Note: Descriptions are shown in the official language in which they were submitted.
1
STEERING DISTRIBUTIONS FOR CONNECTIONS FROM ONLINE
CAMPAIGNS OF UNIQUELY IDENTIFIABLE OBJECTS (UMS) BASED ON
PREDICTED DISTRIBUTIONS
Field of the Invention
[001-2] The invention relates generally to online connections from UI0s, and
more
specifically, to steering distributions for connections for online campaigns.
Background
[003] Online advertisements are important revenue generators for publishers.
As consumers
turn away from traditional media to spend more time on the Internet, and
increase spending
there as well, online advertisements are becoming even more critical. One key
concern for
marketers is how to effectively target online advertisements to consumers that
are interested
and likely to buy the specific or even uniquely identifiable product or
service, or even inquire
or pay attention to the pitch.
[004] When a marketer configures a marketing campaign to a server, a PPC value
is assigned
to advertisements to entice publishers to prominently display the
advertisements amongst other
published content. However, a marketer may not know what PPC values (or
CA 3014859 2019-08-26
CA 03014859 2018-08-16
WO 2017/149374
PCT/IB2017/000152
2
other values drawn from user interaction) to start out with. Also, a marketer
can have an
overall budget that needs to be allocated among several advertisements. An
inefficiently
configured marketing campaign is likely not an effective one.
[005] What is needed is a technique for steering distributions for connections
for campaigns
of uniquely identifiable objects (UI0s) based on predicted distributions. The
technique
should also update initial weightings dynamically in response to actual
connection data as it
comes available.
Summary
[006] To meet the above-described needs, methods, computer program products,
and
systems for steering distributions of connections from campaigns of UlOs based
on predicted
distributions.
[007] In one embodiment, a plurality of UI0s and parameters for configuring a
campaign
for the plurality of UI0s is received. One example is an advertisement
campaign for click-
throughs to a specific page on a car dealer web site which as more-detailed
information about
the vehicle. UI0s comparable to each of the plurality of UI0s of the
advertisement campaign
are automatically identified. A distribution of connections from a web site
displaying a MO
.. to a web site offering the UI0 (e.g., hyperlink clicks) is predicted for
the plurality of UI0s of
the campaign, based on historical information of connections (e.g., connection
velocity or
connection volume) for the identified comparable UI0s.
[008] In another embodiment, an initial connection weighting is automatically
generated for
each of the plurality of UI0s, wherein the initial weighted score affects how
often an
associated UI0 is displayed as part of an advertisement. The display can be
singular or part
of an array of options, and weighting can affect placement within the array of
options (e.g.,
higher up on the list). The plurality of UI0s are sent for display on
computing devices
according to the initial weighted scores.
[009] In yet another embodiment, feedback is received of actual connections
for each of the
plurality of UTOs for generating an actual connection distribution. Responsive
to the actual
3
connection distribution, at least part of the initial connections weights are
adjusted based on a
comparison of the actual advertisement connection distribution to the
predicted connection
distribution. The plurality of UI0s continue to be sent for display on
computing devices
according to the adjusted weighted scores. Another embodiment considers not
only the
campaign budget and connections monitored for the internal system, but also
campaigns and
resulting connections observed for external systems. As a result, connection
weights can be
adjusted in response to the success or failure of connections from external
campaigns.
[010] Advantageously, UI0 advertisements are weighted even before connection
data is
available. Weighting is dynamically adjusted based on real-time connection
data on once
available.
[010a] In yet another embodiment, there is provided a computer-implemented
method at a
connection management server for steering distributions for uniquely
identifiable object
(UI0) connections based on predicted connections, the method comprising the
steps of:
receiving, by the connection management server, data describing a plurality of
UI0s and
parameters for configuring an online advertisement campaign for the plurality
of UI0s;
automatically identifying, by the connection management server, UI0s
comparable to each of
the plurality of UI0s of the online advertisement campaign; automatically
predicting, by the
connection management server, a distribution of connections resulting from
displaying
graphical advertisements describing the plurality of UI0s of the online
advertisement
campaign with multiple webpages associated with the identified comparable UI0s
based on
historical information of connections for the identified comparable UI0s, each
graphical
advertisement including a hyperlink that, when selected, connects a browser
with a
termination point website associated with a selected UIO; automatically
generating, by the
connection management server, an initial connection weight for each of the
plurality of UI0s,
wherein the initial connection weight affects how often the graphical
advertisement for an
associated UI0 is displayed as part of an online advertisement; serving, by
the connection
management server, the graphical advertisements for the plurality of UI0s for
display in
association with the multiple webpages on mutiple computing devices according
to the initial
connection weights; receiving, by the connection management server, data
describing inputs
associated with the graphical advertisements sent for display on the multiple
computing
devices for each of the plurality of UI0s and generating an actual connection
distribution
Date Recue/Date Received 2021-06-15
3a
based on the data describing the inputs; responsive to the actual connection
distribution,
automatically adjusting, by the connection management server, at least part of
the initial
connection weights based on a comparison of the actual connection distribution
to the
predicted distribution of connections; and serving, by the connection
management server, the
.. graphical advertisements for the plurality of UI0s for display in
association with the multiple
webpages on the multiple computing devices at a frequency according to the
adjusted initial
connection weights.
1010b] In yet another embodiment, a non-transitory computer-readable medium
storing
.. source code that, when executed by a processor, performs a method in a
connection
management server for steering distributions for uniquely identifiable object
(UI0)
connections based on predicted connections, the method comprising the steps
of: receiving
data describing a plurality of UI0s and parameters for configuring an online
advertisement
campaign for the plurality of UI0s; automatically identifying UI0s comparable
to each of the
plurality of UI0s of the online advertisement campaign; predicting a
distribution of
connections resulting from displaying graphical advertisements describing the
plurality of
UI0s of the online advertisement campaign with one or more webpages associated
with the
identified comparable UI0s based on historical information of connections for
the identified
comparable UI0s, the graphical advertisements including a hyperlink that, when
selected,
.. connects a browser with a termination point associated with a selected UIO;
automatically
generating an initial connection weight for each of the plurality of UI0s,
wherein the initial
connection weight affects how often the graphical advertisement for an
associated UI0 is
displayed as part of an online advertisement; serving the graphical
advertisements for the
plurality of UI0s for display in association with the one or more webpages on
one or more
.. computing devices according to the initial connection weights; receiving
data describing
inputs associated with the graphical advertisements sent for display on the
one or more
computing devices for each of the plurality of UI0s and generating an actual
connection
distribution based on the data describing the inputs; responsive to the actual
connection
distribution, adjusting at least part of the initial connection weights based
on a comparison of
the actual connection distribution to the predicted distribution of
connections; and serving the
graphical advertisements for the plurality of UI0s for display in association
with the one or
more webpages on the one or more computing devices at a frequency according to
the
adjusted initial connection weights.
Date Recue/Date Received 2021-06-15
3b
[010c] In yet another embodiment, a connection management server for steering
distributions
for uniquely identifiable object (UI0) connections based on predicted
connections is
provided, the connection management server comprising: a processor; a network
interface;
and a memory communicatively coupled to the processor and the network
interface and
storing: a first module to receive data describing a plurality of UI0s and
parameters for
configuring an online advertisement campaign for the plurality of UI0s; a
second module to
automatically identify UI0s comparable to each of the plurality of UI0s of the
online
advertisement campaign; a third module to predict a distribution of
connections resulting
from displaying graphical advertisements describing the plurality of UI0s of
the online
advertisement campaign with one or more webpages associated with the
identified
comparable UI0s based on historical information of connections for the
identified
comparable UI0s, the graphical advertisements including a hyperlink that, when
selected,
connects a browser with a termination point associated with a selected UIO;
and a fourth
module to automatically generate an initial connection weight for each of the
plurality of
UI0s, wherein the initial connection weight affects how often the graphical
advertisement for
an associated UI0 is displayed as part of an online advertisement, wherein the
first module
serves the graphical advertisements for the plurality of UI0s for display in
association with
the one or more webpages on one or more computing devices according to the
initial
connection weights, wherein the first module receives data describing inputs
associated with
the graphical advertisements sent for display on the one or more computing
devices for each
of the plurality of UI0s and generating an actual connection distribution
based on the data
describing the inputs, and wherein the fourth module, responsive to the actual
connection
distribution, adjusts at least part of the initial connection weights based on
a comparison of
the actual connection distribution to the predicted distribution of
connections, and the fourth
module serving the graphical advertisements for the plurality of UI0s for
display in
association with the one or more webpages on the one or more computing devices
at a
frequency according to the adjusted initial connection weights.
Date Recue/Date Received 2021-01-18
3c
Brief Description of the Drawin2s
[011] In the following drawings, like reference numbers are used to refer to
like elements.
Although the following figures depict various examples of the invention, the
invention is not
limited to the examples depicted in the figures.
[012] FIG. 1 is a high-level block diagram illustrating a system for steering
distributions for
connections for campaigns (e.g., advertisement campaigns) of UI0s based on
predicted
distributions, according to one embodiment.
[013] FIG. 2 is a more detailed block diagram illustrating a remote
advertisement server (or
connection management server) of the system of FIG. 1, according to one
embodiment.
[014] FIG. 3 is a flow diagram illustrating a method for steering
distributions for
connections for advertisement campaigns of UI0s based on predicted
distributions, according
to one embodiment.
[015] FIG. 4 is a flow diagram illustrating a step of determining initial
connection weights
based on a cold-start prediction of connections for the method of FIG. 3,
according to one
embodiment.
Date Recue/Date Received 2021-01-18
CA 03014859 2018-08-16
WO 2017/149374
PCT/IB2017/000152
4
[016] FIG. 5 is a flow diagram illustrating a step of adjusting initial
connection weights for
the method of FIG. 3, according to one embodiment.
[017] FIGS. 6A-B are schematic diagrams illustrating web pages with exemplary
online
advertisement displays of uniquely identifiable objects, according to some
embodiments.
[018] FIGS. 6C-D are schematic diagrams illustrating web pages with exemplary
connections for the online advertisements of uniquely identifiable objects,
according to some
embodiments.
[019] FIG. 7 is a block diagram illustrating an exemplary computing device,
according to
one embodiment.
Detailed Description
[020] The present invention provides methods, computer program products, and
systems for
steering distributions for connections from campaigns (e.g., advertisement
campaigns) of
UI0s based on predicted distributions. In general, a uniquely identifiable
object can be
individually distinguished over all other objects, and even those that have
nearly identical
manufacturing specifications, by a unique identification code (e.g., a serial
number or other
type of distinguishing data).
[021] In one exemplary embodiment, an automobile appears on a web site, such
as
Craigslist.com. Metadata in the web page, or content displayed in the web
page, provides a
VIN number assigned to the automobile. A client identifies the VIN number for
submission
to a remote server that stores online advertisements. As a unique identifier,
the VIN number
can reveal characteristics of the automobile such as model, make, year, and
the like. The
characteristics can be indicative of certain types of owner categories that
are useful for
advertising impressions. Therefore, the remote server returns PPC (pay per
click) sponsored
online advertisements associated with the VIN number, specific to that VIN
itself, and any
related ads to display along with the uniquely identified object, such as a
related vehicles or
affinity offerings (i.e., related goods are services that are not necessarily
unique).
CA 03014859 2018-08-16
WO 2017/149374
PCT/IB2017/000152
[022] In another exemplary embodiment, as set forth more fully in US
Application No.
13/555,987 an automobile marketplace web site is a publisher that hosts direct
revenue-
generating online advertisements or free advertisements submitted directly by
sellers. A
search engine integrated within the website returns search results to user
queries for specific
5 types of automobiles as a SERP (search engine results page). To augment
the results, the
automobile marketplace web site can also list third-party sponsored
advertisements or search
results within or along with the other search results. As a result, the
automobile marketplace
web site can generate additional revenue from connections (e.g., click-
throughs to the third
party sponsored search results of uniquely identifiable objects, or exposure
of connection
data).
[023] As used herein, the terms "third party" and "remote" are relative. From
a publisher
point of view, an advertisement server is a remote, third party. From an
advertisement server
point of view, the publisher is a remote, third party. The point of views and
above terms are
used interchangeably herein.
[024] Also, please note that the disclosure herein often refers to automobiles
solely for the
purposes of thoroughly delineating a preferred embodiment. However, one of
ordinary skill
in the art will recognize, given the disclosure, that the techniques can be
applied to other
individually distinguishable products such as real estate or artwork. Further,
services such as
dentistry can be uniquely identified by time, location and/ or a particular
dentist within an
office.
[025] I. Connection Steering Systems (FIGS. 1-2)
[026] FIG. 1 is a high-level block diagram illustrating a system 100 for
steering
distributions for advertisement connections for advertisement campaigns of
UI0s based on
predicted distributions, according to one embodiment. The system 100 comprises
a remote
advertisement server 110 (or more generally. a connection management server),
a publisher
120, a sponsor web site (or termination point) 130, and a consumer 140. The
components can
be coupled to a network 199, such as the Internet, a local network or a
cellular network,
through any suitable wired (e.g., Ethernet) or wireless (e.g., Wi-Fi or 3G)
medium, or
combination. Other embodiments of communication channels are possible,
including hybrid
CA 03014859 2018-08-16
WO 2017/149374
PCT/1132017/000152
6
networks. Additional network components can also be part of the system 100,
such as
firewalls, virus scanners, routers, switches, application servers, databases,
and the like.
[027] The remote advertisement server 110 can serve advertisements according
to a
predicted advertisement connection distribution. For instance, past listings
of vehicles having
certain characteristics can help estimate future listings of vehicles with
similar characteristics.
One embodiment of the remote advertisement server 110 responds to a feedback
of actual
connection data. In more detail, an actual distribution of advertisement
connections is
compared to a targeted distribution of advertisement connections. The targeted
distribution
can be received from a marketer, or can be preloaded in server software. As a
result, some
UFOs can be sent for display more or less often in order draw more or fewer
click-throughs,
relative to other UI0s of an advertising campaign. Some UICos have higher
click-through
rates, and as a result, require fewer impressions until sold. So a flashy
vehicle, such as a red
corvette, may naturally draw more click-throughs than an old beat up truck.
[028] In one embodiment, a weighted score is calculated in order to increase
or decrease the
number of connections to UI0s. A vehicle that naturally draws fewer click-
throughs can be
assigned a higher weighting and get served more often. The weighting of an
embodiment is
relative to other vehicles, or other Ms, of a particular advertisement
campaign.
Consequently, more exposure of one advertisement results in less exposure of
one or more
other advertisements. In one embodiment, weighting adjusts predicted
distributions from
historical data to target distributions preferred by a marketer. In another
embodiment,
weighting adjusts actual distributions to target distributions. In still
another embodiment,
weighting adjusts actual distributions to predicted distributions. For
instance, marketer may
be paying for a specific guarantee of service as predicted and computing
resources can be
conserved when outperforming the predictions, and computing resources can be
reserved as
needed. Adjustments can occur in real-time, periodically, or responsive to
reaching a
threshold.
[029] Another embodiment considers not only the campaign budget and
connections
monitored for the internal system, but also campaigns and resulting
connections observed for
external systems. As a result, connection weights can be adjusted in response
to the success
CA 03014859 2018-08-16
WO 2017/149374
PCT/1B2017/000152
7
or failure of connections from external campaigns. External connection
statistics can be sent
from a daemon that intercepts feed data, by providing log-in information to
external systems,
by pushing data between systems, available third party data sources, and by
any other
appropriate technique of sharing statistics.
[030] Yet another embodiment, adjusts weighting not only responsive to the
number of
connections but also in consideration of the quality of connections. A click-
through to a web
site that is immediately discontinued has less value than a click-through to a
web site that
results in a 10-minute view or a sale. Data for tracking quality of
connections can be
obtained from daemons on the marketer web site, apps running on the client,
inter-system
compatibility, or the like.
[031] In some embodiments, the remote advertisement server 110 stores online
advertisements related to uniquely identifiable objects. The online
advertisements can be
sponsored advertisements having a PPC and campaign budget set by marketers.
The online
advertisements can include direct marketing of an object, or indirect
marketing of related
advertisements. For example, a vehicle can be marketed by a used car dealer,
and peripheral
products to the vehicle, such as CarFax reports, offers to buy your existing
car for cash,
finance, insurance, new car options, used vehicle parts, and oil change
locations can be
marketed to potential buyers of the vehicle. Characteristics can be associated
with the
peripheral products to allow matching to characteristics of objects.
[032] The remote advertisement server 110 can be a computer, a set of
distributed
computers, a server, a virtual server, or the like. The remote advertisement
server 110 can be
operated by an entity that is independent from marketers of U1Os and
publishers of UlOs,
thereby providing an advertising exchange.
[033] More detailed embodiments of the remote advertisement server 110 are
discussed in
more detail below in association with FIG. 2.
CA 03014859 2018-08-16
WO 2017/149374
PCT/1B2017/000152
8
[034] The publisher (or third party) 120 can be a computer, a set of
distributed computers, a
server, a virtual server, or the like. Examples of the publisher include
general search engines
(e.g., Google or Bing), marketplaces with vertical search engines (e.g., Cars.
com or
Amazon.com), marketers, bloggers, news sources, classified publishers, social
media
providers, third parties, and the like.
[035] Web pages 122 and an optional code parser 124 are included in the
illustrated
embodiment of the publisher 120. However, in other embodiments, the components
are not
integrated, and can be owned and/ or operated by a third party. The web pages
122 can be
static or dynamic web pages that are incorporated to a web site available at a
particular
domain, or be individual or sub-groups of pages. The content of web pages 122
can relate to
commercial products offered by the publisher 120, such as vehicles or resold
products. The
content of web pages 122 can also be blogs, news articles, social media,
multimedia such as
images or video, or any other type of content suitable for online viewing.
Search
functionality can be included. One of more online advertisements are included
with a web
page. The online advertisements can be provided in real time (e.g., for
dynamically
generated web pages), or can be pre-processed in batch (e.g., for static web
pages).
[036] The code parser 124 can be a process, daemon, operating system module,
other
application, or hardware that checks the web pages 122 for unique identifiers.
In response,
the unique identifier is used to retrieve online advertisements for insertion
to web pages. In
other embodiments, a code parser 124 is not necessary, such as when online
advertisements
are directly associated with uniquely identifiable objects by a marketer.
[037] The sponsor web site (which is an exemplary termination point) 130 can
be an online
inventory of products or services of a marketer. In some embodiments, a search
engine sends
a consumer to the sponsor web site 130. In the case of advertisements based on
a search
query, the sponsor web site 130 can be displayed within code of a SERP web
page controlled
by the search engine 125. Alternatively, a new tab or browser instance can be
spawned.
Note that the sponsor web site 130 is just one example of a connection medium
associated
with an advertisement. Connection mediums can include a termination point for
CA 03014859 2018-08-16
WO 2017/149374
PCT/1B2017/000152
9
automatically forwarding a user, or mere connection data for the user to
select how to
connect. Exemplary connection mediums include an SMS message, click to expose
a
complete telephone number, click to initiate a telephone call, an address that
can include
directions, click for navigation, a VDP (vehicle description page) or other
deep link, and the
like.
[038] The consumer (or customer or user) 140 can be an individual, a corporate
entity, an
automated process, or the like, using a computing device to use the third-
party web site. The
computing device can be mobile (e.g., a smartphone) or stationary. The
consumer 140 can
submit queries and click-on sponsored advertisements. For example, the
consumer 140 can
select a deep hyperlink from a UI0 advertisement that is connected to a
specific white-
colored Nissan Pathfinder located at a specific dealership nearby to the
consumer 140.
Aggregate data can be collected for a number of times a UI0 is shown to
consumers, a
number of times consumers click on a hyperlink for a UIO, a type of hyperlink
that is clicked
among several types of hyperlinks (e.g., other types of hyperlink engagements
include a web
page hyperlink, an SMS hyperlink, an e-mail hyperlink, a navigation web page
hyperlink, or
the like), and other types of data. Some types of client engagement outside of
click-throughs
can occur including hover-overs, Java interactions, gestures, user reactions,
future HTML
features, and the like. The consumer data can be raw data or pre-processed
data.
[039] FIG. 2 is a more detailed block diagram illustrating a remote
advertisement server 110
of the system 100 of FIG. 1, according to one embodiment. The remote
advertisement server
110 includes a statistical analysis engine 201 and a connection steering
engine 202 which acts
as a controller for the process, in some embodiments. Peripheral components
include an
advertisement selection engine 203, an account manager 210, a uniquely
identifiable object
database 220, a search engine interface 230, and a network module 240. The
components can
be implemented in hardware, software, or a combination of both.
[040] The statistical analysis engine 201 and the connection steering engine
204 are the core
components of one implementation. In cooperation, a predicted distribution
helps in a cold-
start situation for a particular UI0 by leveraging historical data for similar
UI0s. Histograms
can be generated and used to model past data. A resulting statistical model
quantifies the
CA 03014859 2018-08-16
WO 2017/149374
PCT/1B2017/000152
likelihood of a desired amount of exposure. The exposure is measured by click-
throughs to
deep links to a page with detailed information about a U10, or by other
measurements. The
connection steering engine 202 calculates and/ or applies weighting based on
predicted and/
or actual connections. One case starts out in a first mode by using predicted
connections,
5 shifts to a second mode by using a hybrid of rolling feedback for actual
connections, and
finally shifts to a third mode of using actual connections to drive future
weighting.
[041] The advertisement selection engine 203 identifies a UI0 advertisement
and accesses
memory at a corresponding location to retrieve an advertisement for download
to a publisher,
10 according to the weighting. Without modification by weighting,
advertisement selection can
be based on a random model, a round-robin or other parity model, a guaranteed
service level,
or the like. The weighting either completely or partially modifies the
preexisting selection
algorithms. A digital image file can be sent from the remote advertisement
server 110 for
display at the publisher 120.
[042] The account manager 210 is an interface for a marketer to manage online
advertisements. Settings can be global to the account, or particular to a
uniquely identifiable
object. Exemplary settings include a PPC (pay-per-click) value, an overall
budget (e.g., a
maximum spending allowance, a maximum number of clicks), a per uniquely
identifiable
object budget, a per uniquely identifiable object bid amount, a bid geography,
a fallback
termination point, and a marketing agency. In some embodiments, the account
manager 210
logs click-throughs. A marketer may access the account manager 210 using a
viewer such as
a web browser or a local client or API.
[043] The uniquely identifiable object database 220 stores data entered by a
marketer about
uniquely identifiable objects through the user interface, and stores data
harvested from
external data sources. For instance, a relational database stores a record for
each uniquely
identifiable object. Other data stored can include a uniquely identifiable
object key (e.g., a
unique identification code, such as a serial number, a VIN, or the like), a
marketer
identification, a bid amount, a termination point, a daily maximum, and a time
stamp (i.e.,
record last updated), and external data (e.g., Carfax.com data). A user
interface allows the
marketer to control PPC values and budgets. There are many ways for a user to
control PPC
CA 03014859 2018-08-16
WO 2017/149374
PCT/1B2017/000152
11
values (e.g., programmatically, selectively, overriding, nil es, algorithms,
etc.).
[044] In one embodiment, a historical data module (not shown in drawing)
collects and
aggregates data of past UI0s number of displays, number of click-throughs and
UI0
characteristics. The data can be collected on a per-click real-time basis or
be periodically
uploaded. Historical data can be categorized by characteristics. Some
embodiments operate
within the remote advertisement server 110 and others embodiments operate
externally on
dedicated servers. An embodiment stores the historical data or analysis
thereof within the
uniquely identifiable database 220.
[045] Advertisements can be indexed in the uniquely identifiable object
database 220
according to identification data. The identification data can be, for example,
unique
identification codes, serial numbers, VINs, addresses, or any other
appropriate data to
uniquely identify each object. The objects can be distinguished even if
manufactured to the
same specifications. In one implementation, VINs are not available for cars
manufactured
prior to 1981, so a unique identification code can be generated. The method
for generating
the code can be made available to a publisher to utilize locally and/ or
alternatively, recipients
of the generated code can be provided a key. Other data can also be sent along
with the
identification data. If an advertisement is an enriched listing, information
can be downloaded
ahead of time (e.g., multimedia files).
[046] Additionally, advertisement identifiers can be stored in a record. The
advertisement
identifiers are used as pointers to online advertisements stored either
locally or on a separate
advertisement server. In further details of the local_ storage implementation,
online
.. advertisements can be stored in the uniquely identifiable object database
220. Each online
advertisement can include characteristics for matching against uniquely
identifiable objects.
For example, a certain car model and a range of years can be specified as
appropriate for a
CarFax online advertisement. In US Application No. 123/523,703, techniques for
populating
a uniquely identifiable object database are described in further detail.
[047] The publisher interface 230 communicates with search engines concerning
sponsored
search results or sponsored advertisements (e.g., eligibility of
advertisements). In a first
instance, the publisher interface 230 provides listings from the uniquely
identifiable object
CA 03014859 2018-08-16
WO 2017/149374
PCT/1B2017/000152
12
database 220 to include in SERPs. In an embodiment, identification data for
the listings are
provided in batch during overnight data transfers. The listings can be updated
during
specified intervals between batch transfers. For example, VINs, bids, and
other data for
automobiles can be provided to a website like Lemonfree.com or Amazon. corn.
[048] The network module 240 provides lower-level interfaces with external
devices such
as remote search engines. The network module 240 translates commands sent and
received
according to a specific protocol or API (application programming interface).
Moreover,
network cards manage the transmission of data over a network with the external
devices
using IEEE standards.
[049] II. Connection Steering Methods (FIGS. 3-5)
[050] FIG. 3 is a flow diagram illustrating a method 300 for steering
distributions for
connections from advertisement campaigns of U1Os based on predicted
distributions,
according to one embodiment. The method 300 can be implemented by, for
example, the
remote advertisement server 110 of FIG. 1.
[051] At step 310, a plurality of UI0s and parameters for configuring an
advertisement
campaign are received. Marketers can log-on and upload individual or batch
lists of UFOs.
Also, web crawlers can automatically pull Ms from a dealer web site. Web site
pages for
particular UI0s can be abstracted to a template for improved data scraping.
Additionally,
information and UI0s can be received from feeds.
[052] At step 320. UI0 advertisements are displayed based on a cold-start mode
utilizing a
predicted distribution of advertisement connections. Additional details are
described below
with respect to FIG. 4.
[053] At step 330, when feedback of actual connections for displayed UI0s is
received, at
step 340, UI0 display is continued with adjustments to initial advertisement
weightings
CA 03014859 2018-08-16
WO 2017/149374
PCT/1B2017/000152
13
responsive to statistics of the actual connection feedback, as is described
below in association
with FIG. 5.
[054] FIG. 4 is a flow diagram illustrating the step 320 of determining
initial advertisement
weights based on a cold-start prediction of advertisement connections for the
method of FIG.
3, according to one embodiment.
[055] At step 410, UlOs comparable to the UlOs of the advertisement campaign
are
identified. Characteristics of a house are use in one example to identify. for
example, other
houses that are brick, within a certain area, have a certain age, include a
garage, or any
combination of characteristics. In one case, a score indicates close a UI0 is
to one or more
other UlOs. The comparisons can be made as a whole, or with respect to a
specific
characteristic.
[056] At step 420, a distribution of advertisement connections is predicted by
advertisement
displays. The distribution can be formed by statistical analysis of historical
connection data
from similar UlOs. A stochastic process library can store modeling data for
Gaussian
processes, Markov processes, Bernoulli processes, and the others.
[057] At step 430, initial connection weights are generated for UlOs to affect
the predicted
connection distribution. A marketer may not have a high enough budget to
support some
predicted exposure levels. Charges or fees can also relate to predictions.
[058] At step 440, selected UlOs are sent for display according to initial
advertisement
weights. The online advertisement can be displayed numerous different ways. In
one
example, display is integrated with a web page (e.g., SERP results). In
another example,
display is separated into a separate web page, tab, or shadow box as shown in
FIGS. 6A-B.
Display of various termination points are shown in FIGS. 6C-D. Additional
display
examples include a banner (static or dynamic rotational) or a container (e.g.,
an ad box) of a
web page. In the ad box, one or more uniquely identifiable object listings are
displayed along
CA 03014859 2018-08-16
WO 2017/149374
PCT/1B2017/000152
14
with one or more PPC termination. In still another example, the advertisement
can be
displayed outside of the browser, such as directly on a desktop or within an
application
window.
[059] A user is connected to a connection medium (e.g., a termination point or
connection
data) associated with the sponsored advertisement. In a preferred embodiment,
the
termination point is a URL (Universal Resource Locator) reachable through a
network. Other
exemplary termination points include maps or an address with directions,
instantiation of
instant messenger boxes connecting a customer to a marketer, a telephone
number, SMS
message, e-mail message, and the like.
[060] In an embodiment, connecting a consumer to a termination point, includes
displaying
an online advertisement is displayed until selected by a user. In some
instances, the online
advertisement rotates among other online advertisements, or reaches a time
out. In other
instances, a browser window containing the advertisement can be closed.
Different
termination points can vary in PPC values. The user clicks are recorded in
real-time, the
updates of available advertisements can be provided in substantially real-time
or in periodic
batches (e.g., hourly or daily) and those batches can be updated locally based
on the
publisher's schedule. If authorization is not given due to an expired budget,
execution of the
termination point can be denied. In this case, a default termination point can
be substituted,
or the user can simply receive an exception message. If the PPC budget has not
been expired,
authorization to direct the user to a termination point or otherwise make
available the
sponsored connection method associated with the advertisement is received. In
some cases,
the termination point is sent with the online advertisement. In other cases,
the termination
point is sent upon selection (e.g., a dynamic termination point).
[061] FIG. 5 is a flow diagram illustrating the step 340 of adjusting initial
connection
weights for the method of FIG. 3, according to one embodiment.
[062] At step 510, a distribution of actual connection is generated.
CA 03014859 2018-08-16
WO 2017/149374
PCT/1B2017/000152
[063] At step 520, one or more initial connection weights can be adjusted
based on
comparisons of actual connection distributions compared to predicted
advertisement
connection distributions.
5 [064] At step 530, selected UI0s are sent for display in accordance with
adjusted
connection weights. The same displays of FIGS. 6A-6D are examples of output to
consumers.
[065] III. Computing Devices (FIG. 7)
[066] FIG. 7 is a block diagram illustrating an exemplary computing device 700
for use in
the system 100 of FIG. 1, according to one embodiment. The computing device
600 is an
exemplary device that is implementable for each of the components of the
system 100,
including the remote advertisement server 110, the publisher 120, the sponsor
web site 130,
and the consumer 140. Additionally, the system 100 is merely an example
implementation
itself, since the system 100 can also be fully or partially implemented with
laptop computers,
tablet computers, smart cell phones. Internet appliances, and the like.
[067] The computing device 700, of the present embodiment, includes a memory
710, a
processor 720, a hard drive 730, and an I/O port 740. Each of the components
is coupled for
electronic communication via a bus 799. Communication can be digital and/ or
analog, and
use any suitable protocol.
[068] The memory 710 further comprises network applications 720 and an
operating system
722. The network applications 720 can include the modules of advertisement
server 110 of
FIG. 2. Other network applications can include a web browser, a mobile
application, an
application that uses networking, a remote application executing locally, a
network protocol
application, a network management application, a network routing application,
or the like.
[069] The operating system 722 can be one of the Microsoft Windows family of
operating
systems (e.g., Windows 95, 98, Me, Windows NT, Windows 2000, Windows XP,
Windows
XP x64 Edition, Windows Vista, Widows 7, Windows CE, Windows Mobile), Linux,
HP-
CA 03014859 2018-08-16
WO 2017/149374
PCT/1B2017/000152
16
UX, UNIX, Sun OS, Solaris, Mac OS X, Alpha OS, AIX, IRIX32, or IRIX64. Other
operating systems may be used. Microsoft Windows is a trademark of Microsoft
Corporation.
[070] The processor 720 can be a network processor (e.g., optimized for IEEE
802.11), a
general purpose processor, an application-specific integrated circuit (ASIC),
a field
programmable gate array (FPGA), a reduced instruction set controller (RISC)
processor, an
integrated circuit, or the like. Atheros, Broadcom, and Marvell Semiconductors
manufacture
processors that are optimized for IEEE 802.11 devices. The processor 720 can
be single core,
multiple core, or include more than one processing elements. The processor 720
can be
disposed on silicon or any other suitable material. The processor 720 can
receive and execute
instructions and data stored in the memory 710 or the storage drive 730.
[071] The storage drive 730 can be any non-volatile type of storage such as a
magnetic disc,
EEPROM, Flash, or the like. The storage drive 730 stores code and data for
applications.
[072] The I/O port 740 further comprises a user interface 742 and a network
interface 744.
The user interface 742 can output to a display device and receive input from,
for example, a
keyboard. The network interface 744 connects to a medium such as Ethernet or
Wi-Fi for
data input and output.
[073] Many of the functionalities described herein can be implemented with
computer
software, computer hardware, or a combination.
[074] Computer software products (e.g., non-transitory computer products
storing source
code) may be written in any of various suitable programming languages, such as
C, C++, C#,
Java, JavaScript, PHP, Python, Perl, Ruby, and AJAX. The computer software
product may
be an independent application with data input and data display modules.
Alternatively, the
computer software products may be classes that are instantiated as distributed
objects. The
computer software products may also be component software such as Java Beans
(from Sun
Microsystems) or Enterprise Java Beans (from Sun Microsystems).
[075] Furthermore, the computer that is running the previously mentioned
computer
software may be connected to a network and may interface to other computers
using this
CA 03014859 2018-08-16
WO 2017/149374
PCT/1B2017/000152
17
network. The network may be on an intranet or the Internet, among others. The
network
may be a wired network (e.g., using copper), telephone network, packet
network, an optical
network (e.g., using optical fiber), or a wireless network, or any combination
of these. For
example, data and other information may be passed between the computer and
components
(or steps) of a system of the invention using a wireless network using a
protocol such as Wi-
Fi (IEEE standards 802.11, 802.11a, 802.11b, 802.11e, 802.11g, 802.11i, and
802.11n, just to
name a few examples). For example, signals from a computer may be transferred,
at least in
part, vvirelessly to components or other computers.
[076] In an embodiment, with a Web browser executing on a computer workstation
system,
a user accesses a system on the World Wide Web (WWW) through a network such as
the
Internet. The Web browser is used to download web pages or other content in
various formats
including HTML, XML, text, PDF, and postscript, and may be used to upload
information to
other parts of the system. The Web browser may use uniform resource
identifiers (URLs) to
identify resources on the Web and hypertext transfer protocol (HTTP) in
transferring files on
the Web.
[077] This description of the invention has been presented for the purposes of
illustration
and description. It is not intended to be exhaustive or to limit the invention
to the precise
form described, and many modifications and variations are possible in light of
the teaching
above. The embodiments were chosen and described in order to best explain the
principles of
the invention and its practical applications. This description will enable
others skilled in the
art to best utilize and practice the invention in various embodiments and with
various
modifications as are suited to a particular use. The scope of the invention is
defined by the
following claims.