Note: Descriptions are shown in the official language in which they were submitted.
CA 02743693 2011-06-17
SYSTEMS AND METHODS TO SELECT TARGETED ADVERTISING
FIELD OF THE DISCLOSURE
[00011 This disclosure relates generally to targeted advertising and, more
particularly, to methods and apparatus to select targeted advertising.
BACKGROUND
[0002] 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
[0003] FIG. I is a block diagram of an example system to provide individually-
targeted advertising to different households based on an opportunity metric.
[0004] FIG. 2 is a more detailed block diagram of an example implementation of
the headend system of the cable provider described in connection with FIG. 1.
-1-
CA 02743693 2011-06-17
[0005] FIG. 3A is a block diagram illustrating example data flows that may
occur
in the example systems illustrated in FIGS. I and 2.
[0006] FIG. 3B illustrates an example advertisement saturation characteristic
for
an advertisement of a product with respect to a household.
[0007] FIG. 4 is a table illustrating example weekly purchases for several
households and products.
[0008] FIG. 5 is a table illustrating example expected weekly purchases per
household by segment for several products.
[0009] FIG. 6 is a table illustrating an example advertising log for several
households and several advertisements.
[0010] FIG. 7 is a flowchart representative of example machine readable
instructions which may be executed to select targeted advertising for display.
[0011] 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.
100121 FIG. 9 is a flowchart representative of example machine readable
instructions which may be executed to select an advertisement for delivery to
a
household.
100131 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.
[0014] FIG. 1 1 is a diagram of an example processor system that may be used
to
execute the example instructions of FIGS. 7, 8, 9, and 10 to implement the
example
systems of FIGS. I and 2.
CA 02743693 2011-06-17
DETAILED DESCRIPTION
[00151 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.
[00161 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.
[0017] 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.
-3-
CA 02743693 2011-06-17
[00181 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.
100191 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 advert isements 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
-4-
CA 02743693 2011-06-17
calculations may keep pace with the rate at which consumers make their
purchases
and respond to advert isements.
[0020] 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.
[0021] FIG. 1 is a block diagram of an example system 100 to provide
individually-targeted advertising to different households A 102 and B 104
based on an
opportunity metric. The example system 100 includes a content provider 106
such as
a cable television company to deliver programming and advertisements to a
number
of households A 102 and B 104. The households A 102 and B 104 are provided
with
respective set top boxes 108 and 1 10 coupled to televisions 112 and 114. The
content
provider 106 may broadcast programming via a headend system 107 over a
plurality
of channels, to which the set top boxes 108 and 110 may tune to receive the
broadcast
programming. Additionally or alternatively, the content provider 106 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 set top boxes 108 or 110_ The tuned
or
delivered programming is displayed on the respective televisions 112 and 114.
The
television could alternatively or additionally be implemented by personal
computer
monitors or the like.
100221 In the example system 100, the content provider 106 sells time on each
of
one or more channels to one or more advertisers. The content provider 106 may
agree
CA 02743693 2011-06-17
to broadcast particular advertisements to the set top boxes 108 and 1 10 and,
thus, the
televisions 112 and 114 for display to consumers.
[0023] 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.
[0024] 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,
the example system 100 of FIG. I includes an opportunity calculator 118 to
determine
opportunity metrics for the individual households A 102 and B 104. The
opportunity
metrics are representative of the opportunity a household represents to
advertisers of
particular products based on purchase data 120 and 122 for respective
individual
households A 102 and B 104 and expected purchase data 124.
-6-
CA 02743693 2011-06-17
100251 The household purchase data 120 and 122 identifies purchases made by
the
respective example households A 102 and B 104. The purchase data 120 and 122
may be provided to the opportunity calculator 118 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.
100261 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.
100271 The expected purchases, as described in more detail below. may be
determined by a collaborative filter 126 that determines a quantity of a
product that
the households A 102 and B 104 might be expected to consume (e.g., per day,
per
week, per month, per year) based on the segment(s) in which the households A
102
and B 104 may be classified and what those segment(s) typically purchase. By
-7-
CA 02743693 2011-06-17
determining the difference between what quantity of a given product the
respective
households A 102 and B 104 are expected to purchase (e.g., expected purchases
124
of FIG. 1) and what quantity of that product the households A 102 and B 104
actually
purchased as reflected in the purchase data 120 and 122, the opportunity
calculator
118 generates an opportunity metric. The opportunity metric is provided to the
content provider 106 to enable the content provider 106 to select the most
appropriate
(e.g., effective) advertisements to be delivered to the households A 102 and B
104.
100281 To evaluate the relationships between different items, the
collaborative
filter 126 may be populated and/or updated with the purchasing relationships
between
different goods and/or brands. The collaborative filter 126 may be maintained
by, for
example, a research service or agency that performs market research of
consumer
segments and behaviors.
[00291 FIG. 2 is a more detailed block diagram of an example headend system
200
to implement the headend system 107 of the content provider 106 described in
connection with FIG. 1. The headend system 200 may be used to select and
display
advertisements for individual households (e.g., the households A 102 and B
104)
based on one or more opportunity metrics. For clarity and brevity, the
following
description of FIG. 2 will reference an example where the headend system 200
considers the household A 102 of FIG. 1. The example headend system 200
includes
an advertisement selector 202 to select an advertisement for delivery to a
household
via a media deliverer 204. The media deliverer 204 may be implemented via, for
example, a cable broadcast system. In the illustrated example, the media
deliverer
204 is in communication with multiple subscriber set top boxes (e.g., the set
top boxes
108 and 110 of F IG. 1),
-8-
CA 02743693 2011-06-17
[00301 The advertisement selector 202 is provided with an advertisement
database
206, from which the advertisement selector 202 may choose advertisements
associated with one or more products. The advertisements may be provided to
the
database 206 by one or more advertisers (e.g., the advertisers 116 of FIG. 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.
[00311 The advertisement selector 202 receives several inputs to determine an
advertisement that should be delivered to a particular household. For example,
the
advertisement selector 202 receives the selection of a household A 102 from a
household selector 208. The selection of the household focuses the
advertisement
selector 202 on the household A 102 that should be considered for individual
advertisement delivery. As mentioned above, the advertisement selector 202
receives
the opportunity metric(s) for one or more products. Because the opportunity
metric(s)
are specific to an individual household, the advertisement selector 202
considers the
opportunity metric(s) corresponding to the household selected by the household
selector 208.
[00321 In the illustrated example, the advertisement selector 202 also
receives
programming association information from an ad/programming associator 210. The
ad/programming associator 210 identifies advertisements for products in the
advertisement database 206 which are appropriate for programming currently
shown
to the household A 102. For example, if a first advertisement and a second
advertisement in the advertisement database correspond to equal or similar
opportunity metric(s) for the household A 102, the associator 210 may identify
an
association between the first advertisement and the program being viewed at
the
household that causes the selector 202 to select the first advertisement for
delivery.
-9-
CA 02743693 2011-06-17
For example, if a young children's program is being viewed at the household A
102,
the associator 210 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, the selector 202 may give weight to the association and select
the
advertisement for cereal.
100331 The example advertisement selector 202 further receives advertisement
response data for the household A 102 from a media response evaluator 212. The
media response evaluator 212 monitors the advertisements that are shown to the
household A 102 (e.g., that are sent from the advertisement selector 202 to
the media
deliverer 204 and/or that are sent to the household A 102 during the course of
regular
programming and advertisement) and reports to the selector 202 whether the
advertisement being considered has been shown to the household A 102 enough to
cause the advertisement to lose its effectiveness with respect to the
household A 102.
For example, when the advertisement selector 202 sends a selected
advertisement to
the media deliverer 204 for delivery to the household A 102, the selected
advertisement (or, alternatively, metadata identifying the advertisement) is
also sent to
the media response evaluator 212. The media response evaluator 212 uses one or
more marketing mix models or saturation algorithms to determine whether the
household A 102 has been presented with the advertisement often enough to
reduce
the effectiveness of the advertisement on future presentations.
100341 As an example of the operation of the media response evaluator 212,
assume that the advertisement selector 202 is considering two advertisements;
one for
product X and one for product Y for delivery to the household A 102. The
opportunity metric for the household A 102 associated with product X is higher
than
the opportunity metric associated with the product Y. However, the
advertisement
-10-
CA 02743693 2011-06-17
associated with product X has been shown to the household A 102 many times,
while
the advertisement associated with product Y has not yet been shown to the
household
A 102. As a result, the media response evaluator 212 determines that the
effectiveness of the advertisement for product X is reduced because the
household A
102 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.
[00351 In operation, the example headend system 200 receives opportunity
metric(s) that are individualized for several households. The household
selector 208
selects one of the households to be considered by the advertisement selector
202. The
advertisement selector 202 receives (e.g., requests, retrieves from storage,
etc.) the
opportunity metric(s) associated with the selected household. The
advertisement
selector 202 evaluates the products for which an opportunity metric is
provided and
determines whether the household is saturated with the advert isement(s) for
any of the
product(s). Saturation information may be requested from the media response
evaluator 212, which determines the saturation and/or the future media
response of
the household to the potential advertisement(s). The media response evaluator
212
may perform the evaluation on request from the advertisement selector 202
and/or
have previously prepared evaluations stored. For those advertisement(s) that
have
reached saturation, the advertisement selector 202 applies a penalty.
100361 The advertisement selector 202 loads a pricing structure, pricing
characteristic, or pricing framework from the pricing database 214 and
determines,
based on the opportunity metric(s) and the saturation of the advertisements,
which
-ll-
CA 02743693 2011-06-17
advertisement(s) may command the highest price from advertisers due to the
likelihood that the advertisement(s) will be effective. The advertisement
selector 202
provides the selected advert isement(s) to the media deliverer 204, which
determines
the set top box address associated with the selected household from the
address
database 216 and delivers the advertisement(s) to the address at the
appropriate time.
The advert isement(s) are further provided to the media response evaluator 212
to
determine the response of the selected household to further presentations of
the
selected advert isement(s).
[0037] FIG. 3A is a block diagram illustrating example data flows that may
occur
in the example systems illustrated in FIGS. I and 2. The example data flows
illustrated in FIG. 3A show the exchanges of data between the household A 102,
the
content provider 106, the set top boxes 108 and 110, and the advertisers 116.
While
the example household A 102 is illustrated in FIG. 3A, the example exchanges
of data
may be used in a similar or identical manner with respect to additional
households
(e.g., the household B 104). The data flows 300 of FIG. 3A further detail the
exchange of data between the advertisement selector 202, the media deliverer
204, the
advertising database 206, the household selector 208, the ad/programming
associator
210, the media response evaluator 212, the pricing database 214, and the
address
database 216 within the example headend system 200 of FIG. 2,
100381 For clarity and brevity, the example data flows 300 of FIG. 3A will
refer to
the selection of advertisement(s) for delivery to the household A 102 of FIG.
I
without regard to the household B 104. However, the examples described herein
may
be extended to include any number of households.
[0039] The example household A 102 of FIG. I is associated with purchase data
302 (e.g.. the products and quantities purchased by the household during a
given time
-12-
CA 02743693 2011-06-17
period, etc.) and geodemographic data 304 (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.). The purchase data 302 and the geodemographic
data
304 may be provided by, for example, self-reporting by the household A 102
and/or
monitoring techniques practiced within the household A 102. The purchase data
302
may additionally or alternatively be provided through the use of frequent
shopper card
data associated with frequent shopper cards carried by members of the
household A
102.
100401 Based on the purchase 302 and demographic data 304, the household A 102
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 data 306 may be developed and/or updated to reflect, for example, new
product offerings, purchasing trends, and/or changes in membership
characteristics.
Additionally, retailer purchase panel data 308 is collected at different
points of sale.
In some examples, the product purchase data 308 is collected from as many
retailers
or other points of sale as possible. Additionally or alternatively, the
retailer purchase
panel data 308 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 the household A 1 02 and/or the segment.
100411 The purchase data 302 and the geodemographic data 304 from the
household A 102, the relevant consumer segment data 306, and the relevant
retailer
purchase panel data 308 are input to a collaborative filter 126. The purchase
data 302
provided to the collaborative filter 126 may include product identifiers and
quantities
-13-
CA 02743693 2011-06-17
per time period. The geodemographic data 304 of the illustrated example
includes
sufficient geodemographic information to place the household A 102 into a
consumer
segment. The example consumer segment data 306 of the illustrated example
provides at least the segment purchase information of the products purchased
by the
household A 102. The consumer segment data 306 may additionally provide
available product purchase information for the consumer segment into which the
household A 102 fits. The retailer purchase panel data 308 provides bills of
sale that
illustrate the combinations of products that are often purchased together.
[00421 The collaborative filter 126 receives the purchase data 302 and the
geodemographic data 304 from the household A 102, the consumer segment data
306,
and the retailer purchase panel data 308, and determines one or more
opportunity
metric(s) for one or more products based on the received information. For
example,
the purchase data 302 includes a quantity of a product X (e.g., a brand of a
product)
purchased weekly by the household A 102. The geodemographic data 304 allows
the
collaborative filter 126 (or some other classification device) to generate a
purchase
expectation for the household A 102 based on placement of the household A 102
into
a segment (e.g., segment N). The placement of the household A 102 into the
segment
N may be based on other data besides geodemographic data, such as the purchase
data
302.
100431 The consumer segment data 306 provides the typical (e.g., average)
weekly
purchases of one or more products, including product X, by households in the
segment N. Thus, the collaborative filter 126 may determine the amount by
which the
household A 102 is below or above the typical weekly purchases for the segment
N.
Retailer purchase panel data 308 allows the collaborative filter 126 to
evaluate the
combinations and quantities of products that are often purchased together,
which may
-14-
CA 02743693 2011-06-17
then be applied to the purchase data 302 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.
[0044] Unlike known collaborative filters, the example collaborative filter
126
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, the collaborative filter 126 of FIG. 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, the
collaborative
filter 126 may also determine or account for products that are generally
purchased
infrequently.
[0045] The collaborative filter 126 determines, based on the purchases 302 of
the
household A 102 and historical purchasing trends of the related segment N.
items that
may be desirable substitutes for and/or supplements to products the household
A 102
currently purchases. By determining the segment (e.g., N) of the household A
102,
the collaborative filter 126 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 the household A 102 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 the household A 102. The
-15-
CA 02743693 2011-06-17
collaborative filter 126 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 retailer purchase panel data 308. In some examples, the retailer
purchase
panel data 308 is at least partially used to determine the consumer segment
data 306.
100461 The collaborative filter 126 outputs one or more opportunity metrics
312
for each product (e.g.. products X. B, C, and D) that are identified as having
an
opportunity associated with the household A 102. The opportunity metric of a
product X is based on the purchase data 302 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.
100471 For example, assume the household A 102 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 the household A 102 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, the collaborative filter 126
determines that
the opportunity for motivating the household A 102 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 the
household A 102 purchases none of soft drink X compared to the average of 12
cans
of soft drink X per week. Therefore, advertising soft drink X to the household
A 102
may be highly likely to influence the household A 102 to increase its
purchases of soft
drink X. Additionally, the collaborative filter 126 determines that the
opportunity for
potato chips B is high because the average purchases for potato chips B among
-16-
CA 02743693 2011-06-17
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 the household A 102 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 the
collaborative filter 126.
[00481 The opportunity metric(s) 312 developed by the opportunity calculator
118
of FIG. I are input to the advertisement selector 202 of FIG. 2. In addition
to the
opportunity metric(s) 312, the advertisement selector 202 receives pricing
information
314 (e.g., from the pricing database 214 of FIG. 2), advert
isement/programming
associations 316 (e.g.. from the ad/programming associator 210 of FIG. 2),
advertisements 318 (e.g., from the advertisement database 206 of FIG. 2), and
household saturation information (e.g., a saturation metric) from an
advertising
response monitor 320 (e.g.. from the media response evaluator 212 of FIG. 2).
The
selection of the household A 102 (e.g., from the household selector 208 of
FIG. 2) is
identified expressly or implicitly in the delivery of the opportunity
metric(s) 312 for
household A 102.
100491 While the example opportunity metric(s) 312 are shown in FIG. 3A as a
score, the opportunity metric(s) 3 12 may he presented or measured in any
suitable
manner. For example, the opportunity metric(s) of the household A 102 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.
-17-
CA 02743693 2011-06-17
[0050] The pricing information 314 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, the pricing information 314 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.
[00511 In some example advertisement pricing models, a media provider (e.g.,
the
media provider 200 of FIG. 2) contracts with an advertiser (e.g., the
advertisers 116)
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,
the media provider 200 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 the advertiser 116.
The
advertiser 116 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 the media provider 200 and the advertisers
116.
[0052] In some examples, the media provider 200 may determine the opportunity
metric for a type of product as opposed to a particular brand. For example,
cola is a
-18-
CA 02743693 2011-06-17
type of product where Coke and Pepsi are particular brands. To increase
revenue,
the media provider 200 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.
100531 The advertising/programming associations 316 may include, for example,
broadcast programs that advertisers identify as preferable to identify with
advertised
products. For example, the ad/programming associations 316 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.
[00541 The advertisements 318 include at least the advertisements for products
having an opportunity metric provided by the collaborative filter 126. In some
examples, the advertisements 318 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. the advertisement selector
202
may show different advertisements for the same product to the household A 102,
thereby decreasing the saturation of the household A 102 to advertisements for
a
product. By changing the advertisements for a product, the advertisement
selector
202 may maintain higher revenue for an advertisement mix sent to the household
A
102 by maintaining a high effectiveness of the advertising mix.
100551 The ad response monitor 320 monitors the advertising sent to the
household
A 102. The ad response monitor 320 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
-19-
CA 02743693 2011-06-17
manner, such as by reading an identification code embedded in (or otherwise
broadcast with) the broadcast advertisement. Based on the advertising sent to
the
household A 102, the ad response monitor 320 determines saturation metric(s)
of the
household A 102 with respect to one or more advertisements. As the number of
times
a particular advertisement is presented to household A 102 increases, the ad
response
monitor 320 determines that the incremental effectiveness of that
advertisement
decreases with additional showings (i.e., the total effectiveness increases
more
slowly). The example ad response monitor 320 thus provides a penalty to be
applied
to certain advertisements 3 18 that may be selected by the advertisement
selector 202
when the advertisements 3 18 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
monitor 320 may change over time. Therefore, the ad response monitor 320 may
be
provided and/or updated by, for example, a consumer research service or agency
that
specializes in consumer behavior.
100561 As the number of times an advertisement 318 is shown increases beyond a
saturation point, the ad response monitor 320 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 monitor 320 to apply the penalty. An example of
such a
saturation characteristic is illustrated in FIG. 3B. Thus, while
advertisements for
products having mid- to high-range opportunity metrics with respect to the
household
A 102 may be sufficiently high to overcome the penalties and, thus, continues
to be
shown, after an advertisement 318 has been shown often enough the price for
that
-20-
CA 02743693 2011-06-17
advertisement 3 18 will be overtaken by another advertisement offering a
higher price
because it is expected to exhibit higher advertising effectiveness.
100571 When the advertisement selector 202 has received the pricing
information
314, the advertisement/programming associations 3 16. the advertisements 318,
and
the saturation metric(s) from the ad response monitor 320, the advertisement
selector
202 selects an advertisement for delivery and/or presentation to the household
A 102
(e.g., via the set top box 108 and the television 112 of FIG. 1). The selected
advertisement and the destination for the advertisement (e.g., household A
102) are
provided to the media deliverer 204. The media deliverer 204 also receives
household
address information 322 corresponding to the set top box 108 in the household
A 102.
The address information may include, for example, a media access control (MAC)
address, an Internet protocol (I P) address, or any other type of network
layer or other
type of address that uniquely identifies the set top box 108 of household A
102. The
media deliverer 204 delivers the selected advertisement to the set top box 108
in
household A 102 at the appropriate time, such as shortly prior to the time
space sold
by the content provider 200.
100581 The media deliverer 204 may also be responsible for broadcasting (e.g.,
to a
large portion of the possible audience) programming and/or advertisements to
the
households 102 and 104. 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 the
household A
102, the media deliverer 204 further provides and/or identifies the selected
advertisements and broadcast advertisements to the ad response monitor 320. As
mentioned above, the ad response monitor 320 monitors the advertisements
presented
to household A 102 and provides household saturation information (e.g., a
saturation
-21-
CA 02743693 2011-06-17
metric) to the advertisement selector 202 to select future advertisements for
delivery
to the household A 102. Thus, the ad response monitor 320 uses the identified
advertisements to update the saturation metric or level.
[0059] The advertisements provided to the set top box 108 and/or presented to
the
household A 102 may be further fed back to the advertisement selector through
advert isement-driven purchasing by the household A 102. For example, when an
advertisement stimulates purchases of product X that previously had a high
opportunity metric, the collaborative filter 126 may determine that the
opportunity
metric for the household A 102 for product X decreases because the quantity of
product X that the household A 102 is expected to purchase has been constant
for the
relevant time period, unless the household A 102 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 the household A 102 at a higher effectiveness and, thus,
a higher
price.
[00601 The example data flows illustrated in FIG. 3A may be performed by any
one or more parties- In some examples, a cable or other media provider (e.g.,
the
content provider 106) of FIG. I may implement at least the advertisement
selector
202, the media deliverer 204, and the ad response monitor 320 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 the purchase data 302, the geodemographic data 304, the
consumer segment data 306, the retail purchase data 308, and/or implement the
collaborative filter 126 to generate opportunity metrics. The media research
organization then provides the opportunity metrics to a content provider to
select and
-22-
CA 02743693 2011-06-17
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.
100611 FIG. 3B illustrates an example advertisement saturation characteristic
324
for an advertisement of a product X with respect to the example household A
102.
The horizontal axis is representative of the number of presentations of the
advertisement to the household A 102. The vertical axis is representative of
the
expected likelihood that the household A 102 will exhibit a response or
behavior (e.g.,
purchasing the advertised product). As illustrated in FIG. 313, 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.).
[0062] The example saturation characteristic 324 may be provided to, for
example,
the ad response monitor 320 to determine a saturation metric for the product X
and
the household A 102. The ad response monitor 320 determines the number of
presentations of the advertisement to the household A 102 and generates a
saturation
metric (e.g., a penalty). The ad response monitor 320 provides the saturation
metric
to the advertisement selector 202. 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 the example household A 102. The
advertisement selector 202 may then select an advertisement for delivery to
the
-23-
CA 02743693 2011-06-17
household A 102 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.
[0063] According to the example saturation characteristic 324, the ad response
monitor 320 may cause the saturation metric of the household A 102 to the
advertisement to be higher after the household A 102 has been presented the
advertisement five times than after the household A 102 has been presented the
advertisement two times. However, according to the saturation characteristic
324, the
saturation metric after the household A 102 has been presented the
advertisement
once may be very similar to the saturation metric after the household A 102
has been
presented the advertisement five times.
[0064] The example saturation characteristic 324 may be represented by the
equation v = 1 + e '+m 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.
[00651 While an example saturation characteristic 324 is illustrated in FIG.
3B,
saturation characteristics 324 may be additionally and/or alternatively
represented by,
for example, a mathematical algorithm, a lookup table, or any other
appropriate
representation. Further, the saturation characteristic 324 may differ between
household, segment, and/or product combinations.
[0066] FIG. 4 is a table 400 illustrating example expected weekly purchases
per
household by segment. The example table 400 may be representative of the
consumer
segment data 306 of FIG. 3A for example segments N, Q, R, S, and T. The table
400
-24-
CA 02743693 2011-06-17
includes the expected purchases of several products, including beer Z, chips
B, bread
H, bread J, and milk M. The expected purchases provided by table 400 may be
used
by the collaborative filter 126 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.
100671 FIG. 5 is a table 500 illustrating example weekly purchases for several
households by product. The example table 500 may be provided to the
collaborative
filter 126 and used in combination with the example table 400 to determine the
opportunity metrics of the households 1-5 listed in the table 400 with respect
to the
listed products. The example table 500 further includes the segment that each
household 1-5 falls into based on similar purchases and/or geodemographic
information.
100681 To generate the opportunity metrics for the households 1-5, the
collaborative filter 126 compares the purchases of each of the relevant
products in the
example weekly (or other time period) purchases table 500 with the expected
purchases of the corresponding products found in the appropriate segment row
of the
expected purchases table 400. For example, the collaborative filter 126
compares the
weekly purchases of chips B for household 1. which is in segment N. with the
expected weekly purchases of chips B by households in segment N in the
expected
purchases table 400. 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
-25-
CA 02743693 2011-06-17
than the expected purchases, the opportunity metric for that product with
respect to
that household will be high.
100691 FIG. 6 is a table 600 illustrating an example advertising log for
several
households and several advertisements. The example table 600 may be maintained
by
the example media response evaluator 212 of FIG. 2 to evaluate the media
response of
different households 1-5 at the appropriate times. In the example table 600,
the
advertisements I 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 the content provider 106. In contrast, the
example
advertisement 3 is targeted at the households I 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 the content provider 106 and not presented as part of a blanket
advertising campaign. Any number of advertisements and/or households may be
stored in the example table 600 by the media response evaluator 212 and used
to
evaluate media response to advertisements presented to the households.
100701 When evaluating the media responses according to the example table 600,
the media response evaluator 212 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 of FIG.
6,
when the media response evaluator 212 evaluates advertisements to be delivered
to
household 4, the media response evaluator 212 applies a penalty to
advertisements I,
2, and 9.
-26-
CA 02743693 2011-06-17
100711 While an example manner of implementing the system 100 and the headend
system 107 has been illustrated in FIGS. 1. 2 and/or 3A, one or more of the
elements,
processes and/or devices illustrated in FIGS. 1, 2 and/or 3A may be combined.
divided, re-arranged, omitted, eliminated and/or implemented in any other way.
Further, the example set top boxes 108 and/or 110, the example content
provider 106,
the example headend system 107, the example opportunity calculator 118, the
example collaborative filter 126, the example advertisement selector 202, the
example
media deliverer 204, the example advertisement database 206, the example
household
selector 208, the example advertisement/programming selector 210, the example
media response evaluator 212, the example pricing database 214, the example
address
database 216 and/or, more generally, the example system 100 may be implemented
by
hardware, software, firmware and/or any combination of hardware, software
and/or
firmware. Thus, for example, any of the example set top boxes 108 and/or 110,
the
example content provider 106, the example headend system 107, the example
opportunity calculator 118, the example collaborative filter 126, the example
advertisement selector 202, the example media deliverer 204, the example
advertisement database 206, the example household selector 208, the example
advertisement/programming selector 210, the example media response evaluator
212,
the example pricing database 214, the example address database 216 and/or,
more
generally, the example system 100 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 set top boxes 108
and/or
110, the example content provider 106, the example headend system 107, the
example
-27-
CA 02743693 2011-06-17
opportunity calculator 118, the example collaborative filter 126, the example
advertisement selector 202, the example media deliverer 204, the example
advertisement database 206, the example household selector 208, the example
advertisement/programming selector 210, the example media response evaluator
212,
the example pricing database 214, and/or the example address database 216 are
hereby expressly defined to include a tangible medium such as a memory, DVD,
CD,
etc. storing the software and/or firmware. Further still, the example system
100 may
include one or more elements, processes and/or devices in addition to, or
instead of,
those illustrated in FIGS. 1, 2, and/or 3, and/or may include more than one of
any or
all of the illustrated elements, processes and devices.
100721 FIG. 7 is a flowchart representative of example machine readable
instructions 700 which may be executed to implement the example system 100 of
FIGS. 1-3A. The example instructions 700 may be executed by, for example, the
processor 1 102 of FIG. 11 to implement the headend system 200 illustrated in
FIG. 2
to deliver targeting advertising to individual households. In some examples,
the
instructions 700 may be executed separately for each household for which
targeted
advertising will be selected and delivered. The example instructions 700 begin
by
determining one or more opportunity metric(s) for one or more products for a
household (e.g., the household A 102 or the household B 104 of FIG. 1) (block
702).
An example process to implement block 702 is described in more detail below
with
reference to FIG. 8.
100731 The headend system 200 (e.g., via the advertisement selector 202 of
FIG. 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 (block
704).
The selection of the advertisement may be based further on a pricing
structure, an
-28-
CA 02743693 2011-06-17
advertisement/programming association, and/or a saturation characteristic of
the
household with respect to different advertisements. An example process to
implement
block 704 is described in more detail below with reference to FIG. 9. The
headend
system 200 (e.g., via the media deliverer 204 of FIG. 2) delivers the selected
advertisement to the household (block 706). The media deliverer 204 may
deliver the
selected advertisement during a time slot owned by the content provider 106.
In some
examples, the media deliverer 204 receives an address corresponding to a set
top box
(e.g., the set top box 108 in the household A 102 of FIG. 1) and transmits the
advertisement to the address.
100741 The headend system 200 (e.g., via the media response evaluator 212 of
FIG. 2) then determines the media response of the household to the
advertisement
(block 708). For example, the media response evaluator 212 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 implement block 708 is described in more detail below with
reference to FIG. 9. The media response evaluator 212 then feeds back the
media
response to the advertisement selector 202 (block 710). The advertisement
selector
202 may use the media response in selecting future advertisements for delivery
to the
household (e.g.. future iterations of block 704).
100751 FIG. 8 is a flowchart representative of example machine readable
instructions 800 which may be executed to determine one or more opportunity
metrics
for a household. The example instructions 800 may be executed by, for example,
the
processor 1 102 of FIG. 1 I to implement the collaborative filter 126 of FIG.
3A and/or
block 702 of FIG. 7.
-29-
CA 02743693 2011-06-17
100761 The example collaborative filter 126 receives purchasing information
(e.g.,
the purchasing information 302 of FIG. 3A) and geodemographic information
(e.g.,
the geodemographic information 304 of FIG. 3A) from a household (e.g., the
household A 102 of FIG. 3A) (block 802). The collaborative filter 126 further
receives purchasing and geodemographic data from one or more retail panels
(block
804). In some examples, block 804 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 example
collaborative filter 126 may receive data specific to the segment and/or
geographic
area to which the household A 102 belongs. The collaborative filter 126
determines
the segment to which the household A 102 belongs based on the purchase
information
302 and the geodemographic information 304 (block 806). In some examples, the
segments are determined based on the purchase information 302, the
geodemographic
information 304. and/or the data received from retail panels. Alternatively,
the
segments may be predetermined by an outside provider. The collaborative filter
126
may use additional and/or alternative data to determine the household segment.
100771 The collaborative filter 126 then identifies products for which an
opportunity metric may be calculated (block 808). Products identified by the
collaborative filter 126 may include products purchased by the household A 102
and/or complements to such products. For example, the collaborative filter 126
may
determine, based on the purchase data from different points of sale, that
particular
products purchased by the household A 102 have popular substitutes and/or
complementary products. For instance, the collaborative filter 126 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
-30-
CA 02743693 2011-06-17
complementary item. Similarly, the collaborative filter 126 may determine that
purchasers who buy combinations of certain cereals and candy also tend to
purchase a
particular brand of flavored drink mix.
[00781 The collaborative filter 126 selects one of the products identified in
block
808 to determine an opportunity metric for the product with respect to the
household
A 102 (block 810). Based on the points of sale data and the segment of the
household
A 102, the collaborative filter 126 determines the expected and/or average
purchases
of the selected product for the household A 102 and/or the segment to which
the
household A 102 belongs (block 812). Based on the expected and/or average
purchases of the selected product and the current purchases of the selected
product by
the household A 102, the collaborative filter 126 determines the opportunity
metric
for the selected product in the selected household A 102 (block 814). The
opportunity
metric is based on the difference between the expected purchases (e.g.,
weekly) by the
household A 102 and the actual purchases by the household A 102.
100791 The collaborative filter 126 then determines whether there are
additional
identified products for which an opportunity metric must be generated (block
816). If
there are additional products (block 816), control returns to block 810 to
select
another identified product. Blocks 810-816 iterate to determine opportunity
metric(s)
for the products identified in block 808. When there are no more products for
which
an opportunity metric is to be generated (block 816), control returns to block
704 of
FIG. 7.
[00801 FIG. 9 is a flowchart representative of example machine readable
instructions 900 which may be executed to select an advertisement for delivery
to a
household. The example instructions 900 may be executed by, for example, the
-31-
CA 02743693 2011-06-17
processor 1 102 of FIG. I Ito implement advertisement selector 202 of FIG. 2
and/or
block 704 of FIG. 7.
100811 The advertisement selector 202 first selects a household and receives
opportunity metric information for the household A 102 (e.g., the household A
102 of
FIG. 1) (block 902). The household A 102 may be selected based on opportunity
metric information received from, for example, the collaborative filter 126.
The
advertisement selector 202 selects a first product from the opportunity
metrics
associated with the household A 102 (block 904). For example, the
advertisement
selector 202 may select the product having the highest opportunity metric with
respect
to the household A 102 or may select a product at random from a list of
products for
which an opportunity metric is provided.
10082] The advertisement selector 202 then determines whether the opportunity
metric associated with the selected product is greater than a threshold (block
906). If
the opportunity is not greater than the threshold (block 906), control returns
to block
904 to select another product. If the opportunity metric for the selected
product is
greater than the threshold (block 906), the advertisement selector 202 selects
an
advertisement associated with the selected product (block 908). For example,
the
advertisement selector 202 may select an advertisement associated with the
product
from the advertisement database 206 of FIG. 2. In some examples, the
advertisement
database 206 includes multiple advertisements associated with a particular
product
and the advertisement selector 202 picks between the same as explained below.
100831 The advertisement selector 202 determines the saturation of the
household
A 102 with respect to the selected advertisement (block 910). For example, the
advertisement selector 202 may have previously received a count or other
indication
of a number of times that the selected advertisement has been presented to the
-32-
CA 02743693 2011-06-17
household A 102. Presentation of the advertisement to the household A 102 may
be
counted as a result of general broadcast advertising and/or targeted
advertising using
the systems and methods disclosed herein. Thus, the system 100 of FIG. I
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).
[00841 Based on the collected data for the household A 102 and/or
extrapolation to
the household A 102 if the household A 102 is not directly monitored by an
audience
measurement system, the advertisement selector 202 determines a net
effectiveness
metric for the selected advertisement to represent the saturation of the
household A
102 (block 912). For example, if using discrete saturation levels, the
advertisement
selector 202 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 the household A 102 and the number of times that an
advertisement
must be shown to the household A 102 before saturation begins. If the
advertisement
uses binary saturation levels, the advertisement selector 202 may apply a high
penalty
to (e.g., disqualify) an advertisement to which the household A 102 is
saturated and
apply no penalty to an advertisement to which the household A 102 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.
[00851 After determining the saturation metric and the net effectiveness
metric, the
advertisement selector 202 determines whether the net effectiveness metric
(i.e., the
-33-
CA 02743693 2011-06-17
opportunity metric as modified by the penalty) is sufficiently high for the
advertisement to overcome the saturation of the household A 102 (block 914).
For
example, the advertisement selector 202 may determine if the net effectiveness
metric
for the advertisement is the highest net effectiveness metric calculated for
the
household A 102. If the opportunity for a product is still sufficiently high
at the
household A 102, the net effectiveness metric is sufficiently high and the
selected
advertisement may still influence purchasing decisions despite having been
presented
to the household A 102 a number of times.
100861 After determining the net effectiveness metric, the advertisement
selector
202 determines whether there are additional advertisements that may be
selected for
the selected product (block 914). In some examples, a first advertisement for
a
selected product may have reached saturation at the household A 102, but a
second
advertisement for the selected product has not yet reached saturation. If
there are
additional advertisements available for the selected product (block 914),
control
returns to block 908 to select another advertisement for the selected product.
In
contrast, if there are no additional advertisements available (block 914), the
advertisement selector 202 determines whether there are additional products
for which
an opportunity metric is available (block 916).
100871 If, at block 916. there are no additional products. the advertisement
selector
202 identifies an advertisement for a product, based on the net effectiveness
metrics
and one or more thresholds, to the media deliverer 204 for transmission to the
selected
household A 102 (block 918). For example, the advertisement selector 202 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 the household A 102. In
some
-34-
CA 02743693 2011-06-17
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 the advertisement selector 202 has identified an advertisement for the
selected
household A 102 (block 918), the instructions 900 may end and control returns
to
block 706 of FIG. 7.
100881 FIG. 10 is a flowchart representative of example machine readable
instructions 1000 which may be executed to determine a media response of a
household to a delivered advertisement. The example instructions 1000 may be
executed by, for example, the processor 1 102 of FIG. I I to implement the
media
response evaluator 212 of FIG. 2.
[00891 The media evaluator 212 receives a media response characteristic from,
for
example, a media research organization or other media response characteristic
provider (block 1002). 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) - I /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 example
advertisement
saturation characteristic 324 of FIG. 313. the effectiveness f(n) resembles an
S-shaped
curve where the greatest effectiveness lies after the first presentation of an
advertisement.
-35-
CA 02743693 2011-06-17
100901 In some examples, the media evaluator 212 deduces a media response
characteristic of the household A 102 based on the changing purchases of the
household A 102, the advertisements presented to the household A 102, the
segment
of the household A 102, and/or the media response characteristic of the
segment of
the household A 102. 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, the media evaluator 212 may penalize advertisements less than
for
typical households for the same number of presentations.
100911 The media evaluator 212 receives an identification of an advertisement
delivered to the household A 102 (block 1004). For example, the media
evaluator 212
may be notified via metadata that an advertisement was delivered to the
household A
102 at a particular time. The media evaluator 212 determines the number of
times the
advertisement has been shown to the household A 102 in a measured period
(e.g., one
day, one week, two weeks, one month, etc.) (block 1006). Based on the number
of
times the advertisement has been shown (e.g., including the most recent
presentation)
and the media response characteristic, the media evaluator 212 determines a
media
response value of the household A 102 to future presentations of the
advertisement
(block 1008). The media response values may be stored for future reference by
the
advertisement selector 202_ In some examples, the media evaluator 212 only
determines the household media response on demand when the advertisement
selector
202 selects an advertisement for potential delivery based on a product with a
high
-36-
CA 02743693 2011-06-17
opportunity metric. In such examples, the media evaluator 212 sends the future
response to the advertisement selector 202. When the media evaluator 212 has
predicted the future response, the example instructions 1000 end and control
returns
to block 710 of FIG. 7.
[00921 FIG. 1 1 is a diagram of an example processor system 1100 that may be
used to execute the example machine readable instructions 700, 800, 900,
and/or 1000
described in FIGS. 7-10. as well as to implement the system 100 of FIGS. 1-3A
and
the headend system 200 described in FIG. 2. The example processor system 1 100
includes a processor 1 102 having associated memories, such as a random access
memory (RAM) 1104. a read only memory (ROM) 1106 and a flash memory 1108.
The processor 1102 in communication with an interface, such as a bus 1112 to
which
other components may be interfaced. In the illustrated example, the components
interfaced to the bus 1112 include an input device 1114, a display device
1116. a mass
storage device 1118, a removable storage device drive 1120, and a network
adapter
1122. The removable storage device drive 1120 may include associated removable
storage media 1124 such as magnetic or optical media. The network adapter 1
122
may connect the processor system 1100 to an external network 1126.
[00931 The example processor system 1100 may be, for example, a desktop
personal computer, a notebook computer, a workstation or any other computing
device. The processor 1 102 may be any type of logic device, such as a
microprocessor from the Intel Pcntium4` family of microprocessors, the
Intel"':
Itanium`- family of microprocessors, and/or the Intel XSeale`"' family of
processors.
The memories 1104, 1 106 and 1108 that are in communication with the processor
1102 may be any suitable memory devices and may be sized to fit the storage
-37-
CA 02743693 2011-06-17
demands of the system 1 100. In particular, the flash memory 1 108 may be a
non-
volatile memory that is accessed and erased on a block-by-block basis.
100941 The input device 1114 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 the processor 1102.
100951 The display device 1 1 16 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 the processor 1 102 and a user. The display
device 11 16
includes any additional hardware required to interface a display screen to the
processor 1102.
100961 The mass storage device 11 18 may be, for example, a hard drive or any
other magnetic, optical. or solid state media that is readable by the
processor 1102.
100971 The removable storage device drive 1120 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. The removable storage media 1124 is
complimentary to the removable storage device drive 1120, inasmuch as the
media
1 124 is selected to operate with the drive 1120. For example, if the
removable
storage device drive 1 120 is an optical drive, the removable storage media 1
124 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 removable storage device drive 1 120 is a magnetic
media
device, the removable storage media 1 124 may be, for example, a diskette or
any
other suitable magnetic storage media.
-38-
CA 02743693 2011-06-17
[00981 The network adapter 1 122 may be, for example, an Ethernet adapter, a
wireless local area network (LAN) adapter, a telephony modem, or any other
device
that allows the processor system 1 100 to communicate with one or more other
processor systems over a network. The external network 1] 26 may be a LAN, a
wide
area network (WAN), a wireless network, or any type of network capable of
communicating with the processor system 1 100. Example networks may include
the
Internet, an intranet, and/or an ad hoc network.
101001 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.
-39-
