100 may be implemented by hardware, software, firmware and/or any combination of hardware, software and/or firmware. Thus, for example, any of the example settop boxes108 and/or110, theexample content provider106, theexample headend system107, theexample opportunity calculator118, the examplecollaborative filter126, theexample advertisement selector202, theexample media deliverer204, theexample advertisement database206, theexample household selector208, the example advertisement/programming selector210, the examplemedia response evaluator212, theexample pricing database214, theexample address database216 and/or, more generally, theexample system100 could be implemented by one or more circuit(s), programmable processor(s), application specific integrated circuit(s) (ASIC(s)), programmable logic device(s) (PLD(s)) and/or field programmable logic device(s) (FPLD(s)), etc. When any of the appended claims are read to cover a purely software and/or firmware implementation, at least one of the example settop boxes108 and/or110, theexample content provider106, theexample headend system107, theexample opportunity calculator118, the examplecollaborative filter126, theexample advertisement selector202, theexample media deliverer204, theexample advertisement database206, theexample household selector208, the example advertisement/programming selector210, the examplemedia response evaluator212, theexample pricing database214, and/or theexample address database216 are hereby expressly defined to include a tangible medium such as a memory, DVD, CD, etc. storing the software and/or firmware. Further still, theexample system100 may include one or more elements, processes and/or devices in addition to, or instead of, those illustrated inFIGS. 1, 2, and/or3, and/or may include more than one of any or all of the illustrated elements, processes and devices.
• FIG. 7 is a flowchart representative of example machinereadable instructions700 which may be executed to implement theexample system100 ofFIGS. 1-3A. Theexample instructions700 may be executed by, for example, theprocessor1102 ofFIG. 11 to implement theheadend system200 illustrated inFIG. 2 to deliver targeting advertising to individual households. In some examples, theinstructions700 may be executed separately for each household for which targeted advertising will be selected and delivered. Theexample instructions700 begin by determining one or more opportunity metric(s) for one or more products for a household (e.g., thehousehold A102 or thehousehold B104 ofFIG. 1) (block702). An example process to implementblock702 is described in more detail below with reference toFIG. 8.
• The headend system200 (e.g., via theadvertisement selector202 ofFIG. 2) selects an advertisement for display or presentation to the household based at least on the opportunity metric of the product associated with the advertisement (block704). The selection of the advertisement may be based further on a pricing structure, an advertisement/programming association, and/or a saturation characteristic of the household with respect to different advertisements. An example process to implementblock704 is described in more detail below with reference toFIG. 9. The headend system200 (e.g., via themedia deliverer204 ofFIG. 2) delivers the selected advertisement to the household (block706). Themedia deliverer204 may deliver the selected advertisement during a time slot owned by thecontent provider106. In some examples, themedia deliverer204 receives an address corresponding to a set top box (e.g., the settop box108 in thehousehold A102 ofFIG. 1) and transmits the advertisement to the address.
• The headend system200 (e.g., via themedia response evaluator212 ofFIG. 2) then determines the media response of the household to the advertisement (block708). For example, themedia response evaluator212 may determine that future presentations of the delivered advertisement are likely to exhibit decreased effectiveness because the household has been saturated with the advertisement. An example process to implementblock708 is described in more detail below with reference toFIG. 9. Themedia response evaluator212 then feeds back the media response to the advertisement selector202 (block710). Theadvertisement selector202 may use the media response in selecting future advertisements for delivery to the household (e.g., future iterations of block704).
• FIG. 8 is a flowchart representative of example machinereadable instructions800 which may be executed to determine one or more opportunity metrics for a household. Theexample instructions800 may be executed by, for example, theprocessor1102 ofFIG. 11 to implement thecollaborative filter126 ofFIG. 3A and/or block702 ofFIG. 7.
• The examplecollaborative filter126 receives purchasing information (e.g., the purchasinginformation302 ofFIG. 3A) and geodemographic information (e.g., thegeodemographic information304 ofFIG. 3A) from a household (e.g., thehousehold A102 ofFIG. 3A) (block802). Thecollaborative filter126 further receives purchasing and geodemographic data from one or more retail panels (block804). In some examples, block804 may be accomplished by receiving aggregate and/or processed frequent shopper card purchase data representative of multiple retailers in a geographic area of interest. By receiving the aggregate data, the examplecollaborative filter126 may receive data specific to the segment and/or geographic area to which thehousehold A102 belongs. Thecollaborative filter126 determines the segment to which thehousehold A102 belongs based on thepurchase information302 and the geodemographic information304 (block806). In some examples, the segments are determined based on thepurchase information302, thegeodemographic information304, and/or the data received from retail panels. Alternatively, the segments may be predetermined by an outside provider. Thecollaborative filter126 may use additional and/or alternative data to determine the household segment.
• Thecollaborative filter126 then identifies products for which an opportunity metric may be calculated (block808). Products identified by thecollaborative filter126 may include products purchased by thehousehold A102 and/or complements to such products. For example, thecollaborative filter126 may determine, based on the purchase data from different points of sale, that particular products purchased by thehousehold A102 have popular substitutes and/or complementary products. For instance, thecollaborative filter126 may determine that, based on the points of sale data, purchasers who purchase a particular brand of potato chips also tend to purchase a particular brand or flavor of chip dip as a complementary item. Similarly, thecollaborative filter126 may determine that purchasers who buy combinations of certain cereals and candy also tend to purchase a particular brand of flavored drink mix.
• Thecollaborative filter126 selects one of the products identified in block808 to determine an opportunity metric for the product with respect to the household A102 (block810). Based on the points of sale data and the segment of thehousehold A102, thecollaborative filter126 determines the expected and/or average purchases of the selected product for thehousehold A102 and/or the segment to which thehousehold A102 belongs (block812). Based on the expected and/or average purchases of the selected product and the current purchases of the selected product by thehousehold A102, thecollaborative filter126 determines the opportunity metric for the selected product in the selected household A102 (block814). The opportunity metric is based on the difference between the expected purchases (e.g., weekly) by thehousehold A102 and the actual purchases by thehousehold A102.
• Thecollaborative filter126 then determines whether there are additional identified products for which an opportunity metric must be generated (block816). If there are additional products (block816), control returns to block810 to select another identified product. Blocks810-816 iterate to determine opportunity metric(s) for the products identified in block808. When there are no more products for which an opportunity metric is to be generated (block816), control returns to block704 ofFIG. 7.
• FIG. 9 is a flowchart representative of example machinereadable instructions900 which may be executed to select an advertisement for delivery to a household. Theexample instructions900 may be executed by, for example, theprocessor1102 ofFIG. 11 to implementadvertisement selector202 ofFIG. 2 and/or block704 ofFIG. 7.
• Theadvertisement selector202 first selects a household and receives opportunity metric information for the household A102 (e.g., thehousehold A102 ofFIG. 1) (block902). Thehousehold A102 may be selected based on opportunity metric information received from, for example, thecollaborative filter126. Theadvertisement selector202 selects a first product from the opportunity metrics associated with the household A102 (block904). For example, theadvertisement selector202 may select the product having the highest opportunity metric with respect to thehousehold A102 or may select a product at random from a list of products for which an opportunity metric is provided.
• Theadvertisement selector202 then determines whether the opportunity metric associated with the selected product is greater than a threshold (block906). If the opportunity is not greater than the threshold (block906), control returns to block904 to select another product. If the opportunity metric for the selected product is greater than the threshold (block906), theadvertisement selector202 selects an advertisement associated with the selected product (block908). For example, theadvertisement selector202 may select an advertisement associated with the product from theadvertisement database206 ofFIG. 2. In some examples, theadvertisement database206 includes multiple advertisements associated with a particular product and theadvertisement selector202 picks between the same as explained below.
• Theadvertisement selector202 determines the saturation of thehousehold A102 with respect to the selected advertisement (block910). For example, theadvertisement selector202 may have previously received a count or other indication of a number of times that the selected advertisement has been presented to thehousehold A102. Presentation of the advertisement to thehousehold A102 may be counted as a result of general broadcast advertising and/or targeted advertising using the systems and methods disclosed herein. Thus, thesystem100 ofFIG. 1 preferably includes an audience measurement system that monitors media exposure in the household in order to count advertisement exposure through general broadcasts. Such an audience measurement systems may be any type of system such as the system(s) employed by the Nielsen Company, LLC, to develop television ratings and/or to perform advertisement broadcast monitoring (e.g., the Monitor-Plus® system).
• Based on the collected data for thehousehold A102 and/or extrapolation to thehousehold A102 if thehousehold A102 is not directly monitored by an audience measurement system, theadvertisement selector202 determines a net effectiveness metric for the selected advertisement to represent the saturation of the household A102 (block912). For example, if using discrete saturation levels, theadvertisement selector202 may apply a penalty to the advertisement (e.g., subtract the saturation metric from the corresponding product opportunity metric) using a function of the number of times the selected advertisement has been shown (or is estimated to have been shown) to thehousehold A102 and the number of times that an advertisement must be shown to thehousehold A102 before saturation begins. If the advertisement uses binary saturation levels, theadvertisement selector202 may apply a high penalty to (e.g., disqualify) an advertisement to which thehousehold A102 is saturated and apply no penalty to an advertisement to which thehousehold A102 is not saturated. The penalty may be applied by subtracting a value associated with the penalty from the corresponding opportunity metric for the product in question.
• After determining the saturation metric and the net effectiveness metric, theadvertisement selector202 determines whether the net effectiveness metric (i.e., the opportunity metric as modified by the penalty) is sufficiently high for the advertisement to overcome the saturation of the household A102 (block914). For example, theadvertisement selector202 may determine if the net effectiveness metric for the advertisement is the highest net effectiveness metric calculated for thehousehold A102. If the opportunity for a product is still sufficiently high at thehousehold A102, the net effectiveness metric is sufficiently high and the selected advertisement may still influence purchasing decisions despite having been presented to the household A102 a number of times.
• After determining the net effectiveness metric, theadvertisement selector202 determines whether there are additional advertisements that may be selected for the selected product (block914). In some examples, a first advertisement for a selected product may have reached saturation at thehousehold A102, but a second advertisement for the selected product has not yet reached saturation. If there are additional advertisements available for the selected product (block914), control returns to block908 to select another advertisement for the selected product. In contrast, if there are no additional advertisements available (block914), theadvertisement selector202 determines whether there are additional products for which an opportunity metric is available (block916).
• If, atblock916, there are no additional products, theadvertisement selector202 identifies an advertisement for a product, based on the net effectiveness metrics and one or more thresholds, to themedia deliverer204 for transmission to the selected household A102 (block918). For example, theadvertisement selector202 may identify one or more advertisements that have a net effectiveness metric greater than a threshold. In contrast, if one or more advertisements are less than the threshold, they may not be identified as acceptable for delivery to thehousehold A102. In some examples, the threshold is based on the reach, the lower limit of effectiveness desired by an advertiser, and/or other factors that may be defined in an advertising agreement. When theadvertisement selector202 has identified an advertisement for the selected household A102 (block918), theinstructions900 may end and control returns to block706 ofFIG. 7.
• FIG. 10 is a flowchart representative of example machinereadable instructions1000 which may be executed to determine a media response of a household to a delivered advertisement. Theexample instructions1000 may be executed by, for example, theprocessor1102 ofFIG. 11 to implement themedia response evaluator212 ofFIG. 2.
• Themedia evaluator212 receives a media response characteristic from, for example, a media research organization or other media response characteristic provider (block1002). In some examples, the media response characteristic is determined by statistical models that describe the effectiveness of an advertisement per presentation as a characteristic of the number of times that the advertisement is presented to a given person. In some models, as the number of times the advertisement is shown to a household increases, the effectiveness of each additional showing decreases (e.g., f(n)˜1/n, where f(n) is the effectiveness of the next presentation and n is the total number of presentations) and/or the overall effectiveness of all presentations of the advertisement flattens out (e.g., t(n)˜log(n), where t(n) is the total effectiveness of all presentations and n is the total number of presentations). However, in some other models such as the exampleadvertisement saturation characteristic324 ofFIG. 3B, the effectiveness f(n) resembles an S-shaped curve where the greatest effectiveness lies after the first presentation of an advertisement.
• In some examples, themedia evaluator212 deduces a media response characteristic of thehousehold A102 based on the changing purchases of thehousehold A102, the advertisements presented to thehousehold A102, the segment of thehousehold A102, and/or the media response characteristic of the segment of thehousehold A102. For example, some households may respond to advertising more readily and saturate with respect to a given advertisement more quickly than average. Thus, it may be desirable to avoid repeating advertisements to such households very often, and new advertisements could potentially command a higher price for presentation than repeated advertisements. In contrast, some households may be less prone to saturation with respect to any given advertisement than average. In such a case, themedia evaluator212 may penalize advertisements less than for typical households for the same number of presentations.
• Themedia evaluator212 receives an identification of an advertisement delivered to the household A102 (block1004). For example, themedia evaluator212 may be notified via metadata that an advertisement was delivered to thehousehold A102 at a particular time. Themedia evaluator212 determines the number of times the advertisement has been shown to thehousehold A102 in a measured period (e.g., one day, one week, two weeks, one month, etc.) (block1006). Based on the number of times the advertisement has been shown (e.g., including the most recent presentation) and the media response characteristic, themedia evaluator212 determines a media response value of thehousehold A102 to future presentations of the advertisement (block1008). The media response values may be stored for future reference by theadvertisement selector202. In some examples, themedia evaluator212 only determines the household media response on demand when theadvertisement selector202 selects an advertisement for potential delivery based on a product with a high opportunity metric. In such examples, themedia evaluator212 sends the future response to theadvertisement selector202. When themedia evaluator212 has predicted the future response, theexample instructions1000 end and control returns to block710 ofFIG. 7.
• FIG. 11 is a diagram of anexample processor system1100 that may be used to execute the example machinereadable instructions700,800,900, and/or1000 described inFIGS. 7-10, as well as to implement thesystem100 ofFIGS. 1-3A and theheadend system200 described inFIG. 2. Theexample processor system1100 includes aprocessor1102 having associated memories, such as a random access memory (RAM)1104, a read only memory (ROM)1106 and aflash memory1108. Theprocessor1102 in communication with an interface, such as abus1112 to which other components may be interfaced. In the illustrated example, the components interfaced to thebus1112 include aninput device1114, adisplay device1116, amass storage device1118, a removablestorage device drive1120, and anetwork adapter1122. The removablestorage device drive1120 may include associatedremovable storage media1124 such as magnetic or optical media. Thenetwork adapter1122 may connect theprocessor system1100 to anexternal network1126.
• Theexample processor system1100 may be, for example, a desktop personal computer, a notebook computer, a workstation or any other computing device. Theprocessor1102 may be any type of logic device, such as a microprocessor from the Intel® Pentium® family of microprocessors, the Intel® Itanium® family of microprocessors, and/or the Intel XScale family of processors. Thememories1104,1106 and1108 that are in communication with theprocessor1102 may be any suitable memory devices and may be sized to fit the storage demands of thesystem1100. In particular, theflash memory1108 may be a non-volatile memory that is accessed and erased on a block-by-block basis.
• Theinput device1114 may be implemented using a keyboard, a mouse, a touch screen, a track pad, a barcode scanner or any other device that enables a user to provide information to theprocessor1102.
• Thedisplay device1116 may be, for example, a liquid crystal display (LCD) monitor, a cathode ray tube (CRT) monitor or any other suitable device that acts as an interface between theprocessor1102 and a user. Thedisplay device1116 includes any additional hardware required to interface a display screen to theprocessor1102.
• Themass storage device1118 may be, for example, a hard drive or any other magnetic, optical, or solid state media that is readable by theprocessor1102.
• The removablestorage device drive1120 may, for example, be an optical drive, such as a compact disk-recordable (CD-R) drive, a compact disk-rewritable (CD-RW) drive, a digital versatile disk (DVD) drive or any other optical drive. It may alternatively be, for example, a magnetic media drive and/or a solid state universal serial bus (USB) storage drive. Theremovable storage media1124 is complimentary to the removablestorage device drive1120, inasmuch as themedia1124 is selected to operate with thedrive1120. For example, if the removablestorage device drive1120 is an optical drive, theremovable storage media1124 may be a CD-R disk, a CD-RW disk, a DVD disk or any other suitable optical disk. On the other hand, if the removablestorage device drive1120 is a magnetic media device, theremovable storage media1124 may be, for example, a diskette or any other suitable magnetic storage media.
• Thenetwork adapter1122 may be, for example, an Ethernet adapter, a wireless local area network (LAN) adapter, a telephony modem, or any other device that allows theprocessor system1100 to communicate with one or more other processor systems over a network. Theexternal network1126 may be a LAN, a wide area network (WAN), a wireless network, or any type of network capable of communicating with theprocessor system1100. Example networks may include the Internet, an intranet, and/or an ad hoc network.
• Although this patent discloses example systems including software or firmware executed on hardware, it should be noted that such systems are merely illustrative and should not be considered as limiting. For example, it is contemplated that any or all of these hardware and software components could be implemented exclusively in hardware, exclusively in software, exclusively in firmware or in some combination of hardware, firmware and/or software. Accordingly, while the above specification described example systems, methods and articles of manufacture, the examples are not the only way to implement such systems, methods and articles of manufacture. Therefore, although certain example methods, apparatus and articles of manufacture have been described herein, the scope of coverage of this patent is not limited thereto. On the contrary, this patent covers all methods, apparatus and articles of manufacture fairly falling within the scope of the claims either literally or under the doctrine of equivalents.

Claims (20)

What is claimed is:

1. An apparatus comprising:

at least one memory;

instructions in the apparatus; and

processor circuitry to execute the instructions to at least:

access an address of a media device associated with a first household;

determine, based on first purchase activity data associated with the first household, a first quantity of a first product purchased by the first household during a first period of time;

determine, based on second purchase activity data associated with a plurality of consumers included in a consumer segment associated with the first household, a second quantity of the first product expected to be purchased by the first household during the first period of time;

determine an opportunity metric for the first product based on the first quantity of the first product purchased by the first household during the first period of time and the second quantity of the first product expected to be purchased by the first household during the first period of time;

select, based on at least the opportunity metric, a first advertisement associated with the first product to deliver to the media device; and

transmit the first advertisement to the address of the media device.

2. The apparatus ofclaim 1, wherein the media device corresponds to a set top box, and the address corresponds to at least one a media access control address or an Internet protocol address.

3. The apparatus ofclaim 1, wherein the processor circuitry is to determine the consumer segment associated with the first household based on demographic data obtained for the first household.

4. The apparatus ofclaim 1, wherein the processor circuitry is to determine the opportunity metric based on a difference between the second quantity and the first quantity.

5. The apparatus ofclaim 1, wherein the first purchase activity data is reported from a logging device that is to log product purchase activity at the first household, the opportunity metric is a first opportunity metric, and the processor circuitry is to:

access point-of-sale data to identify a second product associated with the first product; and

determine a second opportunity metric for the second product to be purchased by the first household, the second opportunity metric to be determined based on the first purchase activity data reported from the logging device and the second purchase activity data associated with the plurality of consumers included in the consumer segment.

6. The apparatus ofclaim 1, wherein the processor circuitry is to determine a saturation metric for the first advertisement, the saturation metric associated with the first household, the saturation metric to be determined based on detection of identification codes embedded in prior transmissions of the first advertisement to the media device.

7. The apparatus ofclaim 6, wherein to select the first advertisement, the processor circuitry is to:

determine the saturation metric based on a number of the prior transmissions of the first advertisement received by the media device;

adjust the opportunity metric based on the saturation metric to determine an effectiveness metric for the first advertisement; and

select the first advertisement when the effectiveness metric satisfies a threshold.

8. The apparatus ofclaim 7, wherein to determine the effectiveness metric, the processor circuitry is to subtract the saturation metric from the opportunity metric.

9. At least one storage device or storage disk comprising computer readable instructions that, when executed, cause at least one processor to at least:

access an address of a media device associated with a first household;

determine, based on first purchase activity data associated with the first household, a first quantity of a first product purchased by the first household during a first period of time;

determine, based on second purchase activity data associated with a plurality of consumers included in a consumer segment associated with the first household, a second quantity of the first product expected to be purchased by the first household during the first period of time;

determine an opportunity metric for the first product based on the first quantity of the first product purchased by the first household during the first period of time and the second quantity of the first product expected to be purchased by the first household during the first period of time;

select, based on at least the opportunity metric, a first advertisement associated with the first product to deliver to the media device; and

transmit the first advertisement to the address of the media device.

10. The at least one storage device or storage disk ofclaim 9, wherein the media device corresponds to a set top box, and the address corresponds to at least one a media access control address or an Internet protocol address.

11. The at least one storage device or storage disk ofclaim 9, wherein the instructions cause the at least one processor to determine the consumer segment associated with the first household based on demographic data obtained for the first household.

12. The at least one storage device or storage disk ofclaim 9, wherein the instructions cause the at least one processor to determine the opportunity metric based on a difference between the second quantity and the first quantity.

13. The at least one storage device or storage disk ofclaim 9, wherein the first purchase activity data is reported from a logging device that is to log product purchase activity at the first household, the opportunity metric is a first opportunity metric, and the instructions cause the at least one processor to:

access point-of-sale data to identify a second product associated with the first product; and

determine a second opportunity metric for the second product to be purchased by the first household, the second opportunity metric to be determined based on the first purchase activity data reported from the logging device and the second purchase activity data associated with the plurality of consumers included in the consumer segment.

14. The at least one storage device or storage disk ofclaim 9, wherein the instructions cause the at least one processor to determine a saturation metric for the first advertisement, the saturation metric associated with the first household, the saturation metric to be determined based on detection of identification codes embedded in prior transmissions of the first advertisement to the media device.

15. The at least one storage device or storage disk ofclaim 14, wherein the instructions cause the at least one processor to select the first advertisement by:

determining the saturation metric based on a number of the prior transmissions of the first advertisement received by the media device;

adjusting the opportunity metric based on the saturation metric to determine an effectiveness metric for the first advertisement; and

selecting the first advertisement when the effectiveness metric satisfies a threshold.

16. The at least one storage device or storage disk ofclaim 15, wherein the instructions cause the at least one processor to subtract the saturation metric from the opportunity metric to determine the effectiveness metric.

17. A method comprising:

accessing an address of a media device associated with a first household;

determining, by executing an instruction with at least one processor, a first quantity of a first product purchased by the first household during a first period of time, the first quantity determined based on first purchase activity data associated with the first household;

determining, by executing an instruction with the at least one processor, a second quantity of the first product expected to be purchased by the first household during the first period of time, the second quantity determined based on second purchase activity data associated with a plurality of consumers included in a consumer segment associated with the first household ;

determining, by executing an instruction with the at least one processor, an opportunity metric for the first product based on the first quantity of the first product purchased by the first household during the first period of time and the second quantity of the first product expected to be purchased by the first household during the first period of time;

selecting, by executing an instruction with the at least one processor, a first advertisement associated with the first product to deliver to the media device, the selecting based on at least the opportunity metric; and

transmitting the first advertisement to the address of the media device.

18. The method ofclaim 17, wherein the media device corresponds to a set top box, and the address corresponds to at least one a media access control address or an Internet protocol address.

19. The method ofclaim 17, further including determining a saturation metric for the first advertisement, the saturation metric associated with the first household, the saturation metric determined based on detection of identification codes embedded in prior transmissions of the first advertisement to the media device.

20. The method ofclaim 19, wherein the selecting of the first advertisement includes:

determining the saturation metric based on a number of the prior transmissions of the first advertisement received by the media device;

subtracting the saturation metric from the opportunity metric to determine an effectiveness metric for the first advertisement; and

selecting the first advertisement when the effectiveness metric satisfies a threshold.

Images